summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorAndreas Baumann <mail@andreasbaumann.cc>2017-05-06 16:36:16 +0200
committerAndreas Baumann <mail@andreasbaumann.cc>2017-05-06 16:36:16 +0200
commit19bd8afa6b63dccadc2006292154f1633e2a29c9 (patch)
tree1677f5a2b3caccce45a75fc6ca5857a7d820686c
parent4b2800f8f3ee34d9779dfdcd40d7e68b569380e6 (diff)
downloadmonitoring-plugins-19bd8af.tar.gz
some rework:
- added old style 'redir' function and options along to a new libcurl internal 'follow' parameter 'curl' - moved picohttpparser to it's own subdirectory - added uriparser to be used instead of the home-grown parser in 'redir'
Diffstat
-rw-r--r--.gitignore13
-rw-r--r--ACKNOWLEDGEMENTS13
-rw-r--r--configure.ac38
-rw-r--r--plugins/Makefile.am12
-rw-r--r--plugins/check_curl.c320
-rw-r--r--plugins/picohttpparser/Makefile.am3
-rw-r--r--plugins/picohttpparser/picohttpparser.c (renamed from plugins/picohttpparser.c)0
-rw-r--r--plugins/picohttpparser/picohttpparser.h (renamed from plugins/picohttpparser.h)0
-rw-r--r--plugins/uriparser/Makefile.am9
-rw-r--r--plugins/uriparser/Uri.h777
-rw-r--r--plugins/uriparser/UriBase.h197
-rw-r--r--plugins/uriparser/UriCommon.c567
-rw-r--r--plugins/uriparser/UriCommon.h104
-rw-r--r--plugins/uriparser/UriCompare.c168
-rw-r--r--plugins/uriparser/UriDefsAnsi.h82
-rw-r--r--plugins/uriparser/UriDefsConfig.h101
-rw-r--r--plugins/uriparser/UriDefsUnicode.h82
-rw-r--r--plugins/uriparser/UriEscape.c453
-rw-r--r--plugins/uriparser/UriFile.c226
-rw-r--r--plugins/uriparser/UriIp4.c329
-rw-r--r--plugins/uriparser/UriIp4.h92
-rw-r--r--plugins/uriparser/UriIp4Base.c96
-rw-r--r--plugins/uriparser/UriIp4Base.h59
-rw-r--r--plugins/uriparser/UriNormalize.c728
-rw-r--r--plugins/uriparser/UriNormalizeBase.c119
-rw-r--r--plugins/uriparser/UriNormalizeBase.h53
-rw-r--r--plugins/uriparser/UriParse.c2241
-rw-r--r--plugins/uriparser/UriParseBase.c90
-rw-r--r--plugins/uriparser/UriParseBase.h55
-rw-r--r--plugins/uriparser/UriQuery.c460
-rw-r--r--plugins/uriparser/UriRecompose.c577
-rw-r--r--plugins/uriparser/UriResolve.c316
-rw-r--r--plugins/uriparser/UriShorten.c320
33 files changed, 8661 insertions, 39 deletions
diff --git a/.gitignore b/.gitignore
index 5fa4c813..1c8cbdec 100644
--- a/.gitignore
+++ b/.gitignore
@@ -202,7 +202,18 @@ NP-VERSION-FILE
202/plugins/negate 202/plugins/negate
203/plugins/stamp-h* 203/plugins/stamp-h*
204/plugins/urlize 204/plugins/urlize
205/plugins/libpicohttpparser.a 205
206# /plugins/picohttpparser
207/plugins/picohttpparser/Makefile
208/plugins/picohttpparser/Makefile.in
209/plugins/picohttpparser/.deps
210/plugins/picohttpparser/libpicohttpparser.a
211
212# /plugins/uriparser
213/plugins/uriparser/Makefile
214/plugins/uriparser/Makefile.in
215/plugins/uriparser/.deps
216/plugins/uriparser/liburiparser.a
206 217
207# /plugins/t/ 218# /plugins/t/
208/plugins/t/*.tmp 219/plugins/t/*.tmp
diff --git a/ACKNOWLEDGEMENTS b/ACKNOWLEDGEMENTS
index 50c714c3..06e84e25 100644
--- a/ACKNOWLEDGEMENTS
+++ b/ACKNOWLEDGEMENTS
@@ -31,3 +31,16 @@ Gnulib team
31Copyright (C) 2001, 2003, 2004, 2006 Free Software Foundation, Inc 31Copyright (C) 2001, 2003, 2004, 2006 Free Software Foundation, Inc
32http://www.gnu.org/software/gnulib/ 32http://www.gnu.org/software/gnulib/
33Use of lib files that originally were used from coreutils 33Use of lib files that originally were used from coreutils
34
35Copyright (c) 2009-2014 Kazuho Oku, Tokuhiro Matsuno, Daisuke Murase,
36 Shigeo Mitsunari
37picohttpparser
38https://github.com/h2o/picohttpparser
39Use of the library for HTTP header parsing in check_curl.
40
41Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
42Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
43All rights reserved.
44uriparser - RFC 3986 URI parsing library
45http://uriparser.sourceforge.net/
46Use of the library for URL parsing in check_curl.
diff --git a/configure.ac b/configure.ac
index 04ec5bde..c56163c1 100644
--- a/configure.ac
+++ b/configure.ac
@@ -385,19 +385,31 @@ if test "$ac_cv_header_wtsapi32_h" = "yes"; then
385 AC_SUBST(WTSAPI32LIBS) 385 AC_SUBST(WTSAPI32LIBS)
386fi 386fi
387 387
388dnl Check for cURL library 388AC_ARG_ENABLE(check-curl,
389LIBCURL_CHECK_CONFIG(yes, 7.15.2, [ 389 AC_HELP_STRING([--enable-check-curl],
390 EXTRAS="$EXTRAS check_curl\$(EXEEXT)" 390 [Enables compilation of check_curl (default: no)]),
391 LIBCURLINCLUDE="$LIBCURL_CPPFLAGS" 391 [enable_check_curl=$enableval],
392 LIBCURLLIBS="$LIBCURL" 392 [enable_check_curl=no])
393 LIBCURLCFLAGS="$LIBCURL_CPPFLAGS" 393if test "$enable_check_curl" = "yes" ; then
394 AC_SUBST(LIBCURLINCLUDE) 394 dnl Check for cURL library
395 AC_SUBST(LIBCURLLIBS) 395 LIBCURL_CHECK_CONFIG(yes, 7.15.2, [
396 AC_SUBST(LIBCURLCFLAGS) 396 EXTRAS="$EXTRAS check_curl\$(EXEEXT)"
397], [ 397 LIBCURLINCLUDE="$LIBCURL_CPPFLAGS"
398 AC_MSG_WARN([Skipping curl plugin]) 398 LIBCURLLIBS="$LIBCURL"
399 AC_MSG_WARN([install libcurl libs to compile this plugin (see REQUIREMENTS).]) 399 LIBCURLCFLAGS="$LIBCURL_CPPFLAGS"
400]) 400 AC_SUBST(LIBCURLINCLUDE)
401 AC_SUBST(LIBCURLLIBS)
402 AC_SUBST(LIBCURLCFLAGS)
403 AC_SUBST(PICOHTTPPARSER_DIR, picohttpparser)
404 AC_SUBST(URIPARSER_DIR, uriparser)
405 ], [
406 AC_MSG_WARN([Skipping curl plugin])
407 AC_MSG_WARN([install libcurl libs to compile this plugin (see REQUIREMENTS).])
408 ])
409fi
410AM_CONDITIONAL([WITH_CHECK_CURL], [test "$enable_check_curl" = "yes"])
411AM_COND_IF([WITH_CHECK_CURL],
412 [AC_CONFIG_FILES([plugins/picohttpparser/Makefile plugins/uriparser/Makefile])])
401 413
402dnl Fallback to who(1) if the system doesn't provide an utmpx(5) interface 414dnl Fallback to who(1) if the system doesn't provide an utmpx(5) interface
403if test "$ac_cv_header_utmpx_h" = "no" -a "$ac_cv_header_wtsapi32_h" = "no" 415if test "$ac_cv_header_utmpx_h" = "no" -a "$ac_cv_header_wtsapi32_h" = "no"
diff --git a/plugins/Makefile.am b/plugins/Makefile.am
index ff745c39..34a7da8b 100644
--- a/plugins/Makefile.am
+++ b/plugins/Makefile.am
@@ -40,17 +40,17 @@ EXTRA_PROGRAMS = check_mysql check_radius check_pgsql check_snmp check_hpjd \
40 check_nagios check_by_ssh check_dns check_nt check_ide_smart \ 40 check_nagios check_by_ssh check_dns check_nt check_ide_smart \
41 check_procs check_mysql_query check_apt check_dbi check_curl 41 check_procs check_mysql_query check_apt check_dbi check_curl
42 42
43SUBDIRS = @PICOHTTPPARSER_DIR@ @URIPARSER_DIR@
44
43EXTRA_DIST = t tests 45EXTRA_DIST = t tests
44 46
45PLUGINHDRS = common.h 47PLUGINHDRS = common.h
46 48
47noinst_LIBRARIES = libnpcommon.a libpicohttpparser.a 49noinst_LIBRARIES = libnpcommon.a
48 50
49libnpcommon_a_SOURCES = utils.c netutils.c sslutils.c runcmd.c \ 51libnpcommon_a_SOURCES = utils.c netutils.c sslutils.c runcmd.c \
50 popen.c utils.h netutils.h popen.h common.h runcmd.c runcmd.h 52 popen.c utils.h netutils.h popen.h common.h runcmd.c runcmd.h
51 53
52libpicohttpparser_a_SOURCES = picohttpparser.c
53
54BASEOBJS = libnpcommon.a ../lib/libmonitoringplug.a ../gl/libgnu.a 54BASEOBJS = libnpcommon.a ../lib/libmonitoringplug.a ../gl/libgnu.a
55NETOBJS = $(BASEOBJS) $(EXTRA_NETOBLS) 55NETOBJS = $(BASEOBJS) $(EXTRA_NETOBLS)
56NETLIBS = $(NETOBJS) $(SOCKETLIBS) 56NETLIBS = $(NETOBJS) $(SOCKETLIBS)
@@ -71,9 +71,9 @@ test-debug:
71 71
72check_apt_LDADD = $(BASEOBJS) 72check_apt_LDADD = $(BASEOBJS)
73check_cluster_LDADD = $(BASEOBJS) 73check_cluster_LDADD = $(BASEOBJS)
74check_curl_CFLAGS = $(AM_CFLAGS) $(LIBCURLCFLAGS) 74check_curl_CFLAGS = $(AM_CFLAGS) $(LIBCURLCFLAGS) -Ipicohttpparser -Iuriparser
75check_curl_CPPFLAGS = $(AM_CPPFLAGS) $(LIBCURLINCLUDE) 75check_curl_CPPFLAGS = $(AM_CPPFLAGS) $(LIBCURLINCLUDE) -Ipicohttpparser -Iuriparser
76check_curl_LDADD = $(NETLIBS) $(LIBCURLLIBS) $(SSLOBJS) libpicohttpparser.a 76check_curl_LDADD = $(NETLIBS) $(LIBCURLLIBS) $(SSLOBJS) picohttpparser/libpicohttpparser.a uriparser/liburiparser.a
77check_dbi_LDADD = $(NETLIBS) $(DBILIBS) 77check_dbi_LDADD = $(NETLIBS) $(DBILIBS)
78check_dig_LDADD = $(NETLIBS) 78check_dig_LDADD = $(NETLIBS)
79check_disk_LDADD = $(BASEOBJS) 79check_disk_LDADD = $(BASEOBJS)
diff --git a/plugins/check_curl.c b/plugins/check_curl.c
index 5f6905f9..72db27ad 100644
--- a/plugins/check_curl.c
+++ b/plugins/check_curl.c
@@ -51,17 +51,35 @@ const char *email = "devel@monitoring-plugins.org";
51 51
52#include "picohttpparser.h" 52#include "picohttpparser.h"
53 53
54#include "uriparser/Uri.h"
55
56#include <arpa/inet.h>
57
54#define MAKE_LIBCURL_VERSION(major, minor, patch) ((major)*0x10000 + (minor)*0x100 + (patch)) 58#define MAKE_LIBCURL_VERSION(major, minor, patch) ((major)*0x10000 + (minor)*0x100 + (patch))
55 59
56#define DEFAULT_BUFFER_SIZE 2048 60#define DEFAULT_BUFFER_SIZE 2048
57#define DEFAULT_SERVER_URL "/" 61#define DEFAULT_SERVER_URL "/"
58#define HTTP_EXPECT "HTTP/1." 62#define HTTP_EXPECT "HTTP/1."
63#define DEFAULT_MAX_REDIRS 15
64#define INET_ADDR_MAX_SIZE INET6_ADDRSTRLEN
59enum { 65enum {
66 MAX_IPV4_HOSTLENGTH = 255,
60 HTTP_PORT = 80, 67 HTTP_PORT = 80,
61 HTTPS_PORT = 443, 68 HTTPS_PORT = 443,
62 MAX_PORT = 65535 69 MAX_PORT = 65535
63}; 70};
64 71
72enum {
73 STICKY_NONE = 0,
74 STICKY_HOST = 1,
75 STICKY_PORT = 2
76};
77
78enum {
79 FOLLOW_HTTP_CURL = 0,
80 FOLLOW_LIBCURL = 1
81};
82
65/* for buffers for header and body */ 83/* for buffers for header and body */
66typedef struct { 84typedef struct {
67 char *buf; 85 char *buf;
@@ -128,6 +146,8 @@ int verbose = 0;
128int show_extended_perfdata = FALSE; 146int show_extended_perfdata = FALSE;
129int min_page_len = 0; 147int min_page_len = 0;
130int max_page_len = 0; 148int max_page_len = 0;
149int redir_depth = 0;
150int max_depth = DEFAULT_MAX_REDIRS;
131char *http_method = NULL; 151char *http_method = NULL;
132char *http_post_data = NULL; 152char *http_post_data = NULL;
133char *http_content_type = NULL; 153char *http_content_type = NULL;
@@ -155,8 +175,12 @@ char string_expect[MAX_INPUT_BUFFER] = "";
155char server_expect[MAX_INPUT_BUFFER] = HTTP_EXPECT; 175char server_expect[MAX_INPUT_BUFFER] = HTTP_EXPECT;
156int server_expect_yn = 0; 176int server_expect_yn = 0;
157char user_auth[MAX_INPUT_BUFFER] = ""; 177char user_auth[MAX_INPUT_BUFFER] = "";
178char **http_opt_headers;
179int http_opt_headers_count = 0;
158int display_html = FALSE; 180int display_html = FALSE;
159int onredirect = STATE_OK; 181int onredirect = STATE_OK;
182int followmethod = FOLLOW_HTTP_CURL;
183int followsticky = STICKY_NONE;
160int use_ssl = FALSE; 184int use_ssl = FALSE;
161int use_sni = TRUE; 185int use_sni = TRUE;
162int check_cert = FALSE; 186int check_cert = FALSE;
@@ -181,6 +205,7 @@ curlhelp_ssl_library ssl_library = CURLHELP_SSL_LIBRARY_UNKNOWN;
181int process_arguments (int, char**); 205int process_arguments (int, char**);
182void handle_curl_option_return_code (CURLcode res, const char* option); 206void handle_curl_option_return_code (CURLcode res, const char* option);
183int check_http (void); 207int check_http (void);
208void redir (curlhelp_write_curlbuf*);
184void print_help (void); 209void print_help (void);
185void print_usage (void); 210void print_usage (void);
186void print_curl_version (void); 211void print_curl_version (void);
@@ -295,6 +320,8 @@ check_http (void)
295{ 320{
296 int result = STATE_OK; 321 int result = STATE_OK;
297 int page_len = 0; 322 int page_len = 0;
323 int i;
324 char *force_host_header = NULL;
298 325
299 /* initialize curl */ 326 /* initialize curl */
300 if (curl_global_init (CURL_GLOBAL_DEFAULT) != CURLE_OK) 327 if (curl_global_init (CURL_GLOBAL_DEFAULT) != CURLE_OK)
@@ -371,6 +398,28 @@ check_http (void)
371 snprintf (http_header, DEFAULT_BUFFER_SIZE, "Connection: close"); 398 snprintf (http_header, DEFAULT_BUFFER_SIZE, "Connection: close");
372 header_list = curl_slist_append (header_list, http_header); 399 header_list = curl_slist_append (header_list, http_header);
373 400
401 /* check if Host header is explicitly set in options */
402 if (http_opt_headers_count) {
403 for (i = 0; i < http_opt_headers_count ; i++) {
404 if (strncmp(http_opt_headers[i], "Host:", 5) == 0) {
405 force_host_header = http_opt_headers[i];
406 }
407 }
408 }
409
410 /* attach additional headers supplied by the user */
411 /* optionally send any other header tag */
412 if (http_opt_headers_count) {
413 for (i = 0; i < http_opt_headers_count ; i++) {
414 if (force_host_header != http_opt_headers[i]) {
415 header_list = curl_slist_append (header_list, http_opt_headers[i]);
416 }
417 }
418 /* This cannot be free'd here because a redirection will then try to access this and segfault */
419 /* Covered in a testcase in tests/check_http.t */
420 /* free(http_opt_headers); */
421 }
422
374 /* set HTTP headers */ 423 /* set HTTP headers */
375 handle_curl_option_return_code (curl_easy_setopt( curl, CURLOPT_HTTPHEADER, header_list ), "CURLOPT_HTTPHEADER"); 424 handle_curl_option_return_code (curl_easy_setopt( curl, CURLOPT_HTTPHEADER, header_list ), "CURLOPT_HTTPHEADER");
376 425
@@ -481,13 +530,29 @@ check_http (void)
481 530
482 /* handle redirections */ 531 /* handle redirections */
483 if (onredirect == STATE_DEPENDENT) { 532 if (onredirect == STATE_DEPENDENT) {
484 handle_curl_option_return_code (curl_easy_setopt (curl, CURLOPT_FOLLOWLOCATION, 1), "CURLOPT_FOLLOWLOCATION"); 533 if( followmethod == FOLLOW_LIBCURL ) {
485 /* TODO: handle the following aspects of redirection 534 handle_curl_option_return_code (curl_easy_setopt (curl, CURLOPT_FOLLOWLOCATION, 1), "CURLOPT_FOLLOWLOCATION");
486 CURLOPT_POSTREDIR: method switch 535
487 CURLINFO_REDIRECT_URL: custom redirect option 536 /* default -1 is infinite, not good, could lead to zombie plugins!
488 CURLOPT_REDIRECT_PROTOCOLS 537 Setting it to one bigger than maximal limit to handle errors nicely below
489 CURLINFO_REDIRECT_COUNT 538 */
490 */ 539 handle_curl_option_return_code (curl_easy_setopt (curl, CURLOPT_MAXREDIRS, max_depth+1), "CURLOPT_MAXREDIRS");
540
541 /* for now allow only http and https (we are a http(s) check plugin in the end) */
542#if LIBCURL_VERSION_NUM >= MAKE_LIBCURL_VERSION(7, 19, 4)
543 handle_curl_option_return_code (curl_easy_setopt (curl, CURLOPT_REDIR_PROTOCOLS, CURLPROTO_HTTP | CURLPROTO_HTTPS), "CURLOPT_REDIRECT_PROTOCOLS");
544#endif /* LIBCURL_VERSION_NUM >= MAKE_LIBCURL_VERSION(7, 19, 4) */
545
546 /* TODO: handle the following aspects of redirection, make them
547 * command line options too later:
548 CURLOPT_POSTREDIR: method switch
549 CURLINFO_REDIRECT_URL: custom redirect option
550 CURLOPT_REDIRECT_PROTOCOLS: allow people to step outside safe protocols
551 CURLINFO_REDIRECT_COUNT: get the number of redirects, print it, maybe a range option here is nice like for expected page size?
552 */
553 } else {
554 // old style redirection is handled below
555 }
491 } 556 }
492 557
493 /* no-body */ 558 /* no-body */
@@ -533,8 +598,8 @@ check_http (void)
533 printf ("**** REQUEST CONTENT ****\n%s\n", http_post_data); 598 printf ("**** REQUEST CONTENT ****\n%s\n", http_post_data);
534 599
535 /* free header and server IP resolve lists, we don't need it anymore */ 600 /* free header and server IP resolve lists, we don't need it anymore */
536 curl_slist_free_all (header_list); 601 curl_slist_free_all (header_list); header_list = NULL;
537 curl_slist_free_all (server_ips); 602 curl_slist_free_all (server_ips); server_ips = NULL;
538 603
539 /* Curl errors, result in critical Nagios state */ 604 /* Curl errors, result in critical Nagios state */
540 if (res != CURLE_OK) { 605 if (res != CURLE_OK) {
@@ -691,18 +756,39 @@ GOT_FIRST_CERT:
691 /* check redirected page if specified */ 756 /* check redirected page if specified */
692 } else if (code >= 300) { 757 } else if (code >= 300) {
693 if (onredirect == STATE_DEPENDENT) { 758 if (onredirect == STATE_DEPENDENT) {
694 code = status_line.http_code; 759 if( followmethod == FOLLOW_LIBCURL ) {
760 code = status_line.http_code;
761 } else {
762 /* old check_http style redirection, if we come
763 * back here, we are in the same status as with
764 * the libcurl method
765 */
766 redir (&header_buf);
767 }
768 } else {
769 /* this is a specific code in the command line to
770 * be returned when a redirection is encoutered
771 */
695 } 772 }
696 result = max_state_alt (onredirect, result); 773 result = max_state_alt (onredirect, result);
697 /* TODO: make sure the last status line has been
698 parsed into the status_line structure
699 */
700 /* all other codes are considered ok */ 774 /* all other codes are considered ok */
701 } else { 775 } else {
702 result = STATE_OK; 776 result = STATE_OK;
703 } 777 }
704 } 778 }
705 779
780 /* libcurl redirection internally, handle error states here */
781 if( followmethod == FOLLOW_LIBCURL ) {
782 handle_curl_option_return_code (curl_easy_getinfo (curl, CURLINFO_REDIRECT_COUNT, &redir_depth), "CURLINFO_REDIRECT_COUNT");
783 if (verbose >= 2)
784 printf(_("* curl LIBINFO_REDIRECT_COUNT is %d\n"), redir_depth);
785 if (redir_depth > max_depth) {
786 snprintf (msg, DEFAULT_BUFFER_SIZE, "maximum redirection depth %d exceeded in libcurl",
787 max_depth);
788 die (STATE_WARNING, "HTTP WARNING - %s", msg);
789 }
790 }
791
706 /* check status codes, set exit status accordingly */ 792 /* check status codes, set exit status accordingly */
707 if( status_line.http_code != code ) { 793 if( status_line.http_code != code ) {
708 die (STATE_CRITICAL, _("HTTP CRITICAL HTTP/%d.%d %d %s - different HTTP codes (cUrl has %ld)\n"), 794 die (STATE_CRITICAL, _("HTTP CRITICAL HTTP/%d.%d %d %s - different HTTP codes (cUrl has %ld)\n"),
@@ -813,6 +899,188 @@ GOT_FIRST_CERT:
813 return result; 899 return result;
814} 900}
815 901
902int
903uri_strcmp (const UriTextRangeA range, const char* s)
904{
905 if (!range.first) return -1;
906 if (range.afterLast - range.first < strlen (s)) return -1;
907 return strncmp (s, range.first, min( range.afterLast - range.first, strlen (s)));
908}
909
910char*
911uri_string (const UriTextRangeA range, char* buf, size_t buflen)
912{
913 if (!range.first) return "(null)";
914 strncpy (buf, range.first, max (buflen, range.afterLast - range.first));
915 buf[max (buflen, range.afterLast - range.first)] = '\0';
916 buf[range.afterLast - range.first] = '\0';
917 return buf;
918}
919
920void
921redir (curlhelp_write_curlbuf* header_buf)
922{
923 char *location = NULL;
924 curlhelp_statusline status_line;
925 struct phr_header headers[255];
926 size_t nof_headers = 255;
927 size_t msglen;
928 char buf[DEFAULT_BUFFER_SIZE];
929 char ipstr[INET_ADDR_MAX_SIZE];
930 int new_port;
931 char *new_host;
932 char *new_url;
933
934 int res = phr_parse_response (header_buf->buf, header_buf->buflen,
935 &status_line.http_minor, &status_line.http_code, &status_line.msg, &msglen,
936 headers, &nof_headers, 0);
937
938 location = get_header_value (headers, nof_headers, "location");
939
940 if (verbose >= 2)
941 printf(_("* Seen redirect location %s\n"), location);
942
943 if (++redir_depth > max_depth)
944 die (STATE_WARNING,
945 _("HTTP WARNING - maximum redirection depth %d exceeded - %s\n"),
946 max_depth, location, (display_html ? "</A>" : ""));
947
948 UriParserStateA state;
949 UriUriA uri;
950 state.uri = &uri;
951 if (uriParseUriA (&state, location) != URI_SUCCESS) {
952 if (state.errorCode == URI_ERROR_SYNTAX) {
953 die (STATE_UNKNOWN,
954 _("HTTP UNKNOWN - Could not parse redirect location '%s'%s\n"),
955 location, (display_html ? "</A>" : ""));
956 } else if (state.errorCode == URI_ERROR_MALLOC) {
957 die (STATE_UNKNOWN, _("HTTP UNKNOWN - Could not allocate URL\n"));
958 }
959 }
960
961 if (verbose >= 2) {
962 printf (_("** scheme: %s\n"),
963 uri_string (uri.scheme, buf, DEFAULT_BUFFER_SIZE));
964 printf (_("** host: %s\n"),
965 uri_string (uri.hostText, buf, DEFAULT_BUFFER_SIZE));
966 printf (_("** port: %s\n"),
967 uri_string (uri.portText, buf, DEFAULT_BUFFER_SIZE));
968 if (uri.hostData.ip4) {
969 inet_ntop (AF_INET, uri.hostData.ip4->data, ipstr, sizeof (ipstr));
970 printf (_("** IPv4: %s\n"), ipstr);
971 }
972 if (uri.hostData.ip6) {
973 inet_ntop (AF_INET, uri.hostData.ip6->data, ipstr, sizeof (ipstr));
974 printf (_("** IPv6: %s\n"), ipstr);
975 }
976 if (uri.pathHead) {
977 printf (_("** path: "));
978 const UriPathSegmentA* p = uri.pathHead;
979 for (; p; p = p->next) {
980 printf ("/%s", uri_string (p->text, buf, DEFAULT_BUFFER_SIZE));
981 }
982 puts ("");
983 }
984 if (uri.query.first) {
985 printf (_("** query: %s\n"),
986 uri_string (uri.query, buf, DEFAULT_BUFFER_SIZE));
987 }
988 if (uri.fragment.first) {
989 printf (_("** fragment: %s\n"),
990 uri_string (uri.fragment, buf, DEFAULT_BUFFER_SIZE));
991 }
992 }
993
994 use_ssl = !uri_strcmp (uri.scheme, "https");
995
996 /* we do a sloppy test here only, because uriparser would have failed
997 * above, if the port would be invalid, we just check for MAX_PORT
998 */
999 if (uri.portText.first) {
1000 new_port = atoi (uri_string (uri.portText, buf, DEFAULT_BUFFER_SIZE));
1001 } else {
1002 new_port = HTTP_PORT;
1003 if (use_ssl)
1004 new_port = HTTPS_PORT;
1005 }
1006 if (new_port > MAX_PORT)
1007 die (STATE_UNKNOWN,
1008 _("HTTP UNKNOWN - Redirection to port above %d - %s\n"),
1009 MAX_PORT, location, display_html ? "</A>" : "");
1010
1011 /* by RFC 7231 relative URLs in Location should be taken relative to
1012 * the original URL, so wy try to form a new absolute URL here
1013 */
1014 if (!uri.scheme.first && !uri.hostText.first) {
1015 /* TODO: implement */
1016 die (STATE_UNKNOWN, _("HTTP UNKNOWN - non-absolute location, not implemented yet!\n"));
1017 new_host = strdup (host_name ? host_name : server_address);
1018 } else {
1019 new_host = strdup (uri_string (uri.hostText, buf, DEFAULT_BUFFER_SIZE));
1020 }
1021
1022 new_url = (char *)calloc( 1, DEFAULT_BUFFER_SIZE);
1023 if (uri.pathHead) {
1024 const UriPathSegmentA* p = uri.pathHead;
1025 for (; p; p = p->next) {
1026 strncat (new_url, "/", DEFAULT_BUFFER_SIZE);
1027 strncat (new_url, uri_string (p->text, buf, DEFAULT_BUFFER_SIZE), DEFAULT_BUFFER_SIZE);
1028 }
1029 }
1030
1031 if (server_port==new_port &&
1032 !strncmp(server_address, new_host, MAX_IPV4_HOSTLENGTH) &&
1033 (host_name && !strncmp(host_name, new_host, MAX_IPV4_HOSTLENGTH)) &&
1034 !strcmp(server_url, new_url))
1035 die (STATE_WARNING,
1036 _("HTTP WARNING - redirection creates an infinite loop - %s://%s:%d%s%s\n"),
1037 use_ssl ? "https" : "http", new_host, new_port, new_url, (display_html ? "</A>" : ""));
1038
1039 /* set new values for redirected request */
1040
1041 free (host_name);
1042 host_name = strndup (new_host, MAX_IPV4_HOSTLENGTH);
1043 free(new_host);
1044
1045 server_port = (unsigned short)new_port;
1046
1047 /* reset virtual port */
1048 virtual_port = server_port;
1049
1050 if (!(followsticky & STICKY_HOST)) {
1051 free (server_address);
1052 server_address = strndup (new_host, MAX_IPV4_HOSTLENGTH);
1053 }
1054 if (!(followsticky & STICKY_PORT)) {
1055 server_port = new_port;
1056 }
1057
1058 free (server_url);
1059 server_url = new_url;
1060
1061 uriFreeUriMembersA (&uri);
1062
1063 if (verbose)
1064 printf (_("Redirection to %s://%s:%d%s\n"), use_ssl ? "https" : "http",
1065 host_name ? host_name : server_address, server_port, server_url);
1066
1067 /* TODO: the hash component MUST be taken from the original URL and
1068 * attached to the URL in Location
1069 */
1070
1071 check_http ();
1072}
1073
1074#if 0
1075
1076int main(int argc, char *argv[]) {
1077
1078 for (; i < argc; i++) {
1079
1080 }
1081 printf("\n");
1082#endif
1083
816/* check whether a file exists */ 1084/* check whether a file exists */
817void 1085void
818test_file (char *path) 1086test_file (char *path)
@@ -974,7 +1242,11 @@ process_arguments (int argc, char **argv)
974 snprintf (user_agent, DEFAULT_BUFFER_SIZE, optarg); 1242 snprintf (user_agent, DEFAULT_BUFFER_SIZE, optarg);
975 break; 1243 break;
976 case 'k': /* Additional headers */ 1244 case 'k': /* Additional headers */
977 header_list = curl_slist_append(header_list, optarg); 1245 if (http_opt_headers_count == 0)
1246 http_opt_headers = malloc (sizeof (char *) * (++http_opt_headers_count));
1247 else
1248 http_opt_headers = realloc (http_opt_headers, sizeof (char *) * (++http_opt_headers_count));
1249 http_opt_headers[http_opt_headers_count - 1] = optarg;
978 break; 1250 break;
979 case 'L': /* show html link */ 1251 case 'L': /* show html link */
980 display_html = TRUE; 1252 display_html = TRUE;
@@ -1121,6 +1393,14 @@ process_arguments (int argc, char **argv)
1121 onredirect = STATE_UNKNOWN; 1393 onredirect = STATE_UNKNOWN;
1122 else if (!strcmp (optarg, "follow")) 1394 else if (!strcmp (optarg, "follow"))
1123 onredirect = STATE_DEPENDENT; 1395 onredirect = STATE_DEPENDENT;
1396 else if (!strcmp (optarg, "stickyport"))
1397 onredirect = STATE_DEPENDENT, followmethod = FOLLOW_HTTP_CURL, followsticky = STICKY_HOST|STICKY_PORT;
1398 else if (!strcmp (optarg, "sticky"))
1399 onredirect = STATE_DEPENDENT, followmethod = FOLLOW_HTTP_CURL, followsticky = STICKY_HOST;
1400 else if (!strcmp (optarg, "follow"))
1401 onredirect = STATE_DEPENDENT, followmethod = FOLLOW_HTTP_CURL, followsticky = STICKY_NONE;
1402 else if (!strcmp (optarg, "curl"))
1403 onredirect = STATE_DEPENDENT, followmethod = FOLLOW_LIBCURL;
1124 else usage2 (_("Invalid onredirect option"), optarg); 1404 else usage2 (_("Invalid onredirect option"), optarg);
1125 if (verbose >= 2) 1405 if (verbose >= 2)
1126 printf(_("* Following redirects set to %s\n"), state_text(onredirect)); 1406 printf(_("* Following redirects set to %s\n"), state_text(onredirect));
@@ -1264,7 +1544,6 @@ print_help (void)
1264 print_revision (progname, NP_VERSION); 1544 print_revision (progname, NP_VERSION);
1265 1545
1266 printf ("Copyright (c) 1999 Ethan Galstad <nagios@nagios.org>\n"); 1546 printf ("Copyright (c) 1999 Ethan Galstad <nagios@nagios.org>\n");
1267 printf ("Copyright (c) 2017 Andreas Baumann <mail@andreasbaumann.cc>\n");
1268 printf (COPYRIGHT, copyright, email); 1547 printf (COPYRIGHT, copyright, email);
1269 1548
1270 printf ("%s\n", _("This plugin tests the HTTP service on the specified host. It can test")); 1549 printf ("%s\n", _("This plugin tests the HTTP service on the specified host. It can test"));
@@ -1365,8 +1644,11 @@ print_help (void)
1365 printf (" %s\n", _("Print additional performance data")); 1644 printf (" %s\n", _("Print additional performance data"));
1366 printf (" %s\n", "-L, --link"); 1645 printf (" %s\n", "-L, --link");
1367 printf (" %s\n", _("Wrap output in HTML link (obsoleted by urlize)")); 1646 printf (" %s\n", _("Wrap output in HTML link (obsoleted by urlize)"));
1368 printf (" %s\n", "-f, --onredirect=<ok|warning|critical|follow>"); 1647 printf (" %s\n", "-f, --onredirect=<ok|warning|critical|follow|sticky|stickyport|curl>");
1369 printf (" %s\n", _("How to handle redirected pages.")); 1648 printf (" %s\n", _("How to handle redirected pages. sticky is like follow but stick to the"));
1649 printf (" %s\n", _("specified IP address. stickyport also ensures port stays the same."));
1650 printf (" %s\n", _("follow uses the old redirection algorithm of check_http."));
1651 printf (" %s\n", _("curl uses CURL_FOLLOWLOCATION built into libcurl."));
1370 printf (" %s\n", "-m, --pagesize=INTEGER<:INTEGER>"); 1652 printf (" %s\n", "-m, --pagesize=INTEGER<:INTEGER>");
1371 printf (" %s\n", _("Minimum page size required (bytes) : Maximum page size required (bytes)")); 1653 printf (" %s\n", _("Minimum page size required (bytes) : Maximum page size required (bytes)"));
1372 1654
@@ -1436,7 +1718,7 @@ print_usage (void)
1436 printf (" %s -H <vhost> | -I <IP-address> [-u <uri>] [-p <port>]\n",progname); 1718 printf (" %s -H <vhost> | -I <IP-address> [-u <uri>] [-p <port>]\n",progname);
1437 printf (" [-J <client certificate file>] [-K <private key>] [--ca-cert <CA certificate file>]\n"); 1719 printf (" [-J <client certificate file>] [-K <private key>] [--ca-cert <CA certificate file>]\n");
1438 printf (" [-w <warn time>] [-c <critical time>] [-t <timeout>] [-L] [-E] [-a auth]\n"); 1720 printf (" [-w <warn time>] [-c <critical time>] [-t <timeout>] [-L] [-E] [-a auth]\n");
1439 printf (" [-f <ok|warning|critcal|follow>]\n"); 1721 printf (" [-f <ok|warning|critcal|follow|sticky|stickyport|curl>]\n");
1440 printf (" [-e <expect>] [-d string] [-s string] [-l] [-r <regex> | -R <case-insensitive regex>]\n"); 1722 printf (" [-e <expect>] [-d string] [-s string] [-l] [-r <regex> | -R <case-insensitive regex>]\n");
1441 printf (" [-P string] [-m <min_pg_size>:<max_pg_size>] [-4|-6] [-N] [-M <age>]\n"); 1723 printf (" [-P string] [-m <min_pg_size>:<max_pg_size>] [-4|-6] [-N] [-M <age>]\n");
1442 printf (" [-A string] [-k string] [-S <version>] [--sni] [-C <warn_age>[,<crit_age>]]\n"); 1724 printf (" [-A string] [-k string] [-S <version>] [--sni] [-C <warn_age>[,<crit_age>]]\n");
@@ -1739,7 +2021,7 @@ get_content_length (const curlhelp_write_curlbuf* header_buf, const curlhelp_wri
1739 content_length_s += strspn (content_length_s, " \t"); 2021 content_length_s += strspn (content_length_s, " \t");
1740 content_length = atoi (content_length_s); 2022 content_length = atoi (content_length_s);
1741 if (content_length != body_buf->buflen) { 2023 if (content_length != body_buf->buflen) {
1742 /* TODO: should we warn if the actual and the reported body length doen't match? */ 2024 /* TODO: should we warn if the actual and the reported body length don't match? */
1743 } 2025 }
1744 2026
1745 if (content_length_s) free (content_length_s); 2027 if (content_length_s) free (content_length_s);
diff --git a/plugins/picohttpparser/Makefile.am b/plugins/picohttpparser/Makefile.am
new file mode 100644
index 00000000..5a610278
--- /dev/null
+++ b/plugins/picohttpparser/Makefile.am
@@ -0,0 +1,3 @@
1noinst_LIBRARIES = libpicohttpparser.a
2
3libpicohttpparser_a_SOURCES = picohttpparser.c
diff --git a/plugins/picohttpparser.c b/plugins/picohttpparser/picohttpparser.c
index 6a2d872d..6a2d872d 100644
--- a/plugins/picohttpparser.c
+++ b/plugins/picohttpparser/picohttpparser.c
diff --git a/plugins/picohttpparser.h b/plugins/picohttpparser/picohttpparser.h
index a8fad71d..a8fad71d 100644
--- a/plugins/picohttpparser.h
+++ b/plugins/picohttpparser/picohttpparser.h
diff --git a/plugins/uriparser/Makefile.am b/plugins/uriparser/Makefile.am
new file mode 100644
index 00000000..592e694c
--- /dev/null
+++ b/plugins/uriparser/Makefile.am
@@ -0,0 +1,9 @@
1noinst_LIBRARIES = liburiparser.a
2
3liburiparser_a_SOURCES = UriCommon.c UriCompare.c \
4 UriEscape.c UriFile.c UriIp4.c \
5 UriIp4Base.c UriNormalize.c \
6 UriNormalizeBase.c UriParse.c \
7 UriParseBase.c UriQuery.c \
8 UriRecompose.c UriResolve.c \
9 UriShorten.c
diff --git a/plugins/uriparser/Uri.h b/plugins/uriparser/Uri.h
new file mode 100644
index 00000000..4a185808
--- /dev/null
+++ b/plugins/uriparser/Uri.h
@@ -0,0 +1,777 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file Uri.h
42 * Holds the RFC 3986 %URI parser interface.
43 * NOTE: This header includes itself twice.
44 */
45
46#if (defined(URI_PASS_ANSI) && !defined(URI_H_ANSI)) \
47 || (defined(URI_PASS_UNICODE) && !defined(URI_H_UNICODE)) \
48 || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
49/* What encodings are enabled? */
50#include "UriDefsConfig.h"
51#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
52/* Include SELF twice */
53# ifdef URI_ENABLE_ANSI
54# define URI_PASS_ANSI 1
55# include "Uri.h"
56# undef URI_PASS_ANSI
57# endif
58# ifdef URI_ENABLE_UNICODE
59# define URI_PASS_UNICODE 1
60# include "Uri.h"
61# undef URI_PASS_UNICODE
62# endif
63/* Only one pass for each encoding */
64#elif (defined(URI_PASS_ANSI) && !defined(URI_H_ANSI) \
65 && defined(URI_ENABLE_ANSI)) || (defined(URI_PASS_UNICODE) \
66 && !defined(URI_H_UNICODE) && defined(URI_ENABLE_UNICODE))
67# ifdef URI_PASS_ANSI
68# define URI_H_ANSI 1
69# include "UriDefsAnsi.h"
70# else
71# define URI_H_UNICODE 1
72# include "UriDefsUnicode.h"
73# endif
74
75
76
77#ifdef __cplusplus
78extern "C" {
79#endif
80
81
82
83#ifndef URI_DOXYGEN
84# include "UriBase.h"
85#endif
86
87
88
89/**
90 * Specifies a range of characters within a string.
91 * The range includes all characters from <c>first</c>
92 * to one before <c>afterLast</c>. So if both are
93 * non-NULL the difference is the length of the text range.
94 *
95 * @see UriUriA
96 * @see UriPathSegmentA
97 * @see UriHostDataA
98 * @since 0.3.0
99 */
100typedef struct URI_TYPE(TextRangeStruct) {
101 const URI_CHAR * first; /**< Pointer to first character */
102 const URI_CHAR * afterLast; /**< Pointer to character after the last one still in */
103} URI_TYPE(TextRange); /**< @copydoc UriTextRangeStructA */
104
105
106
107/**
108 * Represents a path segment within a %URI path.
109 * More precisely it is a node in a linked
110 * list of path segments.
111 *
112 * @see UriUriA
113 * @since 0.3.0
114 */
115typedef struct URI_TYPE(PathSegmentStruct) {
116 URI_TYPE(TextRange) text; /**< Path segment name */
117 struct URI_TYPE(PathSegmentStruct) * next; /**< Pointer to the next path segment in the list, can be NULL if last already */
118
119 void * reserved; /**< Reserved to the parser */
120} URI_TYPE(PathSegment); /**< @copydoc UriPathSegmentStructA */
121
122
123
124/**
125 * Holds structured host information.
126 * This is either a IPv4, IPv6, plain
127 * text for IPvFuture or all zero for
128 * a registered name.
129 *
130 * @see UriUriA
131 * @since 0.3.0
132 */
133typedef struct URI_TYPE(HostDataStruct) {
134 UriIp4 * ip4; /**< IPv4 address */
135 UriIp6 * ip6; /**< IPv6 address */
136 URI_TYPE(TextRange) ipFuture; /**< IPvFuture address */
137} URI_TYPE(HostData); /**< @copydoc UriHostDataStructA */
138
139
140
141/**
142 * Represents an RFC 3986 %URI.
143 * Missing components can be {NULL, NULL} ranges.
144 *
145 * @see uriParseUriA
146 * @see uriFreeUriMembersA
147 * @see UriParserStateA
148 * @since 0.3.0
149 */
150typedef struct URI_TYPE(UriStruct) {
151 URI_TYPE(TextRange) scheme; /**< Scheme (e.g. "http") */
152 URI_TYPE(TextRange) userInfo; /**< User info (e.g. "user:pass") */
153 URI_TYPE(TextRange) hostText; /**< Host text (set for all hosts, excluding square brackets) */
154 URI_TYPE(HostData) hostData; /**< Structured host type specific data */
155 URI_TYPE(TextRange) portText; /**< Port (e.g. "80") */
156 URI_TYPE(PathSegment) * pathHead; /**< Head of a linked list of path segments */
157 URI_TYPE(PathSegment) * pathTail; /**< Tail of the list behind pathHead */
158 URI_TYPE(TextRange) query; /**< Query without leading "?" */
159 URI_TYPE(TextRange) fragment; /**< Query without leading "#" */
160 UriBool absolutePath; /**< Absolute path flag, distincting "a" and "/a" */
161 UriBool owner; /**< Memory owner flag */
162
163 void * reserved; /**< Reserved to the parser */
164} URI_TYPE(Uri); /**< @copydoc UriUriStructA */
165
166
167
168/**
169 * Represents a state of the %URI parser.
170 * Missing components can be NULL to reflect
171 * a components absence.
172 *
173 * @see uriFreeUriMembersA
174 * @since 0.3.0
175 */
176typedef struct URI_TYPE(ParserStateStruct) {
177 URI_TYPE(Uri) * uri; /**< Plug in the %URI structure to be filled while parsing here */
178 int errorCode; /**< Code identifying the occured error */
179 const URI_CHAR * errorPos; /**< Pointer to position in case of a syntax error */
180
181 void * reserved; /**< Reserved to the parser */
182} URI_TYPE(ParserState); /**< @copydoc UriParserStateStructA */
183
184
185
186/**
187 * Represents a query element.
188 * More precisely it is a node in a linked
189 * list of query elements.
190 *
191 * @since 0.7.0
192 */
193typedef struct URI_TYPE(QueryListStruct) {
194 const URI_CHAR * key; /**< Key of the query element */
195 const URI_CHAR * value; /**< Value of the query element, can be NULL */
196
197 struct URI_TYPE(QueryListStruct) * next; /**< Pointer to the next key/value pair in the list, can be NULL if last already */
198} URI_TYPE(QueryList); /**< @copydoc UriQueryListStructA */
199
200
201
202/**
203 * Parses a RFC 3986 URI.
204 *
205 * @param state <b>INOUT</b>: Parser state with set output %URI, must not be NULL
206 * @param first <b>IN</b>: Pointer to the first character to parse, must not be NULL
207 * @param afterLast <b>IN</b>: Pointer to the character after the last to parse, must not be NULL
208 * @return 0 on success, error code otherwise
209 *
210 * @see uriParseUriA
211 * @see uriToStringA
212 * @since 0.3.0
213 */
214int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state,
215 const URI_CHAR * first, const URI_CHAR * afterLast);
216
217
218
219/**
220 * Parses a RFC 3986 %URI.
221 *
222 * @param state <b>INOUT</b>: Parser state with set output %URI, must not be NULL
223 * @param text <b>IN</b>: Text to parse, must not be NULL
224 * @return 0 on success, error code otherwise
225 *
226 * @see uriParseUriExA
227 * @see uriToStringA
228 * @since 0.3.0
229 */
230int URI_FUNC(ParseUri)(URI_TYPE(ParserState) * state,
231 const URI_CHAR * text);
232
233
234
235/**
236 * Frees all memory associated with the members
237 * of the %URI structure. Note that the structure
238 * itself is not freed, only its members.
239 *
240 * @param uri <b>INOUT</b>: %URI structure whose members should be freed
241 *
242 * @since 0.3.0
243 */
244void URI_FUNC(FreeUriMembers)(URI_TYPE(Uri) * uri);
245
246
247
248/**
249 * Percent-encodes all unreserved characters from the input string and
250 * writes the encoded version to the output string.
251 * Be sure to allocate <b>3 times</b> the space of the input buffer for
252 * the output buffer for <c>normalizeBreaks == URI_FALSE</c> and <b>6 times</b>
253 * the space for <c>normalizeBreaks == URI_TRUE</c>
254 * (since e.g. "\x0d" becomes "%0D%0A" in that case)
255 *
256 * @param inFirst <b>IN</b>: Pointer to first character of the input text
257 * @param inAfterLast <b>IN</b>: Pointer after the last character of the input text
258 * @param out <b>OUT</b>: Encoded text destination
259 * @param spaceToPlus <b>IN</b>: Wether to convert ' ' to '+' or not
260 * @param normalizeBreaks <b>IN</b>: Wether to convert CR and LF to CR-LF or not.
261 * @return Position of terminator in output string
262 *
263 * @see uriEscapeA
264 * @see uriUnescapeInPlaceExA
265 * @since 0.5.2
266 */
267URI_CHAR * URI_FUNC(EscapeEx)(const URI_CHAR * inFirst,
268 const URI_CHAR * inAfterLast, URI_CHAR * out,
269 UriBool spaceToPlus, UriBool normalizeBreaks);
270
271
272
273/**
274 * Percent-encodes all unreserved characters from the input string and
275 * writes the encoded version to the output string.
276 * Be sure to allocate <b>3 times</b> the space of the input buffer for
277 * the output buffer for <c>normalizeBreaks == URI_FALSE</c> and <b>6 times</b>
278 * the space for <c>normalizeBreaks == URI_TRUE</c>
279 * (since e.g. "\x0d" becomes "%0D%0A" in that case)
280 *
281 * @param in <b>IN</b>: Text source
282 * @param out <b>OUT</b>: Encoded text destination
283 * @param spaceToPlus <b>IN</b>: Wether to convert ' ' to '+' or not
284 * @param normalizeBreaks <b>IN</b>: Wether to convert CR and LF to CR-LF or not.
285 * @return Position of terminator in output string
286 *
287 * @see uriEscapeExA
288 * @see uriUnescapeInPlaceA
289 * @since 0.5.0
290 */
291URI_CHAR * URI_FUNC(Escape)(const URI_CHAR * in, URI_CHAR * out,
292 UriBool spaceToPlus, UriBool normalizeBreaks);
293
294
295
296/**
297 * Unescapes percent-encoded groups in a given string.
298 * E.g. "%20" will become " ". Unescaping is done in place.
299 * The return value will be point to the new position
300 * of the terminating zero. Use this value to get the new
301 * length of the string. NULL is only returned if <c>inout</c>
302 * is NULL.
303 *
304 * @param inout <b>INOUT</b>: Text to unescape/decode
305 * @param plusToSpace <b>IN</b>: Whether to convert '+' to ' ' or not
306 * @param breakConversion <b>IN</b>: Line break conversion mode
307 * @return Pointer to new position of the terminating zero
308 *
309 * @see uriUnescapeInPlaceA
310 * @see uriEscapeExA
311 * @since 0.5.0
312 */
313const URI_CHAR * URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout,
314 UriBool plusToSpace, UriBreakConversion breakConversion);
315
316
317
318/**
319 * Unescapes percent-encoded groups in a given string.
320 * E.g. "%20" will become " ". Unescaping is done in place.
321 * The return value will be point to the new position
322 * of the terminating zero. Use this value to get the new
323 * length of the string. NULL is only returned if <c>inout</c>
324 * is NULL.
325 *
326 * NOTE: '+' is not decoded to ' ' and line breaks are not converted.
327 * Use the more advanced UnescapeInPlaceEx for that features instead.
328 *
329 * @param inout <b>INOUT</b>: Text to unescape/decode
330 * @return Pointer to new position of the terminating zero
331 *
332 * @see uriUnescapeInPlaceExA
333 * @see uriEscapeA
334 * @since 0.3.0
335 */
336const URI_CHAR * URI_FUNC(UnescapeInPlace)(URI_CHAR * inout);
337
338
339
340/**
341 * Performs reference resolution as described in
342 * <a href="http://tools.ietf.org/html/rfc3986#section-5.2.2">section 5.2.2 of RFC 3986</a>.
343 * NOTE: On success you have to call uriFreeUriMembersA on \p absoluteDest manually later.
344 *
345 * @param absoluteDest <b>OUT</b>: Result %URI
346 * @param relativeSource <b>IN</b>: Reference to resolve
347 * @param absoluteBase <b>IN</b>: Base %URI to apply
348 * @return Error code or 0 on success
349 *
350 * @see uriRemoveBaseUriA, uriAddBaseUriExA
351 * @since 0.4.0
352 */
353int URI_FUNC(AddBaseUri)(URI_TYPE(Uri) * absoluteDest,
354 const URI_TYPE(Uri) * relativeSource,
355 const URI_TYPE(Uri) * absoluteBase);
356
357
358
359/**
360 * Performs reference resolution as described in
361 * <a href="http://tools.ietf.org/html/rfc3986#section-5.2.2">section 5.2.2 of RFC 3986</a>.
362 * NOTE: On success you have to call uriFreeUriMembersA on \p absoluteDest manually later.
363 *
364 * @param absoluteDest <b>OUT</b>: Result %URI
365 * @param relativeSource <b>IN</b>: Reference to resolve
366 * @param absoluteBase <b>IN</b>: Base %URI to apply
367 * @param options <b>IN</b>: Configuration to apply
368 * @return Error code or 0 on success
369 *
370 * @see uriRemoveBaseUriA, uriAddBaseUriA
371 * @since 0.8.1
372 */
373int URI_FUNC(AddBaseUriEx)(URI_TYPE(Uri) * absoluteDest,
374 const URI_TYPE(Uri) * relativeSource,
375 const URI_TYPE(Uri) * absoluteBase,
376 UriResolutionOptions options);
377
378
379
380/**
381 * Tries to make a relative %URI (a reference) from an
382 * absolute %URI and a given base %URI. This can only work if
383 * the absolute %URI shares scheme and authority with
384 * the base %URI. If it does not the result will still be
385 * an absolute URI (with scheme part if necessary).
386 * NOTE: On success you have to call uriFreeUriMembersA on
387 * \p dest manually later.
388 *
389 * @param dest <b>OUT</b>: Result %URI
390 * @param absoluteSource <b>IN</b>: Absolute %URI to make relative
391 * @param absoluteBase <b>IN</b>: Base %URI
392 * @param domainRootMode <b>IN</b>: Create %URI with path relative to domain root
393 * @return Error code or 0 on success
394 *
395 * @see uriAddBaseUriA, uriAddBaseUriExA
396 * @since 0.5.2
397 */
398int URI_FUNC(RemoveBaseUri)(URI_TYPE(Uri) * dest,
399 const URI_TYPE(Uri) * absoluteSource,
400 const URI_TYPE(Uri) * absoluteBase,
401 UriBool domainRootMode);
402
403
404
405/**
406 * Checks two URIs for equivalence. Comparison is done
407 * the naive way, without prior normalization.
408 * NOTE: Two <c>NULL</c> URIs are equal as well.
409 *
410 * @param a <b>IN</b>: First %URI
411 * @param b <b>IN</b>: Second %URI
412 * @return <c>URI_TRUE</c> when equal, <c>URI_FAlSE</c> else
413 *
414 * @since 0.4.0
415 */
416UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a, const URI_TYPE(Uri) * b);
417
418
419
420/**
421 * Calculates the number of characters needed to store the
422 * string representation of the given %URI excluding the
423 * terminator.
424 *
425 * @param uri <b>IN</b>: %URI to measure
426 * @param charsRequired <b>OUT</b>: Length of the string representation in characters <b>excluding</b> terminator
427 * @return Error code or 0 on success
428 *
429 * @see uriToStringA
430 * @since 0.5.0
431 */
432int URI_FUNC(ToStringCharsRequired)(const URI_TYPE(Uri) * uri,
433 int * charsRequired);
434
435
436
437/**
438 * Converts a %URI structure back to text as described in
439 * <a href="http://tools.ietf.org/html/rfc3986#section-5.3">section 5.3 of RFC 3986</a>.
440 *
441 * @param dest <b>OUT</b>: Output destination
442 * @param uri <b>IN</b>: %URI to convert
443 * @param maxChars <b>IN</b>: Maximum number of characters to copy <b>including</b> terminator
444 * @param charsWritten <b>OUT</b>: Number of characters written, can be lower than maxChars even if the %URI is too long!
445 * @return Error code or 0 on success
446 *
447 * @see uriToStringCharsRequiredA
448 * @since 0.4.0
449 */
450int URI_FUNC(ToString)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, int maxChars, int * charsWritten);
451
452
453
454/**
455 * Determines the components of a %URI that are not normalized.
456 *
457 * @param uri <b>IN</b>: %URI to check
458 * @return Normalization job mask
459 *
460 * @see uriNormalizeSyntaxA
461 * @since 0.5.0
462 */
463unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)(const URI_TYPE(Uri) * uri);
464
465
466
467/**
468 * Normalizes a %URI using a normalization mask.
469 * The normalization mask decides what components are normalized.
470 *
471 * NOTE: If necessary the %URI becomes owner of all memory
472 * behind the text pointed to. Text is duplicated in that case.
473 *
474 * @param uri <b>INOUT</b>: %URI to normalize
475 * @param mask <b>IN</b>: Normalization mask
476 * @return Error code or 0 on success
477 *
478 * @see uriNormalizeSyntaxA
479 * @see uriNormalizeSyntaxMaskRequiredA
480 * @since 0.5.0
481 */
482int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, unsigned int mask);
483
484
485
486/**
487 * Normalizes all components of a %URI.
488 *
489 * NOTE: If necessary the %URI becomes owner of all memory
490 * behind the text pointed to. Text is duplicated in that case.
491 *
492 * @param uri <b>INOUT</b>: %URI to normalize
493 * @return Error code or 0 on success
494 *
495 * @see uriNormalizeSyntaxExA
496 * @see uriNormalizeSyntaxMaskRequiredA
497 * @since 0.5.0
498 */
499int URI_FUNC(NormalizeSyntax)(URI_TYPE(Uri) * uri);
500
501
502
503/**
504 * Converts a Unix filename to a %URI string.
505 * The destination buffer must be large enough to hold 7 + 3 * len(filename) + 1
506 * characters in case of an absolute filename or 3 * len(filename) + 1 in case
507 * of a relative filename.
508 *
509 * EXAMPLE
510 * Input: "/bin/bash"
511 * Output: "file:///bin/bash"
512 *
513 * @param filename <b>IN</b>: Unix filename to convert
514 * @param uriString <b>OUT</b>: Destination to write %URI string to
515 * @return Error code or 0 on success
516 *
517 * @see uriUriStringToUnixFilenameA
518 * @see uriWindowsFilenameToUriStringA
519 * @since 0.5.2
520 */
521int URI_FUNC(UnixFilenameToUriString)(const URI_CHAR * filename,
522 URI_CHAR * uriString);
523
524
525
526/**
527 * Converts a Windows filename to a %URI string.
528 * The destination buffer must be large enough to hold 8 + 3 * len(filename) + 1
529 * characters in case of an absolute filename or 3 * len(filename) + 1 in case
530 * of a relative filename.
531 *
532 * EXAMPLE
533 * Input: "E:\\Documents and Settings"
534 * Output: "file:///E:/Documents%20and%20Settings"
535 *
536 * @param filename <b>IN</b>: Windows filename to convert
537 * @param uriString <b>OUT</b>: Destination to write %URI string to
538 * @return Error code or 0 on success
539 *
540 * @see uriUriStringToWindowsFilenameA
541 * @see uriUnixFilenameToUriStringA
542 * @since 0.5.2
543 */
544int URI_FUNC(WindowsFilenameToUriString)(const URI_CHAR * filename,
545 URI_CHAR * uriString);
546
547
548
549/**
550 * Extracts a Unix filename from a %URI string.
551 * The destination buffer must be large enough to hold len(uriString) + 1 - 7
552 * characters in case of an absolute %URI or len(uriString) + 1 in case
553 * of a relative %URI.
554 *
555 * @param uriString <b>IN</b>: %URI string to convert
556 * @param filename <b>OUT</b>: Destination to write filename to
557 * @return Error code or 0 on success
558 *
559 * @see uriUnixFilenameToUriStringA
560 * @see uriUriStringToWindowsFilenameA
561 * @since 0.5.2
562 */
563int URI_FUNC(UriStringToUnixFilename)(const URI_CHAR * uriString,
564 URI_CHAR * filename);
565
566
567
568/**
569 * Extracts a Windows filename from a %URI string.
570 * The destination buffer must be large enough to hold len(uriString) + 1 - 8
571 * characters in case of an absolute %URI or len(uriString) + 1 in case
572 * of a relative %URI.
573 *
574 * @param uriString <b>IN</b>: %URI string to convert
575 * @param filename <b>OUT</b>: Destination to write filename to
576 * @return Error code or 0 on success
577 *
578 * @see uriWindowsFilenameToUriStringA
579 * @see uriUriStringToUnixFilenameA
580 * @since 0.5.2
581 */
582int URI_FUNC(UriStringToWindowsFilename)(const URI_CHAR * uriString,
583 URI_CHAR * filename);
584
585
586
587/**
588 * Calculates the number of characters needed to store the
589 * string representation of the given query list excluding the
590 * terminator. It is assumed that line breaks are will be
591 * normalized to "%0D%0A".
592 *
593 * @param queryList <b>IN</b>: Query list to measure
594 * @param charsRequired <b>OUT</b>: Length of the string representation in characters <b>excluding</b> terminator
595 * @return Error code or 0 on success
596 *
597 * @see uriComposeQueryCharsRequiredExA
598 * @see uriComposeQueryA
599 * @since 0.7.0
600 */
601int URI_FUNC(ComposeQueryCharsRequired)(const URI_TYPE(QueryList) * queryList,
602 int * charsRequired);
603
604
605
606/**
607 * Calculates the number of characters needed to store the
608 * string representation of the given query list excluding the
609 * terminator.
610 *
611 * @param queryList <b>IN</b>: Query list to measure
612 * @param charsRequired <b>OUT</b>: Length of the string representation in characters <b>excluding</b> terminator
613 * @param spaceToPlus <b>IN</b>: Wether to convert ' ' to '+' or not
614 * @param normalizeBreaks <b>IN</b>: Wether to convert CR and LF to CR-LF or not.
615 * @return Error code or 0 on success
616 *
617 * @see uriComposeQueryCharsRequiredA
618 * @see uriComposeQueryExA
619 * @since 0.7.0
620 */
621int URI_FUNC(ComposeQueryCharsRequiredEx)(const URI_TYPE(QueryList) * queryList,
622 int * charsRequired, UriBool spaceToPlus, UriBool normalizeBreaks);
623
624
625
626/**
627 * Converts a query list structure back to a query string.
628 * The composed string does not start with '?',
629 * on the way ' ' is converted to '+' and line breaks are
630 * normalized to "%0D%0A".
631 *
632 * @param dest <b>OUT</b>: Output destination
633 * @param queryList <b>IN</b>: Query list to convert
634 * @param maxChars <b>IN</b>: Maximum number of characters to copy <b>including</b> terminator
635 * @param charsWritten <b>OUT</b>: Number of characters written, can be lower than maxChars even if the query list is too long!
636 * @return Error code or 0 on success
637 *
638 * @see uriComposeQueryExA
639 * @see uriComposeQueryMallocA
640 * @see uriComposeQueryCharsRequiredA
641 * @see uriDissectQueryMallocA
642 * @since 0.7.0
643 */
644int URI_FUNC(ComposeQuery)(URI_CHAR * dest,
645 const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten);
646
647
648
649/**
650 * Converts a query list structure back to a query string.
651 * The composed string does not start with '?'.
652 *
653 * @param dest <b>OUT</b>: Output destination
654 * @param queryList <b>IN</b>: Query list to convert
655 * @param maxChars <b>IN</b>: Maximum number of characters to copy <b>including</b> terminator
656 * @param charsWritten <b>OUT</b>: Number of characters written, can be lower than maxChars even if the query list is too long!
657 * @param spaceToPlus <b>IN</b>: Wether to convert ' ' to '+' or not
658 * @param normalizeBreaks <b>IN</b>: Wether to convert CR and LF to CR-LF or not.
659 * @return Error code or 0 on success
660 *
661 * @see uriComposeQueryA
662 * @see uriComposeQueryMallocExA
663 * @see uriComposeQueryCharsRequiredExA
664 * @see uriDissectQueryMallocExA
665 * @since 0.7.0
666 */
667int URI_FUNC(ComposeQueryEx)(URI_CHAR * dest,
668 const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten,
669 UriBool spaceToPlus, UriBool normalizeBreaks);
670
671
672
673/**
674 * Converts a query list structure back to a query string.
675 * Memory for this string is allocated internally.
676 * The composed string does not start with '?',
677 * on the way ' ' is converted to '+' and line breaks are
678 * normalized to "%0D%0A".
679 *
680 * @param dest <b>OUT</b>: Output destination
681 * @param queryList <b>IN</b>: Query list to convert
682 * @return Error code or 0 on success
683 *
684 * @see uriComposeQueryMallocExA
685 * @see uriComposeQueryA
686 * @see uriDissectQueryMallocA
687 * @since 0.7.0
688 */
689int URI_FUNC(ComposeQueryMalloc)(URI_CHAR ** dest,
690 const URI_TYPE(QueryList) * queryList);
691
692
693
694/**
695 * Converts a query list structure back to a query string.
696 * Memory for this string is allocated internally.
697 * The composed string does not start with '?'.
698 *
699 * @param dest <b>OUT</b>: Output destination
700 * @param queryList <b>IN</b>: Query list to convert
701 * @param spaceToPlus <b>IN</b>: Wether to convert ' ' to '+' or not
702 * @param normalizeBreaks <b>IN</b>: Wether to convert CR and LF to CR-LF or not.
703 * @return Error code or 0 on success
704 *
705 * @see uriComposeQueryMallocA
706 * @see uriComposeQueryExA
707 * @see uriDissectQueryMallocExA
708 * @since 0.7.0
709 */
710int URI_FUNC(ComposeQueryMallocEx)(URI_CHAR ** dest,
711 const URI_TYPE(QueryList) * queryList,
712 UriBool spaceToPlus, UriBool normalizeBreaks);
713
714
715
716/**
717 * Constructs a query list from the raw query string of a given URI.
718 * On the way '+' is converted back to ' ', line breaks are not modified.
719 *
720 * @param dest <b>OUT</b>: Output destination
721 * @param itemCount <b>OUT</b>: Number of items found, can be NULL
722 * @param first <b>IN</b>: Pointer to first character <b>after</b> '?'
723 * @param afterLast <b>IN</b>: Pointer to character after the last one still in
724 * @return Error code or 0 on success
725 *
726 * @see uriDissectQueryMallocExA
727 * @see uriComposeQueryA
728 * @see uriFreeQueryListA
729 * @since 0.7.0
730 */
731int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) ** dest, int * itemCount,
732 const URI_CHAR * first, const URI_CHAR * afterLast);
733
734
735
736/**
737 * Constructs a query list from the raw query string of a given URI.
738 *
739 * @param dest <b>OUT</b>: Output destination
740 * @param itemCount <b>OUT</b>: Number of items found, can be NULL
741 * @param first <b>IN</b>: Pointer to first character <b>after</b> '?'
742 * @param afterLast <b>IN</b>: Pointer to character after the last one still in
743 * @param plusToSpace <b>IN</b>: Whether to convert '+' to ' ' or not
744 * @param breakConversion <b>IN</b>: Line break conversion mode
745 * @return Error code or 0 on success
746 *
747 * @see uriDissectQueryMallocA
748 * @see uriComposeQueryExA
749 * @see uriFreeQueryListA
750 * @since 0.7.0
751 */
752int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) ** dest, int * itemCount,
753 const URI_CHAR * first, const URI_CHAR * afterLast,
754 UriBool plusToSpace, UriBreakConversion breakConversion);
755
756
757
758/**
759 * Frees all memory associated with the given query list.
760 * The structure itself is freed as well.
761 *
762 * @param queryList <b>INOUT</b>: Query list to free
763 *
764 * @since 0.7.0
765 */
766void URI_FUNC(FreeQueryList)(URI_TYPE(QueryList) * queryList);
767
768
769
770#ifdef __cplusplus
771}
772#endif
773
774
775
776#endif
777#endif
diff --git a/plugins/uriparser/UriBase.h b/plugins/uriparser/UriBase.h
new file mode 100644
index 00000000..bc63b059
--- /dev/null
+++ b/plugins/uriparser/UriBase.h
@@ -0,0 +1,197 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriBase.h
42 * Holds definitions independent of the encoding pass.
43 */
44
45#ifndef URI_BASE_H
46#define URI_BASE_H 1
47
48
49
50/* Version helper macro */
51#define URI_ANSI_TO_UNICODE(x) L##x
52
53
54
55/* Version */
56#define URI_VER_MAJOR 0
57#define URI_VER_MINOR 8
58#define URI_VER_RELEASE 4
59#define URI_VER_SUFFIX_ANSI ""
60#define URI_VER_SUFFIX_UNICODE URI_ANSI_TO_UNICODE(URI_VER_SUFFIX_ANSI)
61
62
63
64/* More version helper macros */
65#define URI_INT_TO_ANSI_HELPER(x) #x
66#define URI_INT_TO_ANSI(x) URI_INT_TO_ANSI_HELPER(x)
67
68#define URI_INT_TO_UNICODE_HELPER(x) URI_ANSI_TO_UNICODE(#x)
69#define URI_INT_TO_UNICODE(x) URI_INT_TO_UNICODE_HELPER(x)
70
71#define URI_VER_ANSI_HELPER(ma, mi, r, s) \
72 URI_INT_TO_ANSI(ma) "." \
73 URI_INT_TO_ANSI(mi) "." \
74 URI_INT_TO_ANSI(r) \
75 s
76
77#define URI_VER_UNICODE_HELPER(ma, mi, r, s) \
78 URI_INT_TO_UNICODE(ma) L"." \
79 URI_INT_TO_UNICODE(mi) L"." \
80 URI_INT_TO_UNICODE(r) \
81 s
82
83
84
85/* Full version strings */
86#define URI_VER_ANSI URI_VER_ANSI_HELPER(URI_VER_MAJOR, URI_VER_MINOR, URI_VER_RELEASE, URI_VER_SUFFIX_ANSI)
87#define URI_VER_UNICODE URI_VER_UNICODE_HELPER(URI_VER_MAJOR, URI_VER_MINOR, URI_VER_RELEASE, URI_VER_SUFFIX_UNICODE)
88
89
90
91/* Unused parameter macro */
92#ifdef __GNUC__
93# define URI_UNUSED(x) unused_##x __attribute__((unused))
94#else
95# define URI_UNUSED(x) x
96#endif
97
98
99
100typedef int UriBool; /**< Boolean type */
101
102#define URI_TRUE 1
103#define URI_FALSE 0
104
105
106
107/* Shared errors */
108#define URI_SUCCESS 0
109#define URI_ERROR_SYNTAX 1 /* Parsed text violates expected format */
110#define URI_ERROR_NULL 2 /* One of the params passed was NULL
111 although it mustn't be */
112#define URI_ERROR_MALLOC 3 /* Requested memory could not be allocated */
113#define URI_ERROR_OUTPUT_TOO_LARGE 4 /* Some output is to large for the receiving buffer */
114#define URI_ERROR_NOT_IMPLEMENTED 8 /* The called function is not implemented yet */
115#define URI_ERROR_RANGE_INVALID 9 /* The parameters passed contained invalid ranges */
116
117
118/* Errors specific to ToString */
119#define URI_ERROR_TOSTRING_TOO_LONG URI_ERROR_OUTPUT_TOO_LARGE /* Deprecated, test for URI_ERROR_OUTPUT_TOO_LARGE instead */
120
121/* Errors specific to AddBaseUri */
122#define URI_ERROR_ADDBASE_REL_BASE 5 /* Given base is not absolute */
123
124/* Errors specific to RemoveBaseUri */
125#define URI_ERROR_REMOVEBASE_REL_BASE 6 /* Given base is not absolute */
126#define URI_ERROR_REMOVEBASE_REL_SOURCE 7 /* Given base is not absolute */
127
128
129
130#ifndef URI_DOXYGEN
131# include <stdio.h> /* For NULL, snprintf */
132# include <ctype.h> /* For wchar_t */
133# include <string.h> /* For strlen, memset, memcpy */
134# include <stdlib.h> /* For malloc */
135#endif /* URI_DOXYGEN */
136
137
138
139/**
140 * Holds an IPv4 address.
141 */
142typedef struct UriIp4Struct {
143 unsigned char data[4]; /**< Each octet in one byte */
144} UriIp4; /**< @copydoc UriIp4Struct */
145
146
147
148/**
149 * Holds an IPv6 address.
150 */
151typedef struct UriIp6Struct {
152 unsigned char data[16]; /**< Each quad in two bytes */
153} UriIp6; /**< @copydoc UriIp6Struct */
154
155
156
157/**
158 * Specifies a line break conversion mode.
159 */
160typedef enum UriBreakConversionEnum {
161 URI_BR_TO_LF, /**< Convert to Unix line breaks ("\\x0a") */
162 URI_BR_TO_CRLF, /**< Convert to Windows line breaks ("\\x0d\\x0a") */
163 URI_BR_TO_CR, /**< Convert to Macintosh line breaks ("\\x0d") */
164 URI_BR_TO_UNIX = URI_BR_TO_LF, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_LF */
165 URI_BR_TO_WINDOWS = URI_BR_TO_CRLF, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_CRLF */
166 URI_BR_TO_MAC = URI_BR_TO_CR, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_CR */
167 URI_BR_DONT_TOUCH /**< Copy line breaks unmodified */
168} UriBreakConversion; /**< @copydoc UriBreakConversionEnum */
169
170
171
172/**
173 * Specifies which component of a %URI has to be normalized.
174 */
175typedef enum UriNormalizationMaskEnum {
176 URI_NORMALIZED = 0, /**< Do not normalize anything */
177 URI_NORMALIZE_SCHEME = 1 << 0, /**< Normalize scheme (fix uppercase letters) */
178 URI_NORMALIZE_USER_INFO = 1 << 1, /**< Normalize user info (fix uppercase percent-encodings) */
179 URI_NORMALIZE_HOST = 1 << 2, /**< Normalize host (fix uppercase letters) */
180 URI_NORMALIZE_PATH = 1 << 3, /**< Normalize path (fix uppercase percent-encodings and redundant dot segments) */
181 URI_NORMALIZE_QUERY = 1 << 4, /**< Normalize query (fix uppercase percent-encodings) */
182 URI_NORMALIZE_FRAGMENT = 1 << 5 /**< Normalize fragment (fix uppercase percent-encodings) */
183} UriNormalizationMask; /**< @copydoc UriNormalizationMaskEnum */
184
185
186
187/**
188 * Specifies how to resolve %URI references.
189 */
190typedef enum UriResolutionOptionsEnum {
191 URI_RESOLVE_STRICTLY = 0, /**< Full RFC conformance */
192 URI_RESOLVE_IDENTICAL_SCHEME_COMPAT = 1 << 0 /**< Treat %URI to resolve with identical scheme as having no scheme */
193} UriResolutionOptions; /**< @copydoc UriResolutionOptionsEnum */
194
195
196
197#endif /* URI_BASE_H */
diff --git a/plugins/uriparser/UriCommon.c b/plugins/uriparser/UriCommon.c
new file mode 100644
index 00000000..37d6b39d
--- /dev/null
+++ b/plugins/uriparser/UriCommon.c
@@ -0,0 +1,567 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriCommon.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriCommon.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include "UriCommon.h"
67#endif
68
69
70
71/*extern*/ const URI_CHAR * const URI_FUNC(SafeToPointTo) = _UT("X");
72/*extern*/ const URI_CHAR * const URI_FUNC(ConstPwd) = _UT(".");
73/*extern*/ const URI_CHAR * const URI_FUNC(ConstParent) = _UT("..");
74
75
76
77void URI_FUNC(ResetUri)(URI_TYPE(Uri) * uri) {
78 memset(uri, 0, sizeof(URI_TYPE(Uri)));
79}
80
81
82
83/* Compares two text ranges for equal text content */
84int URI_FUNC(CompareRange)(
85 const URI_TYPE(TextRange) * a,
86 const URI_TYPE(TextRange) * b) {
87 int diff;
88
89 /* NOTE: Both NULL means equal! */
90 if ((a == NULL) || (b == NULL)) {
91 return ((a == NULL) ? 0 : 1) - ((b == NULL) ? 0 : 1);
92 }
93
94 /* NOTE: Both NULL means equal! */
95 if ((a->first == NULL) || (b->first == NULL)) {
96 return ((a->first == NULL) ? 0 : 1) - ((b->first == NULL) ? 0 : 1);
97 }
98
99 diff = ((int)(a->afterLast - a->first) - (int)(b->afterLast - b->first));
100 if (diff > 0) {
101 return 1;
102 } else if (diff < 0) {
103 return -1;
104 }
105
106 diff = URI_STRNCMP(a->first, b->first, (a->afterLast - a->first));
107
108 if (diff > 0) {
109 return 1;
110 } else if (diff < 0) {
111 return -1;
112 }
113
114 return diff;
115}
116
117
118
119/* Properly removes "." and ".." path segments */
120UriBool URI_FUNC(RemoveDotSegments)(URI_TYPE(Uri) * uri,
121 UriBool relative) {
122 if (uri == NULL) {
123 return URI_TRUE;
124 }
125 return URI_FUNC(RemoveDotSegmentsEx)(uri, relative, uri->owner);
126}
127
128
129
130UriBool URI_FUNC(RemoveDotSegmentsEx)(URI_TYPE(Uri) * uri,
131 UriBool relative, UriBool pathOwned) {
132 URI_TYPE(PathSegment) * walker;
133 if ((uri == NULL) || (uri->pathHead == NULL)) {
134 return URI_TRUE;
135 }
136
137 walker = uri->pathHead;
138 walker->reserved = NULL; /* Prev pointer */
139 do {
140 UriBool removeSegment = URI_FALSE;
141 int len = (int)(walker->text.afterLast - walker->text.first);
142 switch (len) {
143 case 1:
144 if ((walker->text.first)[0] == _UT('.')) {
145 /* "." segment -> remove if not essential */
146 URI_TYPE(PathSegment) * const prev = walker->reserved;
147 URI_TYPE(PathSegment) * const nextBackup = walker->next;
148
149 /* Is this dot segment essential? */
150 removeSegment = URI_TRUE;
151 if (relative && (walker == uri->pathHead) && (walker->next != NULL)) {
152 const URI_CHAR * ch = walker->next->text.first;
153 for (; ch < walker->next->text.afterLast; ch++) {
154 if (*ch == _UT(':')) {
155 removeSegment = URI_FALSE;
156 break;
157 }
158 }
159 }
160
161 if (removeSegment) {
162 /* Last segment? */
163 if (walker->next != NULL) {
164 /* Not last segment */
165 walker->next->reserved = prev;
166
167 if (prev == NULL) {
168 /* First but not last segment */
169 uri->pathHead = walker->next;
170 } else {
171 /* Middle segment */
172 prev->next = walker->next;
173 }
174
175 if (pathOwned && (walker->text.first != walker->text.afterLast)) {
176 free((URI_CHAR *)walker->text.first);
177 }
178 free(walker);
179 } else {
180 /* Last segment */
181 if (pathOwned && (walker->text.first != walker->text.afterLast)) {
182 free((URI_CHAR *)walker->text.first);
183 }
184
185 if (prev == NULL) {
186 /* Last and first */
187 if (URI_FUNC(IsHostSet)(uri)) {
188 /* Replace "." with empty segment to represent trailing slash */
189 walker->text.first = URI_FUNC(SafeToPointTo);
190 walker->text.afterLast = URI_FUNC(SafeToPointTo);
191 } else {
192 free(walker);
193
194 uri->pathHead = NULL;
195 uri->pathTail = NULL;
196 }
197 } else {
198 /* Last but not first, replace "." with empty segment to represent trailing slash */
199 walker->text.first = URI_FUNC(SafeToPointTo);
200 walker->text.afterLast = URI_FUNC(SafeToPointTo);
201 }
202 }
203
204 walker = nextBackup;
205 }
206 }
207 break;
208
209 case 2:
210 if (((walker->text.first)[0] == _UT('.'))
211 && ((walker->text.first)[1] == _UT('.'))) {
212 /* Path ".." -> remove this and the previous segment */
213 URI_TYPE(PathSegment) * const prev = walker->reserved;
214 URI_TYPE(PathSegment) * prevPrev;
215 URI_TYPE(PathSegment) * const nextBackup = walker->next;
216
217 removeSegment = URI_TRUE;
218 if (relative) {
219 if (prev == NULL) {
220 removeSegment = URI_FALSE;
221 } else if ((prev != NULL)
222 && ((prev->text.afterLast - prev->text.first) == 2)
223 && ((prev->text.first)[0] == _UT('.'))
224 && ((prev->text.first)[1] == _UT('.'))) {
225 removeSegment = URI_FALSE;
226 }
227 }
228
229 if (removeSegment) {
230 if (prev != NULL) {
231 /* Not first segment */
232 prevPrev = prev->reserved;
233 if (prevPrev != NULL) {
234 /* Not even prev is the first one */
235 prevPrev->next = walker->next;
236 if (walker->next != NULL) {
237 walker->next->reserved = prevPrev;
238 } else {
239 /* Last segment -> insert "" segment to represent trailing slash, update tail */
240 URI_TYPE(PathSegment) * const segment = malloc(1 * sizeof(URI_TYPE(PathSegment)));
241 if (segment == NULL) {
242 if (pathOwned && (walker->text.first != walker->text.afterLast)) {
243 free((URI_CHAR *)walker->text.first);
244 }
245 free(walker);
246
247 if (pathOwned && (prev->text.first != prev->text.afterLast)) {
248 free((URI_CHAR *)prev->text.first);
249 }
250 free(prev);
251
252 return URI_FALSE; /* Raises malloc error */
253 }
254 memset(segment, 0, sizeof(URI_TYPE(PathSegment)));
255 segment->text.first = URI_FUNC(SafeToPointTo);
256 segment->text.afterLast = URI_FUNC(SafeToPointTo);
257 prevPrev->next = segment;
258 uri->pathTail = segment;
259 }
260
261 if (pathOwned && (walker->text.first != walker->text.afterLast)) {
262 free((URI_CHAR *)walker->text.first);
263 }
264 free(walker);
265
266 if (pathOwned && (prev->text.first != prev->text.afterLast)) {
267 free((URI_CHAR *)prev->text.first);
268 }
269 free(prev);
270
271 walker = nextBackup;
272 } else {
273 /* Prev is the first segment */
274 if (walker->next != NULL) {
275 uri->pathHead = walker->next;
276 walker->next->reserved = NULL;
277
278 if (pathOwned && (walker->text.first != walker->text.afterLast)) {
279 free((URI_CHAR *)walker->text.first);
280 }
281 free(walker);
282 } else {
283 /* Re-use segment for "" path segment to represent trailing slash, update tail */
284 URI_TYPE(PathSegment) * const segment = walker;
285 if (pathOwned && (segment->text.first != segment->text.afterLast)) {
286 free((URI_CHAR *)segment->text.first);
287 }
288 segment->text.first = URI_FUNC(SafeToPointTo);
289 segment->text.afterLast = URI_FUNC(SafeToPointTo);
290 uri->pathHead = segment;
291 uri->pathTail = segment;
292 }
293
294 if (pathOwned && (prev->text.first != prev->text.afterLast)) {
295 free((URI_CHAR *)prev->text.first);
296 }
297 free(prev);
298
299 walker = nextBackup;
300 }
301 } else {
302 URI_TYPE(PathSegment) * const anotherNextBackup = walker->next;
303 /* First segment -> update head pointer */
304 uri->pathHead = walker->next;
305 if (walker->next != NULL) {
306 walker->next->reserved = NULL;
307 } else {
308 /* Last segment -> update tail */
309 uri->pathTail = NULL;
310 }
311
312 if (pathOwned && (walker->text.first != walker->text.afterLast)) {
313 free((URI_CHAR *)walker->text.first);
314 }
315 free(walker);
316
317 walker = anotherNextBackup;
318 }
319 }
320 }
321 break;
322
323 }
324
325 if (!removeSegment) {
326 if (walker->next != NULL) {
327 walker->next->reserved = walker;
328 } else {
329 /* Last segment -> update tail */
330 uri->pathTail = walker;
331 }
332 walker = walker->next;
333 }
334 } while (walker != NULL);
335
336 return URI_TRUE;
337}
338
339
340
341/* Properly removes "." and ".." path segments */
342UriBool URI_FUNC(RemoveDotSegmentsAbsolute)(URI_TYPE(Uri) * uri) {
343 const UriBool ABSOLUTE = URI_FALSE;
344 return URI_FUNC(RemoveDotSegments)(uri, ABSOLUTE);
345}
346
347
348
349unsigned char URI_FUNC(HexdigToInt)(URI_CHAR hexdig) {
350 switch (hexdig) {
351 case _UT('0'):
352 case _UT('1'):
353 case _UT('2'):
354 case _UT('3'):
355 case _UT('4'):
356 case _UT('5'):
357 case _UT('6'):
358 case _UT('7'):
359 case _UT('8'):
360 case _UT('9'):
361 return (unsigned char)(9 + hexdig - _UT('9'));
362
363 case _UT('a'):
364 case _UT('b'):
365 case _UT('c'):
366 case _UT('d'):
367 case _UT('e'):
368 case _UT('f'):
369 return (unsigned char)(15 + hexdig - _UT('f'));
370
371 case _UT('A'):
372 case _UT('B'):
373 case _UT('C'):
374 case _UT('D'):
375 case _UT('E'):
376 case _UT('F'):
377 return (unsigned char)(15 + hexdig - _UT('F'));
378
379 default:
380 return 0;
381 }
382}
383
384
385
386URI_CHAR URI_FUNC(HexToLetter)(unsigned int value) {
387 /* Uppercase recommended in section 2.1. of RFC 3986 *
388 * http://tools.ietf.org/html/rfc3986#section-2.1 */
389 return URI_FUNC(HexToLetterEx)(value, URI_TRUE);
390}
391
392
393
394URI_CHAR URI_FUNC(HexToLetterEx)(unsigned int value, UriBool uppercase) {
395 switch (value) {
396 case 0: return _UT('0');
397 case 1: return _UT('1');
398 case 2: return _UT('2');
399 case 3: return _UT('3');
400 case 4: return _UT('4');
401 case 5: return _UT('5');
402 case 6: return _UT('6');
403 case 7: return _UT('7');
404 case 8: return _UT('8');
405 case 9: return _UT('9');
406
407 case 10: return (uppercase == URI_TRUE) ? _UT('A') : _UT('a');
408 case 11: return (uppercase == URI_TRUE) ? _UT('B') : _UT('b');
409 case 12: return (uppercase == URI_TRUE) ? _UT('C') : _UT('c');
410 case 13: return (uppercase == URI_TRUE) ? _UT('D') : _UT('d');
411 case 14: return (uppercase == URI_TRUE) ? _UT('E') : _UT('e');
412 default: return (uppercase == URI_TRUE) ? _UT('F') : _UT('f');
413 }
414}
415
416
417
418/* Checks if a URI has the host component set. */
419UriBool URI_FUNC(IsHostSet)(const URI_TYPE(Uri) * uri) {
420 return (uri != NULL)
421 && ((uri->hostText.first != NULL)
422 || (uri->hostData.ip4 != NULL)
423 || (uri->hostData.ip6 != NULL)
424 || (uri->hostData.ipFuture.first != NULL)
425 );
426}
427
428
429
430/* Copies the path segment list from one URI to another. */
431UriBool URI_FUNC(CopyPath)(URI_TYPE(Uri) * dest,
432 const URI_TYPE(Uri) * source) {
433 if (source->pathHead == NULL) {
434 /* No path component */
435 dest->pathHead = NULL;
436 dest->pathTail = NULL;
437 } else {
438 /* Copy list but not the text contained */
439 URI_TYPE(PathSegment) * sourceWalker = source->pathHead;
440 URI_TYPE(PathSegment) * destPrev = NULL;
441 do {
442 URI_TYPE(PathSegment) * cur = malloc(sizeof(URI_TYPE(PathSegment)));
443 if (cur == NULL) {
444 /* Fix broken list */
445 if (destPrev != NULL) {
446 destPrev->next = NULL;
447 }
448 return URI_FALSE; /* Raises malloc error */
449 }
450
451 /* From this functions usage we know that *
452 * the dest URI cannot be uri->owner */
453 cur->text = sourceWalker->text;
454 if (destPrev == NULL) {
455 /* First segment ever */
456 dest->pathHead = cur;
457 } else {
458 destPrev->next = cur;
459 }
460 destPrev = cur;
461 sourceWalker = sourceWalker->next;
462 } while (sourceWalker != NULL);
463 dest->pathTail = destPrev;
464 dest->pathTail->next = NULL;
465 }
466
467 dest->absolutePath = source->absolutePath;
468 return URI_TRUE;
469}
470
471
472
473/* Copies the authority part of an URI over to another. */
474UriBool URI_FUNC(CopyAuthority)(URI_TYPE(Uri) * dest,
475 const URI_TYPE(Uri) * source) {
476 /* From this functions usage we know that *
477 * the dest URI cannot be uri->owner */
478
479 /* Copy userInfo */
480 dest->userInfo = source->userInfo;
481
482 /* Copy hostText */
483 dest->hostText = source->hostText;
484
485 /* Copy hostData */
486 if (source->hostData.ip4 != NULL) {
487 dest->hostData.ip4 = malloc(sizeof(UriIp4));
488 if (dest->hostData.ip4 == NULL) {
489 return URI_FALSE; /* Raises malloc error */
490 }
491 *(dest->hostData.ip4) = *(source->hostData.ip4);
492 dest->hostData.ip6 = NULL;
493 dest->hostData.ipFuture.first = NULL;
494 dest->hostData.ipFuture.afterLast = NULL;
495 } else if (source->hostData.ip6 != NULL) {
496 dest->hostData.ip4 = NULL;
497 dest->hostData.ip6 = malloc(sizeof(UriIp6));
498 if (dest->hostData.ip6 == NULL) {
499 return URI_FALSE; /* Raises malloc error */
500 }
501 *(dest->hostData.ip6) = *(source->hostData.ip6);
502 dest->hostData.ipFuture.first = NULL;
503 dest->hostData.ipFuture.afterLast = NULL;
504 } else {
505 dest->hostData.ip4 = NULL;
506 dest->hostData.ip6 = NULL;
507 dest->hostData.ipFuture = source->hostData.ipFuture;
508 }
509
510 /* Copy portText */
511 dest->portText = source->portText;
512
513 return URI_TRUE;
514}
515
516
517
518UriBool URI_FUNC(FixAmbiguity)(URI_TYPE(Uri) * uri) {
519 URI_TYPE(PathSegment) * segment;
520
521 if ( /* Case 1: absolute path, empty first segment */
522 (uri->absolutePath
523 && (uri->pathHead != NULL)
524 && (uri->pathHead->text.afterLast == uri->pathHead->text.first))
525
526 /* Case 2: relative path, empty first and second segment */
527 || (!uri->absolutePath
528 && (uri->pathHead != NULL)
529 && (uri->pathHead->next != NULL)
530 && (uri->pathHead->text.afterLast == uri->pathHead->text.first)
531 && (uri->pathHead->next->text.afterLast == uri->pathHead->next->text.first))) {
532 /* NOOP */
533 } else {
534 return URI_TRUE;
535 }
536
537 segment = malloc(1 * sizeof(URI_TYPE(PathSegment)));
538 if (segment == NULL) {
539 return URI_FALSE; /* Raises malloc error */
540 }
541
542 /* Insert "." segment in front */
543 segment->next = uri->pathHead;
544 segment->text.first = URI_FUNC(ConstPwd);
545 segment->text.afterLast = URI_FUNC(ConstPwd) + 1;
546 uri->pathHead = segment;
547 return URI_TRUE;
548}
549
550
551
552void URI_FUNC(FixEmptyTrailSegment)(URI_TYPE(Uri) * uri) {
553 /* Fix path if only one empty segment */
554 if (!uri->absolutePath
555 && !URI_FUNC(IsHostSet)(uri)
556 && (uri->pathHead != NULL)
557 && (uri->pathHead->next == NULL)
558 && (uri->pathHead->text.first == uri->pathHead->text.afterLast)) {
559 free(uri->pathHead);
560 uri->pathHead = NULL;
561 uri->pathTail = NULL;
562 }
563}
564
565
566
567#endif
diff --git a/plugins/uriparser/UriCommon.h b/plugins/uriparser/UriCommon.h
new file mode 100644
index 00000000..f6bc2ccf
--- /dev/null
+++ b/plugins/uriparser/UriCommon.h
@@ -0,0 +1,104 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40#if (defined(URI_PASS_ANSI) && !defined(URI_COMMON_H_ANSI)) \
41 || (defined(URI_PASS_UNICODE) && !defined(URI_COMMON_H_UNICODE)) \
42 || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* What encodings are enabled? */
44#include <uriparser/UriDefsConfig.h>
45#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
46/* Include SELF twice */
47# ifdef URI_ENABLE_ANSI
48# define URI_PASS_ANSI 1
49# include "UriCommon.h"
50# undef URI_PASS_ANSI
51# endif
52# ifdef URI_ENABLE_UNICODE
53# define URI_PASS_UNICODE 1
54# include "UriCommon.h"
55# undef URI_PASS_UNICODE
56# endif
57/* Only one pass for each encoding */
58#elif (defined(URI_PASS_ANSI) && !defined(URI_COMMON_H_ANSI) \
59 && defined(URI_ENABLE_ANSI)) || (defined(URI_PASS_UNICODE) \
60 && !defined(URI_COMMON_H_UNICODE) && defined(URI_ENABLE_UNICODE))
61# ifdef URI_PASS_ANSI
62# define URI_COMMON_H_ANSI 1
63# include <uriparser/UriDefsAnsi.h>
64# else
65# define URI_COMMON_H_UNICODE 1
66# include <uriparser/UriDefsUnicode.h>
67# endif
68
69
70
71/* Used to point to from empty path segments.
72 * X.first and X.afterLast must be the same non-NULL value then. */
73extern const URI_CHAR * const URI_FUNC(SafeToPointTo);
74extern const URI_CHAR * const URI_FUNC(ConstPwd);
75extern const URI_CHAR * const URI_FUNC(ConstParent);
76
77
78
79void URI_FUNC(ResetUri)(URI_TYPE(Uri) * uri);
80
81int URI_FUNC(CompareRange)(
82 const URI_TYPE(TextRange) * a,
83 const URI_TYPE(TextRange) * b);
84
85UriBool URI_FUNC(RemoveDotSegmentsAbsolute)(URI_TYPE(Uri) * uri);
86UriBool URI_FUNC(RemoveDotSegments)(URI_TYPE(Uri) * uri, UriBool relative);
87UriBool URI_FUNC(RemoveDotSegmentsEx)(URI_TYPE(Uri) * uri,
88 UriBool relative, UriBool pathOwned);
89
90unsigned char URI_FUNC(HexdigToInt)(URI_CHAR hexdig);
91URI_CHAR URI_FUNC(HexToLetter)(unsigned int value);
92URI_CHAR URI_FUNC(HexToLetterEx)(unsigned int value, UriBool uppercase);
93
94UriBool URI_FUNC(IsHostSet)(const URI_TYPE(Uri) * uri);
95
96UriBool URI_FUNC(CopyPath)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * source);
97UriBool URI_FUNC(CopyAuthority)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * source);
98
99UriBool URI_FUNC(FixAmbiguity)(URI_TYPE(Uri) * uri);
100void URI_FUNC(FixEmptyTrailSegment)(URI_TYPE(Uri) * uri);
101
102
103#endif
104#endif
diff --git a/plugins/uriparser/UriCompare.c b/plugins/uriparser/UriCompare.c
new file mode 100644
index 00000000..6896f64e
--- /dev/null
+++ b/plugins/uriparser/UriCompare.c
@@ -0,0 +1,168 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriCompare.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriCompare.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include <uriparser/UriIp4.h>
67# include "UriCommon.h"
68#endif
69
70
71
72UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a,
73 const URI_TYPE(Uri) * b) {
74 /* NOTE: Both NULL means equal! */
75 if ((a == NULL) || (b == NULL)) {
76 return ((a == NULL) && (b == NULL)) ? URI_TRUE : URI_FALSE;
77 }
78
79 /* scheme */
80 if (URI_FUNC(CompareRange)(&(a->scheme), &(b->scheme))) {
81 return URI_FALSE;
82 }
83
84 /* absolutePath */
85 if ((a->scheme.first == NULL)&& (a->absolutePath != b->absolutePath)) {
86 return URI_FALSE;
87 }
88
89 /* userInfo */
90 if (URI_FUNC(CompareRange)(&(a->userInfo), &(b->userInfo))) {
91 return URI_FALSE;
92 }
93
94 /* Host */
95 if (((a->hostData.ip4 == NULL) != (b->hostData.ip4 == NULL))
96 || ((a->hostData.ip6 == NULL) != (b->hostData.ip6 == NULL))
97 || ((a->hostData.ipFuture.first == NULL)
98 != (b->hostData.ipFuture.first == NULL))) {
99 return URI_FALSE;
100 }
101
102 if (a->hostData.ip4 != NULL) {
103 if (memcmp(a->hostData.ip4->data, b->hostData.ip4->data, 4)) {
104 return URI_FALSE;
105 }
106 }
107
108 if (a->hostData.ip6 != NULL) {
109 if (memcmp(a->hostData.ip6->data, b->hostData.ip6->data, 16)) {
110 return URI_FALSE;
111 }
112 }
113
114 if (a->hostData.ipFuture.first != NULL) {
115 if (URI_FUNC(CompareRange)(&(a->hostData.ipFuture), &(b->hostData.ipFuture))) {
116 return URI_FALSE;
117 }
118 }
119
120 if ((a->hostData.ip4 == NULL)
121 && (a->hostData.ip6 == NULL)
122 && (a->hostData.ipFuture.first == NULL)) {
123 if (URI_FUNC(CompareRange)(&(a->hostText), &(b->hostText))) {
124 return URI_FALSE;
125 }
126 }
127
128 /* portText */
129 if (URI_FUNC(CompareRange)(&(a->portText), &(b->portText))) {
130 return URI_FALSE;
131 }
132
133 /* Path */
134 if ((a->pathHead == NULL) != (b->pathHead == NULL)) {
135 return URI_FALSE;
136 }
137
138 if (a->pathHead != NULL) {
139 URI_TYPE(PathSegment) * walkA = a->pathHead;
140 URI_TYPE(PathSegment) * walkB = b->pathHead;
141 do {
142 if (URI_FUNC(CompareRange)(&(walkA->text), &(walkB->text))) {
143 return URI_FALSE;
144 }
145 if ((walkA->next == NULL) != (walkB->next == NULL)) {
146 return URI_FALSE;
147 }
148 walkA = walkA->next;
149 walkB = walkB->next;
150 } while (walkA != NULL);
151 }
152
153 /* query */
154 if (URI_FUNC(CompareRange)(&(a->query), &(b->query))) {
155 return URI_FALSE;
156 }
157
158 /* fragment */
159 if (URI_FUNC(CompareRange)(&(a->fragment), &(b->fragment))) {
160 return URI_FALSE;
161 }
162
163 return URI_TRUE; /* Equal*/
164}
165
166
167
168#endif
diff --git a/plugins/uriparser/UriDefsAnsi.h b/plugins/uriparser/UriDefsAnsi.h
new file mode 100644
index 00000000..deaefaa5
--- /dev/null
+++ b/plugins/uriparser/UriDefsAnsi.h
@@ -0,0 +1,82 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriDefsAnsi.h
42 * Holds definitions for the ANSI pass.
43 * NOTE: This header is included N times, not once.
44 */
45
46/* Allow multi inclusion */
47#include "UriDefsConfig.h"
48
49
50
51#undef URI_CHAR
52#define URI_CHAR char
53
54#undef _UT
55#define _UT(x) x
56
57
58
59#undef URI_FUNC
60#define URI_FUNC(x) uri##x##A
61
62#undef URI_TYPE
63#define URI_TYPE(x) Uri##x##A
64
65
66
67#undef URI_STRLEN
68#define URI_STRLEN strlen
69#undef URI_STRCPY
70#define URI_STRCPY strcpy
71#undef URI_STRCMP
72#define URI_STRCMP strcmp
73#undef URI_STRNCMP
74#define URI_STRNCMP strncmp
75
76/* TODO Remove on next source-compatibility break */
77#undef URI_SNPRINTF
78#if (defined(__WIN32__) || defined(_WIN32) || defined(WIN32))
79# define URI_SNPRINTF _snprintf
80#else
81# define URI_SNPRINTF snprintf
82#endif
diff --git a/plugins/uriparser/UriDefsConfig.h b/plugins/uriparser/UriDefsConfig.h
new file mode 100644
index 00000000..d87ccb63
--- /dev/null
+++ b/plugins/uriparser/UriDefsConfig.h
@@ -0,0 +1,101 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriDefsConfig.h
42 * Adjusts the internal configuration after processing external definitions.
43 */
44
45#ifndef URI_DEFS_CONFIG_H
46#define URI_DEFS_CONFIG_H 1
47
48
49
50/* Deny external overriding */
51#undef URI_ENABLE_ANSI /* Internal for !URI_NO_ANSI */
52#undef URI_ENABLE_UNICODE /* Internal for !URI_NO_UNICODE */
53
54
55
56/* Encoding */
57#ifdef URI_NO_ANSI
58# ifdef URI_NO_UNICODE
59/* No encoding at all */
60# error URI_NO_ANSI and URI_NO_UNICODE cannot go together.
61# else
62/* Unicode only */
63# define URI_ENABLE_UNICODE 1
64# endif
65#else
66# ifdef URI_NO_UNICODE
67/* ANSI only */
68# define URI_ENABLE_ANSI 1
69# else
70/* Both ANSI and Unicode */
71# define URI_ENABLE_ANSI 1
72# define URI_ENABLE_UNICODE 1
73# endif
74#endif
75
76
77
78/* Function inlining, not ANSI/ISO C! */
79#if (defined(URI_DOXYGEN) || defined(URI_SIZEDOWN))
80# define URI_INLINE
81#elif defined(__INTEL_COMPILER)
82/* Intel C/C++ */
83/* http://predef.sourceforge.net/precomp.html#sec20 */
84/* http://www.intel.com/support/performancetools/c/windows/sb/CS-007751.htm#2 */
85# define URI_INLINE __force_inline
86#elif defined(_MSC_VER)
87/* Microsoft Visual C++ */
88/* http://predef.sourceforge.net/precomp.html#sec32 */
89/* http://msdn2.microsoft.com/en-us/library/ms882281.aspx */
90# define URI_INLINE __forceinline
91#elif (__STDC_VERSION__ >= 199901L)
92/* C99, "inline" is a keyword */
93# define URI_INLINE inline
94#else
95/* No inlining */
96# define URI_INLINE
97#endif
98
99
100
101#endif /* URI_DEFS_CONFIG_H */
diff --git a/plugins/uriparser/UriDefsUnicode.h b/plugins/uriparser/UriDefsUnicode.h
new file mode 100644
index 00000000..fa4befcd
--- /dev/null
+++ b/plugins/uriparser/UriDefsUnicode.h
@@ -0,0 +1,82 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriDefsUnicode.h
42 * Holds definitions for the Unicode pass.
43 * NOTE: This header is included N times, not once.
44 */
45
46/* Allow multi inclusion */
47#include "UriDefsConfig.h"
48
49
50
51#undef URI_CHAR
52#define URI_CHAR wchar_t
53
54#undef _UT
55#define _UT(x) L##x
56
57
58
59#undef URI_FUNC
60#define URI_FUNC(x) uri##x##W
61
62#undef URI_TYPE
63#define URI_TYPE(x) Uri##x##W
64
65
66
67#undef URI_STRLEN
68#define URI_STRLEN wcslen
69#undef URI_STRCPY
70#define URI_STRCPY wcscpy
71#undef URI_STRCMP
72#define URI_STRCMP wcscmp
73#undef URI_STRNCMP
74#define URI_STRNCMP wcsncmp
75
76/* TODO Remove on next source-compatibility break */
77#undef URI_SNPRINTF
78#if (defined(__WIN32__) || defined(_WIN32) || defined(WIN32))
79# define URI_SNPRINTF _snwprintf
80#else
81# define URI_SNPRINTF swprintf
82#endif
diff --git a/plugins/uriparser/UriEscape.c b/plugins/uriparser/UriEscape.c
new file mode 100644
index 00000000..79ee3a68
--- /dev/null
+++ b/plugins/uriparser/UriEscape.c
@@ -0,0 +1,453 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriEscape.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriEscape.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include "UriCommon.h"
67#endif
68
69
70
71URI_CHAR * URI_FUNC(Escape)(const URI_CHAR * in, URI_CHAR * out,
72 UriBool spaceToPlus, UriBool normalizeBreaks) {
73 return URI_FUNC(EscapeEx)(in, NULL, out, spaceToPlus, normalizeBreaks);
74}
75
76
77
78URI_CHAR * URI_FUNC(EscapeEx)(const URI_CHAR * inFirst,
79 const URI_CHAR * inAfterLast, URI_CHAR * out,
80 UriBool spaceToPlus, UriBool normalizeBreaks) {
81 const URI_CHAR * read = inFirst;
82 URI_CHAR * write = out;
83 UriBool prevWasCr = URI_FALSE;
84 if ((out == NULL) || (inFirst == out)) {
85 return NULL;
86 } else if (inFirst == NULL) {
87 if (out != NULL) {
88 out[0] = _UT('\0');
89 }
90 return out;
91 }
92
93 for (;;) {
94 if ((inAfterLast != NULL) && (read >= inAfterLast)) {
95 write[0] = _UT('\0');
96 return write;
97 }
98
99 switch (read[0]) {
100 case _UT('\0'):
101 write[0] = _UT('\0');
102 return write;
103
104 case _UT(' '):
105 if (spaceToPlus) {
106 write[0] = _UT('+');
107 write++;
108 } else {
109 write[0] = _UT('%');
110 write[1] = _UT('2');
111 write[2] = _UT('0');
112 write += 3;
113 }
114 prevWasCr = URI_FALSE;
115 break;
116
117 case _UT('a'): /* ALPHA */
118 case _UT('A'):
119 case _UT('b'):
120 case _UT('B'):
121 case _UT('c'):
122 case _UT('C'):
123 case _UT('d'):
124 case _UT('D'):
125 case _UT('e'):
126 case _UT('E'):
127 case _UT('f'):
128 case _UT('F'):
129 case _UT('g'):
130 case _UT('G'):
131 case _UT('h'):
132 case _UT('H'):
133 case _UT('i'):
134 case _UT('I'):
135 case _UT('j'):
136 case _UT('J'):
137 case _UT('k'):
138 case _UT('K'):
139 case _UT('l'):
140 case _UT('L'):
141 case _UT('m'):
142 case _UT('M'):
143 case _UT('n'):
144 case _UT('N'):
145 case _UT('o'):
146 case _UT('O'):
147 case _UT('p'):
148 case _UT('P'):
149 case _UT('q'):
150 case _UT('Q'):
151 case _UT('r'):
152 case _UT('R'):
153 case _UT('s'):
154 case _UT('S'):
155 case _UT('t'):
156 case _UT('T'):
157 case _UT('u'):
158 case _UT('U'):
159 case _UT('v'):
160 case _UT('V'):
161 case _UT('w'):
162 case _UT('W'):
163 case _UT('x'):
164 case _UT('X'):
165 case _UT('y'):
166 case _UT('Y'):
167 case _UT('z'):
168 case _UT('Z'):
169 case _UT('0'): /* DIGIT */
170 case _UT('1'):
171 case _UT('2'):
172 case _UT('3'):
173 case _UT('4'):
174 case _UT('5'):
175 case _UT('6'):
176 case _UT('7'):
177 case _UT('8'):
178 case _UT('9'):
179 case _UT('-'): /* "-" / "." / "_" / "~" */
180 case _UT('.'):
181 case _UT('_'):
182 case _UT('~'):
183 /* Copy unmodified */
184 write[0] = read[0];
185 write++;
186
187 prevWasCr = URI_FALSE;
188 break;
189
190 case _UT('\x0a'):
191 if (normalizeBreaks) {
192 if (!prevWasCr) {
193 write[0] = _UT('%');
194 write[1] = _UT('0');
195 write[2] = _UT('D');
196 write[3] = _UT('%');
197 write[4] = _UT('0');
198 write[5] = _UT('A');
199 write += 6;
200 }
201 } else {
202 write[0] = _UT('%');
203 write[1] = _UT('0');
204 write[2] = _UT('A');
205 write += 3;
206 }
207 prevWasCr = URI_FALSE;
208 break;
209
210 case _UT('\x0d'):
211 if (normalizeBreaks) {
212 write[0] = _UT('%');
213 write[1] = _UT('0');
214 write[2] = _UT('D');
215 write[3] = _UT('%');
216 write[4] = _UT('0');
217 write[5] = _UT('A');
218 write += 6;
219 } else {
220 write[0] = _UT('%');
221 write[1] = _UT('0');
222 write[2] = _UT('D');
223 write += 3;
224 }
225 prevWasCr = URI_TRUE;
226 break;
227
228 default:
229 /* Percent encode */
230 {
231 const unsigned char code = (unsigned char)read[0];
232 write[0] = _UT('%');
233 write[1] = URI_FUNC(HexToLetter)(code >> 4);
234 write[2] = URI_FUNC(HexToLetter)(code & 0x0f);
235 write += 3;
236 }
237 prevWasCr = URI_FALSE;
238 break;
239 }
240
241 read++;
242 }
243}
244
245
246
247const URI_CHAR * URI_FUNC(UnescapeInPlace)(URI_CHAR * inout) {
248 return URI_FUNC(UnescapeInPlaceEx)(inout, URI_FALSE, URI_BR_DONT_TOUCH);
249}
250
251
252
253const URI_CHAR * URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout,
254 UriBool plusToSpace, UriBreakConversion breakConversion) {
255 URI_CHAR * read = inout;
256 URI_CHAR * write = inout;
257 UriBool prevWasCr = URI_FALSE;
258
259 if (inout == NULL) {
260 return NULL;
261 }
262
263 for (;;) {
264 switch (read[0]) {
265 case _UT('\0'):
266 if (read > write) {
267 write[0] = _UT('\0');
268 }
269 return write;
270
271 case _UT('%'):
272 switch (read[1]) {
273 case _UT('0'):
274 case _UT('1'):
275 case _UT('2'):
276 case _UT('3'):
277 case _UT('4'):
278 case _UT('5'):
279 case _UT('6'):
280 case _UT('7'):
281 case _UT('8'):
282 case _UT('9'):
283 case _UT('a'):
284 case _UT('b'):
285 case _UT('c'):
286 case _UT('d'):
287 case _UT('e'):
288 case _UT('f'):
289 case _UT('A'):
290 case _UT('B'):
291 case _UT('C'):
292 case _UT('D'):
293 case _UT('E'):
294 case _UT('F'):
295 switch (read[2]) {
296 case _UT('0'):
297 case _UT('1'):
298 case _UT('2'):
299 case _UT('3'):
300 case _UT('4'):
301 case _UT('5'):
302 case _UT('6'):
303 case _UT('7'):
304 case _UT('8'):
305 case _UT('9'):
306 case _UT('a'):
307 case _UT('b'):
308 case _UT('c'):
309 case _UT('d'):
310 case _UT('e'):
311 case _UT('f'):
312 case _UT('A'):
313 case _UT('B'):
314 case _UT('C'):
315 case _UT('D'):
316 case _UT('E'):
317 case _UT('F'):
318 {
319 /* Percent group found */
320 const unsigned char left = URI_FUNC(HexdigToInt)(read[1]);
321 const unsigned char right = URI_FUNC(HexdigToInt)(read[2]);
322 const int code = 16 * left + right;
323 switch (code) {
324 case 10:
325 switch (breakConversion) {
326 case URI_BR_TO_LF:
327 if (!prevWasCr) {
328 write[0] = (URI_CHAR)10;
329 write++;
330 }
331 break;
332
333 case URI_BR_TO_CRLF:
334 if (!prevWasCr) {
335 write[0] = (URI_CHAR)13;
336 write[1] = (URI_CHAR)10;
337 write += 2;
338 }
339 break;
340
341 case URI_BR_TO_CR:
342 if (!prevWasCr) {
343 write[0] = (URI_CHAR)13;
344 write++;
345 }
346 break;
347
348 case URI_BR_DONT_TOUCH:
349 default:
350 write[0] = (URI_CHAR)10;
351 write++;
352
353 }
354 prevWasCr = URI_FALSE;
355 break;
356
357 case 13:
358 switch (breakConversion) {
359 case URI_BR_TO_LF:
360 write[0] = (URI_CHAR)10;
361 write++;
362 break;
363
364 case URI_BR_TO_CRLF:
365 write[0] = (URI_CHAR)13;
366 write[1] = (URI_CHAR)10;
367 write += 2;
368 break;
369
370 case URI_BR_TO_CR:
371 write[0] = (URI_CHAR)13;
372 write++;
373 break;
374
375 case URI_BR_DONT_TOUCH:
376 default:
377 write[0] = (URI_CHAR)13;
378 write++;
379
380 }
381 prevWasCr = URI_TRUE;
382 break;
383
384 default:
385 write[0] = (URI_CHAR)(code);
386 write++;
387
388 prevWasCr = URI_FALSE;
389
390 }
391 read += 3;
392 }
393 break;
394
395 default:
396 /* Copy two chars unmodified and */
397 /* look at this char again */
398 if (read > write) {
399 write[0] = read[0];
400 write[1] = read[1];
401 }
402 read += 2;
403 write += 2;
404
405 prevWasCr = URI_FALSE;
406 }
407 break;
408
409 default:
410 /* Copy one char unmodified and */
411 /* look at this char again */
412 if (read > write) {
413 write[0] = read[0];
414 }
415 read++;
416 write++;
417
418 prevWasCr = URI_FALSE;
419 }
420 break;
421
422 case _UT('+'):
423 if (plusToSpace) {
424 /* Convert '+' to ' ' */
425 write[0] = _UT(' ');
426 } else {
427 /* Copy one char unmodified */
428 if (read > write) {
429 write[0] = read[0];
430 }
431 }
432 read++;
433 write++;
434
435 prevWasCr = URI_FALSE;
436 break;
437
438 default:
439 /* Copy one char unmodified */
440 if (read > write) {
441 write[0] = read[0];
442 }
443 read++;
444 write++;
445
446 prevWasCr = URI_FALSE;
447 }
448 }
449}
450
451
452
453#endif
diff --git a/plugins/uriparser/UriFile.c b/plugins/uriparser/UriFile.c
new file mode 100644
index 00000000..9cf788fc
--- /dev/null
+++ b/plugins/uriparser/UriFile.c
@@ -0,0 +1,226 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriFile.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriFile.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66#endif
67
68
69
70static URI_INLINE int URI_FUNC(FilenameToUriString)(const URI_CHAR * filename,
71 URI_CHAR * uriString, UriBool fromUnix) {
72 const URI_CHAR * input = filename;
73 const URI_CHAR * lastSep = input - 1;
74 UriBool firstSegment = URI_TRUE;
75 URI_CHAR * output = uriString;
76 UriBool absolute;
77 UriBool is_windows_network;
78
79 if ((filename == NULL) || (uriString == NULL)) {
80 return URI_ERROR_NULL;
81 }
82
83 is_windows_network = (filename[0] == _UT('\\')) && (filename[1] == _UT('\\'));
84 absolute = fromUnix
85 ? (filename[0] == _UT('/'))
86 : ((filename[0] != _UT('\0')) && (filename[1] == _UT(':'))
87 || is_windows_network);
88
89 if (absolute) {
90 const URI_CHAR * const prefix = fromUnix
91 ? _UT("file://")
92 : is_windows_network
93 ? _UT("file:")
94 : _UT("file:///");
95 const int prefixLen = URI_STRLEN(prefix);
96
97 /* Copy prefix */
98 memcpy(uriString, prefix, prefixLen * sizeof(URI_CHAR));
99 output += prefixLen;
100 }
101
102 /* Copy and escape on the fly */
103 for (;;) {
104 if ((input[0] == _UT('\0'))
105 || (fromUnix && input[0] == _UT('/'))
106 || (!fromUnix && input[0] == _UT('\\'))) {
107 /* Copy text after last seperator */
108 if (lastSep + 1 < input) {
109 if (!fromUnix && absolute && (firstSegment == URI_TRUE)) {
110 /* Quick hack to not convert "C:" to "C%3A" */
111 const int charsToCopy = (int)(input - (lastSep + 1));
112 memcpy(output, lastSep + 1, charsToCopy * sizeof(URI_CHAR));
113 output += charsToCopy;
114 } else {
115 output = URI_FUNC(EscapeEx)(lastSep + 1, input, output,
116 URI_FALSE, URI_FALSE);
117 }
118 }
119 firstSegment = URI_FALSE;
120 }
121
122 if (input[0] == _UT('\0')) {
123 output[0] = _UT('\0');
124 break;
125 } else if (fromUnix && (input[0] == _UT('/'))) {
126 /* Copy separators unmodified */
127 output[0] = _UT('/');
128 output++;
129 lastSep = input;
130 } else if (!fromUnix && (input[0] == _UT('\\'))) {
131 /* Convert backslashes to forward slashes */
132 output[0] = _UT('/');
133 output++;
134 lastSep = input;
135 }
136 input++;
137 }
138
139 return URI_SUCCESS;
140}
141
142
143
144static URI_INLINE int URI_FUNC(UriStringToFilename)(const URI_CHAR * uriString,
145 URI_CHAR * filename, UriBool toUnix) {
146 if ((uriString == NULL) || (filename == NULL)) {
147 return URI_ERROR_NULL;
148 }
149
150 {
151 const UriBool file_two_slashes =
152 URI_STRNCMP(uriString, _UT("file://"), URI_STRLEN(_UT("file://"))) == 0;
153 const UriBool file_three_slashes = file_two_slashes
154 && (URI_STRNCMP(uriString, _UT("file:///"), URI_STRLEN(_UT("file:///"))) == 0);
155
156 const size_t charsToSkip = file_two_slashes
157 ? file_three_slashes
158 ? toUnix
159 /* file:///bin/bash */
160 ? URI_STRLEN(_UT("file://"))
161 /* file:///E:/Documents%20and%20Settings */
162 : URI_STRLEN(_UT("file:///"))
163 /* file://Server01/Letter.txt */
164 : URI_STRLEN(_UT("file://"))
165 : 0;
166 const size_t charsToCopy = URI_STRLEN(uriString + charsToSkip) + 1;
167
168 const UriBool is_windows_network_with_authority =
169 (toUnix == URI_FALSE)
170 && file_two_slashes
171 && ! file_three_slashes;
172
173 URI_CHAR * const unescape_target = is_windows_network_with_authority
174 ? (filename + 2)
175 : filename;
176
177 if (is_windows_network_with_authority) {
178 filename[0] = '\\';
179 filename[1] = '\\';
180 }
181
182 memcpy(unescape_target, uriString + charsToSkip, charsToCopy * sizeof(URI_CHAR));
183 URI_FUNC(UnescapeInPlaceEx)(filename, URI_FALSE, URI_BR_DONT_TOUCH);
184 }
185
186 /* Convert forward slashes to backslashes */
187 if (!toUnix) {
188 URI_CHAR * walker = filename;
189 while (walker[0] != _UT('\0')) {
190 if (walker[0] == _UT('/')) {
191 walker[0] = _UT('\\');
192 }
193 walker++;
194 }
195 }
196
197 return URI_SUCCESS;
198}
199
200
201
202int URI_FUNC(UnixFilenameToUriString)(const URI_CHAR * filename, URI_CHAR * uriString) {
203 return URI_FUNC(FilenameToUriString)(filename, uriString, URI_TRUE);
204}
205
206
207
208int URI_FUNC(WindowsFilenameToUriString)(const URI_CHAR * filename, URI_CHAR * uriString) {
209 return URI_FUNC(FilenameToUriString)(filename, uriString, URI_FALSE);
210}
211
212
213
214int URI_FUNC(UriStringToUnixFilename)(const URI_CHAR * uriString, URI_CHAR * filename) {
215 return URI_FUNC(UriStringToFilename)(uriString, filename, URI_TRUE);
216}
217
218
219
220int URI_FUNC(UriStringToWindowsFilename)(const URI_CHAR * uriString, URI_CHAR * filename) {
221 return URI_FUNC(UriStringToFilename)(uriString, filename, URI_FALSE);
222}
223
224
225
226#endif
diff --git a/plugins/uriparser/UriIp4.c b/plugins/uriparser/UriIp4.c
new file mode 100644
index 00000000..c00910bb
--- /dev/null
+++ b/plugins/uriparser/UriIp4.c
@@ -0,0 +1,329 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriIp4.c
42 * Holds the IPv4 parser implementation.
43 * NOTE: This source file includes itself twice.
44 */
45
46/* What encodings are enabled? */
47#include <uriparser/UriDefsConfig.h>
48#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
49/* Include SELF twice */
50# ifdef URI_ENABLE_ANSI
51# define URI_PASS_ANSI 1
52# include "UriIp4.c"
53# undef URI_PASS_ANSI
54# endif
55# ifdef URI_ENABLE_UNICODE
56# define URI_PASS_UNICODE 1
57# include "UriIp4.c"
58# undef URI_PASS_UNICODE
59# endif
60#else
61# ifdef URI_PASS_ANSI
62# include <uriparser/UriDefsAnsi.h>
63# else
64# include <uriparser/UriDefsUnicode.h>
65# endif
66
67
68
69#ifndef URI_DOXYGEN
70# include <uriparser/UriIp4.h>
71# include "UriIp4Base.h"
72# include <uriparser/UriBase.h>
73#endif
74
75
76
77/* Prototypes */
78static const URI_CHAR * URI_FUNC(ParseDecOctet)(UriIp4Parser * parser,
79 const URI_CHAR * first, const URI_CHAR * afterLast);
80static const URI_CHAR * URI_FUNC(ParseDecOctetOne)(UriIp4Parser * parser,
81 const URI_CHAR * first, const URI_CHAR * afterLast);
82static const URI_CHAR * URI_FUNC(ParseDecOctetTwo)(UriIp4Parser * parser,
83 const URI_CHAR * first, const URI_CHAR * afterLast);
84static const URI_CHAR * URI_FUNC(ParseDecOctetThree)(UriIp4Parser * parser,
85 const URI_CHAR * first, const URI_CHAR * afterLast);
86static const URI_CHAR * URI_FUNC(ParseDecOctetFour)(UriIp4Parser * parser,
87 const URI_CHAR * first, const URI_CHAR * afterLast);
88
89
90
91/*
92 * [ipFourAddress]->[decOctet]<.>[decOctet]<.>[decOctet]<.>[decOctet]
93 */
94int URI_FUNC(ParseIpFourAddress)(unsigned char * octetOutput,
95 const URI_CHAR * first, const URI_CHAR * afterLast) {
96 const URI_CHAR * after;
97 UriIp4Parser parser;
98
99 /* Essential checks */
100 if ((octetOutput == NULL) || (first == NULL)
101 || (afterLast <= first)) {
102 return URI_ERROR_SYNTAX;
103 }
104
105 /* Reset parser */
106 parser.stackCount = 0;
107
108 /* Octet #1 */
109 after = URI_FUNC(ParseDecOctet)(&parser, first, afterLast);
110 if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) {
111 return URI_ERROR_SYNTAX;
112 }
113 uriStackToOctet(&parser, octetOutput);
114
115 /* Octet #2 */
116 after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast);
117 if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) {
118 return URI_ERROR_SYNTAX;
119 }
120 uriStackToOctet(&parser, octetOutput + 1);
121
122 /* Octet #3 */
123 after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast);
124 if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) {
125 return URI_ERROR_SYNTAX;
126 }
127 uriStackToOctet(&parser, octetOutput + 2);
128
129 /* Octet #4 */
130 after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast);
131 if (after != afterLast) {
132 return URI_ERROR_SYNTAX;
133 }
134 uriStackToOctet(&parser, octetOutput + 3);
135
136 return URI_SUCCESS;
137}
138
139
140
141/*
142 * [decOctet]-><0>
143 * [decOctet]-><1>[decOctetOne]
144 * [decOctet]-><2>[decOctetTwo]
145 * [decOctet]-><3>[decOctetThree]
146 * [decOctet]-><4>[decOctetThree]
147 * [decOctet]-><5>[decOctetThree]
148 * [decOctet]-><6>[decOctetThree]
149 * [decOctet]-><7>[decOctetThree]
150 * [decOctet]-><8>[decOctetThree]
151 * [decOctet]-><9>[decOctetThree]
152 */
153static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctet)(UriIp4Parser * parser,
154 const URI_CHAR * first, const URI_CHAR * afterLast) {
155 if (first >= afterLast) {
156 return NULL;
157 }
158
159 switch (*first) {
160 case _UT('0'):
161 uriPushToStack(parser, 0);
162 return first + 1;
163
164 case _UT('1'):
165 uriPushToStack(parser, 1);
166 return (const URI_CHAR *)URI_FUNC(ParseDecOctetOne)(parser, first + 1, afterLast);
167
168 case _UT('2'):
169 uriPushToStack(parser, 2);
170 return (const URI_CHAR *)URI_FUNC(ParseDecOctetTwo)(parser, first + 1, afterLast);
171
172 case _UT('3'):
173 case _UT('4'):
174 case _UT('5'):
175 case _UT('6'):
176 case _UT('7'):
177 case _UT('8'):
178 case _UT('9'):
179 uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9')));
180 return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, afterLast);
181
182 default:
183 return NULL;
184 }
185}
186
187
188
189/*
190 * [decOctetOne]-><NULL>
191 * [decOctetOne]->[DIGIT][decOctetThree]
192 */
193static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetOne)(UriIp4Parser * parser,
194 const URI_CHAR * first, const URI_CHAR * afterLast) {
195 if (first >= afterLast) {
196 return afterLast;
197 }
198
199 switch (*first) {
200 case _UT('0'):
201 case _UT('1'):
202 case _UT('2'):
203 case _UT('3'):
204 case _UT('4'):
205 case _UT('5'):
206 case _UT('6'):
207 case _UT('7'):
208 case _UT('8'):
209 case _UT('9'):
210 uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9')));
211 return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, afterLast);
212
213 default:
214 return first;
215 }
216}
217
218
219
220/*
221 * [decOctetTwo]-><NULL>
222 * [decOctetTwo]-><0>[decOctetThree]
223 * [decOctetTwo]-><1>[decOctetThree]
224 * [decOctetTwo]-><2>[decOctetThree]
225 * [decOctetTwo]-><3>[decOctetThree]
226 * [decOctetTwo]-><4>[decOctetThree]
227 * [decOctetTwo]-><5>[decOctetFour]
228 * [decOctetTwo]-><6>
229 * [decOctetTwo]-><7>
230 * [decOctetTwo]-><8>
231 * [decOctetTwo]-><9>
232*/
233static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetTwo)(UriIp4Parser * parser,
234 const URI_CHAR * first, const URI_CHAR * afterLast) {
235 if (first >= afterLast) {
236 return afterLast;
237 }
238
239 switch (*first) {
240 case _UT('0'):
241 case _UT('1'):
242 case _UT('2'):
243 case _UT('3'):
244 case _UT('4'):
245 uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9')));
246 return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, afterLast);
247
248 case _UT('5'):
249 uriPushToStack(parser, 5);
250 return (const URI_CHAR *)URI_FUNC(ParseDecOctetFour)(parser, first + 1, afterLast);
251
252 case _UT('6'):
253 case _UT('7'):
254 case _UT('8'):
255 case _UT('9'):
256 uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9')));
257 return first + 1;
258
259 default:
260 return first;
261 }
262}
263
264
265
266/*
267 * [decOctetThree]-><NULL>
268 * [decOctetThree]->[DIGIT]
269 */
270static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetThree)(UriIp4Parser * parser,
271 const URI_CHAR * first, const URI_CHAR * afterLast) {
272 if (first >= afterLast) {
273 return afterLast;
274 }
275
276 switch (*first) {
277 case _UT('0'):
278 case _UT('1'):
279 case _UT('2'):
280 case _UT('3'):
281 case _UT('4'):
282 case _UT('5'):
283 case _UT('6'):
284 case _UT('7'):
285 case _UT('8'):
286 case _UT('9'):
287 uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9')));
288 return first + 1;
289
290 default:
291 return first;
292 }
293}
294
295
296
297/*
298 * [decOctetFour]-><NULL>
299 * [decOctetFour]-><0>
300 * [decOctetFour]-><1>
301 * [decOctetFour]-><2>
302 * [decOctetFour]-><3>
303 * [decOctetFour]-><4>
304 * [decOctetFour]-><5>
305 */
306static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetFour)(UriIp4Parser * parser,
307 const URI_CHAR * first, const URI_CHAR * afterLast) {
308 if (first >= afterLast) {
309 return afterLast;
310 }
311
312 switch (*first) {
313 case _UT('0'):
314 case _UT('1'):
315 case _UT('2'):
316 case _UT('3'):
317 case _UT('4'):
318 case _UT('5'):
319 uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9')));
320 return first + 1;
321
322 default:
323 return first;
324 }
325}
326
327
328
329#endif
diff --git a/plugins/uriparser/UriIp4.h b/plugins/uriparser/UriIp4.h
new file mode 100644
index 00000000..885a35ca
--- /dev/null
+++ b/plugins/uriparser/UriIp4.h
@@ -0,0 +1,92 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriIp4.h
42 * Holds the IPv4 parser interface.
43 * NOTE: This header includes itself twice.
44 */
45
46#if (defined(URI_PASS_ANSI) && !defined(URI_IP4_TWICE_H_ANSI)) \
47 || (defined(URI_PASS_UNICODE) && !defined(URI_IP4_TWICE_H_UNICODE)) \
48 || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
49/* What encodings are enabled? */
50#include "UriDefsConfig.h"
51#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
52/* Include SELF twice */
53# ifdef URI_ENABLE_ANSI
54# define URI_PASS_ANSI 1
55# include "UriIp4.h"
56# undef URI_PASS_ANSI
57# endif
58# ifdef URI_ENABLE_UNICODE
59# define URI_PASS_UNICODE 1
60# include "UriIp4.h"
61# undef URI_PASS_UNICODE
62# endif
63/* Only one pass for each encoding */
64#elif (defined(URI_PASS_ANSI) && !defined(URI_IP4_TWICE_H_ANSI) \
65 && defined(URI_ENABLE_ANSI)) || (defined(URI_PASS_UNICODE) \
66 && !defined(URI_IP4_TWICE_H_UNICODE) && defined(URI_ENABLE_UNICODE))
67# ifdef URI_PASS_ANSI
68# define URI_IP4_TWICE_H_ANSI 1
69# include "UriDefsAnsi.h"
70# else
71# define URI_IP4_TWICE_H_UNICODE 1
72# include "UriDefsUnicode.h"
73# include <wchar.h>
74# endif
75
76
77
78/**
79 * Converts a IPv4 text representation into four bytes.
80 *
81 * @param octetOutput Output destination
82 * @param first First character of IPv4 text to parse
83 * @param afterLast Position to stop parsing at
84 * @return Error code or 0 on success
85 */
86int URI_FUNC(ParseIpFourAddress)(unsigned char * octetOutput,
87 const URI_CHAR * first, const URI_CHAR * afterLast);
88
89
90
91#endif
92#endif
diff --git a/plugins/uriparser/UriIp4Base.c b/plugins/uriparser/UriIp4Base.c
new file mode 100644
index 00000000..5cd298fa
--- /dev/null
+++ b/plugins/uriparser/UriIp4Base.c
@@ -0,0 +1,96 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriIp4Base.c
42 * Holds code independent of the encoding pass.
43 */
44
45#ifndef URI_DOXYGEN
46# include "UriIp4Base.h"
47#endif
48
49
50
51void uriStackToOctet(UriIp4Parser * parser, unsigned char * octet) {
52 switch (parser->stackCount) {
53 case 1:
54 *octet = parser->stackOne;
55 break;
56
57 case 2:
58 *octet = parser->stackOne * 10
59 + parser->stackTwo;
60 break;
61
62 case 3:
63 *octet = parser->stackOne * 100
64 + parser->stackTwo * 10
65 + parser->stackThree;
66 break;
67
68 default:
69 ;
70 }
71 parser->stackCount = 0;
72}
73
74
75
76void uriPushToStack(UriIp4Parser * parser, unsigned char digit) {
77 switch (parser->stackCount) {
78 case 0:
79 parser->stackOne = digit;
80 parser->stackCount = 1;
81 break;
82
83 case 1:
84 parser->stackTwo = digit;
85 parser->stackCount = 2;
86 break;
87
88 case 2:
89 parser->stackThree = digit;
90 parser->stackCount = 3;
91 break;
92
93 default:
94 ;
95 }
96}
diff --git a/plugins/uriparser/UriIp4Base.h b/plugins/uriparser/UriIp4Base.h
new file mode 100644
index 00000000..23b838e2
--- /dev/null
+++ b/plugins/uriparser/UriIp4Base.h
@@ -0,0 +1,59 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40#ifndef URI_IP4_BASE_H
41#define URI_IP4_BASE_H 1
42
43
44
45typedef struct UriIp4ParserStruct {
46 unsigned char stackCount;
47 unsigned char stackOne;
48 unsigned char stackTwo;
49 unsigned char stackThree;
50} UriIp4Parser;
51
52
53
54void uriPushToStack(UriIp4Parser * parser, unsigned char digit);
55void uriStackToOctet(UriIp4Parser * parser, unsigned char * octet);
56
57
58
59#endif /* URI_IP4_BASE_H */
diff --git a/plugins/uriparser/UriNormalize.c b/plugins/uriparser/UriNormalize.c
new file mode 100644
index 00000000..49db9fff
--- /dev/null
+++ b/plugins/uriparser/UriNormalize.c
@@ -0,0 +1,728 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriNormalize.c
42 * Holds the RFC 3986 %URI normalization implementation.
43 * NOTE: This source file includes itself twice.
44 */
45
46/* What encodings are enabled? */
47#include <uriparser/UriDefsConfig.h>
48#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
49/* Include SELF twice */
50# ifdef URI_ENABLE_ANSI
51# define URI_PASS_ANSI 1
52# include "UriNormalize.c"
53# undef URI_PASS_ANSI
54# endif
55# ifdef URI_ENABLE_UNICODE
56# define URI_PASS_UNICODE 1
57# include "UriNormalize.c"
58# undef URI_PASS_UNICODE
59# endif
60#else
61# ifdef URI_PASS_ANSI
62# include <uriparser/UriDefsAnsi.h>
63# else
64# include <uriparser/UriDefsUnicode.h>
65# include <wchar.h>
66# endif
67
68
69
70#ifndef URI_DOXYGEN
71# include <uriparser/Uri.h>
72# include "UriNormalizeBase.h"
73# include "UriCommon.h"
74#endif
75
76
77
78static int URI_FUNC(NormalizeSyntaxEngine)(URI_TYPE(Uri) * uri, unsigned int inMask,
79 unsigned int * outMask);
80
81static UriBool URI_FUNC(MakeRangeOwner)(unsigned int * doneMask,
82 unsigned int maskTest, URI_TYPE(TextRange) * range);
83static UriBool URI_FUNC(MakeOwner)(URI_TYPE(Uri) * uri,
84 unsigned int * doneMask);
85
86static void URI_FUNC(FixPercentEncodingInplace)(const URI_CHAR * first,
87 const URI_CHAR ** afterLast);
88static UriBool URI_FUNC(FixPercentEncodingMalloc)(const URI_CHAR ** first,
89 const URI_CHAR ** afterLast);
90static void URI_FUNC(FixPercentEncodingEngine)(
91 const URI_CHAR * inFirst, const URI_CHAR * inAfterLast,
92 const URI_CHAR * outFirst, const URI_CHAR ** outAfterLast);
93
94static UriBool URI_FUNC(ContainsUppercaseLetters)(const URI_CHAR * first,
95 const URI_CHAR * afterLast);
96static UriBool URI_FUNC(ContainsUglyPercentEncoding)(const URI_CHAR * first,
97 const URI_CHAR * afterLast);
98
99static void URI_FUNC(LowercaseInplace)(const URI_CHAR * first,
100 const URI_CHAR * afterLast);
101static UriBool URI_FUNC(LowercaseMalloc)(const URI_CHAR ** first,
102 const URI_CHAR ** afterLast);
103
104static void URI_FUNC(PreventLeakage)(URI_TYPE(Uri) * uri,
105 unsigned int revertMask);
106
107
108
109static URI_INLINE void URI_FUNC(PreventLeakage)(URI_TYPE(Uri) * uri,
110 unsigned int revertMask) {
111 if (revertMask & URI_NORMALIZE_SCHEME) {
112 free((URI_CHAR *)uri->scheme.first);
113 uri->scheme.first = NULL;
114 uri->scheme.afterLast = NULL;
115 }
116
117 if (revertMask & URI_NORMALIZE_USER_INFO) {
118 free((URI_CHAR *)uri->userInfo.first);
119 uri->userInfo.first = NULL;
120 uri->userInfo.afterLast = NULL;
121 }
122
123 if (revertMask & URI_NORMALIZE_HOST) {
124 if (uri->hostData.ipFuture.first != NULL) {
125 /* IPvFuture */
126 free((URI_CHAR *)uri->hostData.ipFuture.first);
127 uri->hostData.ipFuture.first = NULL;
128 uri->hostData.ipFuture.afterLast = NULL;
129 uri->hostText.first = NULL;
130 uri->hostText.afterLast = NULL;
131 } else if ((uri->hostText.first != NULL)
132 && (uri->hostData.ip4 == NULL)
133 && (uri->hostData.ip6 == NULL)) {
134 /* Regname */
135 free((URI_CHAR *)uri->hostText.first);
136 uri->hostText.first = NULL;
137 uri->hostText.afterLast = NULL;
138 }
139 }
140
141 /* NOTE: Port cannot happen! */
142
143 if (revertMask & URI_NORMALIZE_PATH) {
144 URI_TYPE(PathSegment) * walker = uri->pathHead;
145 while (walker != NULL) {
146 URI_TYPE(PathSegment) * const next = walker->next;
147 if (walker->text.afterLast > walker->text.first) {
148 free((URI_CHAR *)walker->text.first);
149 }
150 free(walker);
151 walker = next;
152 }
153 uri->pathHead = NULL;
154 uri->pathTail = NULL;
155 }
156
157 if (revertMask & URI_NORMALIZE_QUERY) {
158 free((URI_CHAR *)uri->query.first);
159 uri->query.first = NULL;
160 uri->query.afterLast = NULL;
161 }
162
163 if (revertMask & URI_NORMALIZE_FRAGMENT) {
164 free((URI_CHAR *)uri->fragment.first);
165 uri->fragment.first = NULL;
166 uri->fragment.afterLast = NULL;
167 }
168}
169
170
171
172static URI_INLINE UriBool URI_FUNC(ContainsUppercaseLetters)(const URI_CHAR * first,
173 const URI_CHAR * afterLast) {
174 if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) {
175 const URI_CHAR * i = first;
176 for (; i < afterLast; i++) {
177 /* 6.2.2.1 Case Normalization: uppercase letters in scheme or host */
178 if ((*i >= _UT('A')) && (*i <= _UT('Z'))) {
179 return URI_TRUE;
180 }
181 }
182 }
183 return URI_FALSE;
184}
185
186
187
188static URI_INLINE UriBool URI_FUNC(ContainsUglyPercentEncoding)(const URI_CHAR * first,
189 const URI_CHAR * afterLast) {
190 if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) {
191 const URI_CHAR * i = first;
192 for (; i + 2 < afterLast; i++) {
193 if (i[0] == _UT('%')) {
194 /* 6.2.2.1 Case Normalization: *
195 * lowercase percent-encodings */
196 if (((i[1] >= _UT('a')) && (i[1] <= _UT('f')))
197 || ((i[2] >= _UT('a')) && (i[2] <= _UT('f')))) {
198 return URI_TRUE;
199 } else {
200 /* 6.2.2.2 Percent-Encoding Normalization: *
201 * percent-encoded unreserved characters */
202 const unsigned char left = URI_FUNC(HexdigToInt)(i[1]);
203 const unsigned char right = URI_FUNC(HexdigToInt)(i[2]);
204 const int code = 16 * left + right;
205 if (uriIsUnreserved(code)) {
206 return URI_TRUE;
207 }
208 }
209 }
210 }
211 }
212 return URI_FALSE;
213}
214
215
216
217static URI_INLINE void URI_FUNC(LowercaseInplace)(const URI_CHAR * first,
218 const URI_CHAR * afterLast) {
219 if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) {
220 URI_CHAR * i = (URI_CHAR *)first;
221 const int lowerUpperDiff = (_UT('a') - _UT('A'));
222 for (; i < afterLast; i++) {
223 if ((*i >= _UT('A')) && (*i <=_UT('Z'))) {
224 *i = (URI_CHAR)(*i + lowerUpperDiff);
225 }
226 }
227 }
228}
229
230
231
232static URI_INLINE UriBool URI_FUNC(LowercaseMalloc)(const URI_CHAR ** first,
233 const URI_CHAR ** afterLast) {
234 int lenInChars;
235 const int lowerUpperDiff = (_UT('a') - _UT('A'));
236 URI_CHAR * buffer;
237 int i = 0;
238
239 if ((first == NULL) || (afterLast == NULL) || (*first == NULL)
240 || (*afterLast == NULL)) {
241 return URI_FALSE;
242 }
243
244 lenInChars = (int)(*afterLast - *first);
245 if (lenInChars == 0) {
246 return URI_TRUE;
247 } else if (lenInChars < 0) {
248 return URI_FALSE;
249 }
250
251 buffer = malloc(lenInChars * sizeof(URI_CHAR));
252 if (buffer == NULL) {
253 return URI_FALSE;
254 }
255
256 for (; i < lenInChars; i++) {
257 if (((*first)[i] >= _UT('A')) && ((*first)[i] <=_UT('Z'))) {
258 buffer[i] = (URI_CHAR)((*first)[i] + lowerUpperDiff);
259 } else {
260 buffer[i] = (*first)[i];
261 }
262 }
263
264 *first = buffer;
265 *afterLast = buffer + lenInChars;
266 return URI_TRUE;
267}
268
269
270
271/* NOTE: Implementation must stay inplace-compatible */
272static URI_INLINE void URI_FUNC(FixPercentEncodingEngine)(
273 const URI_CHAR * inFirst, const URI_CHAR * inAfterLast,
274 const URI_CHAR * outFirst, const URI_CHAR ** outAfterLast) {
275 URI_CHAR * write = (URI_CHAR *)outFirst;
276 const int lenInChars = (int)(inAfterLast - inFirst);
277 int i = 0;
278
279 /* All but last two */
280 for (; i + 2 < lenInChars; i++) {
281 if (inFirst[i] != _UT('%')) {
282 write[0] = inFirst[i];
283 write++;
284 } else {
285 /* 6.2.2.2 Percent-Encoding Normalization: *
286 * percent-encoded unreserved characters */
287 const URI_CHAR one = inFirst[i + 1];
288 const URI_CHAR two = inFirst[i + 2];
289 const unsigned char left = URI_FUNC(HexdigToInt)(one);
290 const unsigned char right = URI_FUNC(HexdigToInt)(two);
291 const int code = 16 * left + right;
292 if (uriIsUnreserved(code)) {
293 write[0] = (URI_CHAR)(code);
294 write++;
295 } else {
296 /* 6.2.2.1 Case Normalization: *
297 * lowercase percent-encodings */
298 write[0] = _UT('%');
299 write[1] = URI_FUNC(HexToLetter)(left);
300 write[2] = URI_FUNC(HexToLetter)(right);
301 write += 3;
302 }
303
304 i += 2; /* For the two chars of the percent group we just ate */
305 }
306 }
307
308 /* Last two */
309 for (; i < lenInChars; i++) {
310 write[0] = inFirst[i];
311 write++;
312 }
313
314 *outAfterLast = write;
315}
316
317
318
319static URI_INLINE void URI_FUNC(FixPercentEncodingInplace)(const URI_CHAR * first,
320 const URI_CHAR ** afterLast) {
321 /* Death checks */
322 if ((first == NULL) || (afterLast == NULL) || (*afterLast == NULL)) {
323 return;
324 }
325
326 /* Fix inplace */
327 URI_FUNC(FixPercentEncodingEngine)(first, *afterLast, first, afterLast);
328}
329
330
331
332static URI_INLINE UriBool URI_FUNC(FixPercentEncodingMalloc)(const URI_CHAR ** first,
333 const URI_CHAR ** afterLast) {
334 int lenInChars;
335 URI_CHAR * buffer;
336
337 /* Death checks */
338 if ((first == NULL) || (afterLast == NULL)
339 || (*first == NULL) || (*afterLast == NULL)) {
340 return URI_FALSE;
341 }
342
343 /* Old text length */
344 lenInChars = (int)(*afterLast - *first);
345 if (lenInChars == 0) {
346 return URI_TRUE;
347 } else if (lenInChars < 0) {
348 return URI_FALSE;
349 }
350
351 /* New buffer */
352 buffer = malloc(lenInChars * sizeof(URI_CHAR));
353 if (buffer == NULL) {
354 return URI_FALSE;
355 }
356
357 /* Fix on copy */
358 URI_FUNC(FixPercentEncodingEngine)(*first, *afterLast, buffer, afterLast);
359 *first = buffer;
360 return URI_TRUE;
361}
362
363
364
365static URI_INLINE UriBool URI_FUNC(MakeRangeOwner)(unsigned int * doneMask,
366 unsigned int maskTest, URI_TYPE(TextRange) * range) {
367 if (((*doneMask & maskTest) == 0)
368 && (range->first != NULL)
369 && (range->afterLast != NULL)
370 && (range->afterLast > range->first)) {
371 const int lenInChars = (int)(range->afterLast - range->first);
372 const int lenInBytes = lenInChars * sizeof(URI_CHAR);
373 URI_CHAR * dup = malloc(lenInBytes);
374 if (dup == NULL) {
375 return URI_FALSE; /* Raises malloc error */
376 }
377 memcpy(dup, range->first, lenInBytes);
378 range->first = dup;
379 range->afterLast = dup + lenInChars;
380 *doneMask |= maskTest;
381 }
382 return URI_TRUE;
383}
384
385
386
387static URI_INLINE UriBool URI_FUNC(MakeOwner)(URI_TYPE(Uri) * uri,
388 unsigned int * doneMask) {
389 URI_TYPE(PathSegment) * walker = uri->pathHead;
390 if (!URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_SCHEME,
391 &(uri->scheme))
392 || !URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_USER_INFO,
393 &(uri->userInfo))
394 || !URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_QUERY,
395 &(uri->query))
396 || !URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_FRAGMENT,
397 &(uri->fragment))) {
398 return URI_FALSE; /* Raises malloc error */
399 }
400
401 /* Host */
402 if ((*doneMask & URI_NORMALIZE_HOST) == 0) {
403 if ((uri->hostData.ip4 == NULL)
404 && (uri->hostData.ip6 == NULL)) {
405 if (uri->hostData.ipFuture.first != NULL) {
406 /* IPvFuture */
407 if (!URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_HOST,
408 &(uri->hostData.ipFuture))) {
409 return URI_FALSE; /* Raises malloc error */
410 }
411 uri->hostText.first = uri->hostData.ipFuture.first;
412 uri->hostText.afterLast = uri->hostData.ipFuture.afterLast;
413 } else if (uri->hostText.first != NULL) {
414 /* Regname */
415 if (!URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_HOST,
416 &(uri->hostText))) {
417 return URI_FALSE; /* Raises malloc error */
418 }
419 }
420 }
421 }
422
423 /* Path */
424 if ((*doneMask & URI_NORMALIZE_PATH) == 0) {
425 while (walker != NULL) {
426 if (!URI_FUNC(MakeRangeOwner)(doneMask, 0, &(walker->text))) {
427 /* Free allocations done so far and kill path */
428
429 /* Kill path to one before walker (if any) */
430 URI_TYPE(PathSegment) * ranger = uri->pathHead;
431 while (ranger != walker) {
432 URI_TYPE(PathSegment) * const next = ranger->next;
433 if ((ranger->text.first != NULL)
434 && (ranger->text.afterLast != NULL)
435 && (ranger->text.afterLast > ranger->text.first)) {
436 free((URI_CHAR *)ranger->text.first);
437 free(ranger);
438 }
439 ranger = next;
440 }
441
442 /* Kill path from walker */
443 while (walker != NULL) {
444 URI_TYPE(PathSegment) * const next = walker->next;
445 free(walker);
446 walker = next;
447 }
448
449 uri->pathHead = NULL;
450 uri->pathTail = NULL;
451 return URI_FALSE; /* Raises malloc error */
452 }
453 walker = walker->next;
454 }
455 *doneMask |= URI_NORMALIZE_PATH;
456 }
457
458 /* Port text, must come last so we don't have to undo that one if it fails. *
459 * Otherwise we would need and extra enum flag for it although the port *
460 * cannot go unnormalized... */
461 if (!URI_FUNC(MakeRangeOwner)(doneMask, 0, &(uri->portText))) {
462 return URI_FALSE; /* Raises malloc error */
463 }
464
465 return URI_TRUE;
466}
467
468
469
470unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)(const URI_TYPE(Uri) * uri) {
471 unsigned int res;
472#if defined(__GNUC__) && ((__GNUC__ > 4) \
473 || ((__GNUC__ == 4) && defined(__GNUC_MINOR__) && (__GNUC_MINOR__ >= 2)))
474 /* Slower code that fixes a warning, not sure if this is a smart idea */
475 URI_TYPE(Uri) writeableClone;
476 memcpy(&writeableClone, uri, 1 * sizeof(URI_TYPE(Uri)));
477 URI_FUNC(NormalizeSyntaxEngine)(&writeableClone, 0, &res);
478#else
479 URI_FUNC(NormalizeSyntaxEngine)((URI_TYPE(Uri) *)uri, 0, &res);
480#endif
481 return res;
482}
483
484
485
486int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, unsigned int mask) {
487 return URI_FUNC(NormalizeSyntaxEngine)(uri, mask, NULL);
488}
489
490
491
492int URI_FUNC(NormalizeSyntax)(URI_TYPE(Uri) * uri) {
493 return URI_FUNC(NormalizeSyntaxEx)(uri, (unsigned int)-1);
494}
495
496
497
498static URI_INLINE int URI_FUNC(NormalizeSyntaxEngine)(URI_TYPE(Uri) * uri, unsigned int inMask, unsigned int * outMask) {
499 unsigned int doneMask = URI_NORMALIZED;
500 if (uri == NULL) {
501 if (outMask != NULL) {
502 *outMask = URI_NORMALIZED;
503 return URI_SUCCESS;
504 } else {
505 return URI_ERROR_NULL;
506 }
507 }
508
509 if (outMask != NULL) {
510 /* Reset mask */
511 *outMask = URI_NORMALIZED;
512 } else if (inMask == URI_NORMALIZED) {
513 /* Nothing to do */
514 return URI_SUCCESS;
515 }
516
517 /* Scheme, host */
518 if (outMask != NULL) {
519 const UriBool normalizeScheme = URI_FUNC(ContainsUppercaseLetters)(
520 uri->scheme.first, uri->scheme.afterLast);
521 const UriBool normalizeHostCase = URI_FUNC(ContainsUppercaseLetters)(
522 uri->hostText.first, uri->hostText.afterLast);
523 if (normalizeScheme) {
524 *outMask |= URI_NORMALIZE_SCHEME;
525 }
526
527 if (normalizeHostCase) {
528 *outMask |= URI_NORMALIZE_HOST;
529 } else {
530 const UriBool normalizeHostPrecent = URI_FUNC(ContainsUglyPercentEncoding)(
531 uri->hostText.first, uri->hostText.afterLast);
532 if (normalizeHostPrecent) {
533 *outMask |= URI_NORMALIZE_HOST;
534 }
535 }
536 } else {
537 /* Scheme */
538 if ((inMask & URI_NORMALIZE_SCHEME) && (uri->scheme.first != NULL)) {
539 if (uri->owner) {
540 URI_FUNC(LowercaseInplace)(uri->scheme.first, uri->scheme.afterLast);
541 } else {
542 if (!URI_FUNC(LowercaseMalloc)(&(uri->scheme.first), &(uri->scheme.afterLast))) {
543 URI_FUNC(PreventLeakage)(uri, doneMask);
544 return URI_ERROR_MALLOC;
545 }
546 doneMask |= URI_NORMALIZE_SCHEME;
547 }
548 }
549
550 /* Host */
551 if (inMask & URI_NORMALIZE_HOST) {
552 if (uri->hostData.ipFuture.first != NULL) {
553 /* IPvFuture */
554 if (uri->owner) {
555 URI_FUNC(LowercaseInplace)(uri->hostData.ipFuture.first,
556 uri->hostData.ipFuture.afterLast);
557 } else {
558 if (!URI_FUNC(LowercaseMalloc)(&(uri->hostData.ipFuture.first),
559 &(uri->hostData.ipFuture.afterLast))) {
560 URI_FUNC(PreventLeakage)(uri, doneMask);
561 return URI_ERROR_MALLOC;
562 }
563 doneMask |= URI_NORMALIZE_HOST;
564 }
565 uri->hostText.first = uri->hostData.ipFuture.first;
566 uri->hostText.afterLast = uri->hostData.ipFuture.afterLast;
567 } else if ((uri->hostText.first != NULL)
568 && (uri->hostData.ip4 == NULL)
569 && (uri->hostData.ip6 == NULL)) {
570 /* Regname */
571 if (uri->owner) {
572 URI_FUNC(FixPercentEncodingInplace)(uri->hostText.first,
573 &(uri->hostText.afterLast));
574 } else {
575 if (!URI_FUNC(FixPercentEncodingMalloc)(
576 &(uri->hostText.first),
577 &(uri->hostText.afterLast))) {
578 URI_FUNC(PreventLeakage)(uri, doneMask);
579 return URI_ERROR_MALLOC;
580 }
581 doneMask |= URI_NORMALIZE_HOST;
582 }
583
584 URI_FUNC(LowercaseInplace)(uri->hostText.first,
585 uri->hostText.afterLast);
586 }
587 }
588 }
589
590 /* User info */
591 if (outMask != NULL) {
592 const UriBool normalizeUserInfo = URI_FUNC(ContainsUglyPercentEncoding)(
593 uri->userInfo.first, uri->userInfo.afterLast);
594 if (normalizeUserInfo) {
595 *outMask |= URI_NORMALIZE_USER_INFO;
596 }
597 } else {
598 if ((inMask & URI_NORMALIZE_USER_INFO) && (uri->userInfo.first != NULL)) {
599 if (uri->owner) {
600 URI_FUNC(FixPercentEncodingInplace)(uri->userInfo.first, &(uri->userInfo.afterLast));
601 } else {
602 if (!URI_FUNC(FixPercentEncodingMalloc)(&(uri->userInfo.first),
603 &(uri->userInfo.afterLast))) {
604 URI_FUNC(PreventLeakage)(uri, doneMask);
605 return URI_ERROR_MALLOC;
606 }
607 doneMask |= URI_NORMALIZE_USER_INFO;
608 }
609 }
610 }
611
612 /* Path */
613 if (outMask != NULL) {
614 const URI_TYPE(PathSegment) * walker = uri->pathHead;
615 while (walker != NULL) {
616 const URI_CHAR * const first = walker->text.first;
617 const URI_CHAR * const afterLast = walker->text.afterLast;
618 if ((first != NULL)
619 && (afterLast != NULL)
620 && (afterLast > first)
621 && (
622 (((afterLast - first) == 1)
623 && (first[0] == _UT('.')))
624 ||
625 (((afterLast - first) == 2)
626 && (first[0] == _UT('.'))
627 && (first[1] == _UT('.')))
628 ||
629 URI_FUNC(ContainsUglyPercentEncoding)(first, afterLast)
630 )) {
631 *outMask |= URI_NORMALIZE_PATH;
632 break;
633 }
634 walker = walker->next;
635 }
636 } else if (inMask & URI_NORMALIZE_PATH) {
637 URI_TYPE(PathSegment) * walker;
638 const UriBool relative = ((uri->scheme.first == NULL)
639 && !uri->absolutePath) ? URI_TRUE : URI_FALSE;
640
641 /* Fix percent-encoding for each segment */
642 walker = uri->pathHead;
643 if (uri->owner) {
644 while (walker != NULL) {
645 URI_FUNC(FixPercentEncodingInplace)(walker->text.first, &(walker->text.afterLast));
646 walker = walker->next;
647 }
648 } else {
649 while (walker != NULL) {
650 if (!URI_FUNC(FixPercentEncodingMalloc)(&(walker->text.first),
651 &(walker->text.afterLast))) {
652 URI_FUNC(PreventLeakage)(uri, doneMask);
653 return URI_ERROR_MALLOC;
654 }
655 walker = walker->next;
656 }
657 doneMask |= URI_NORMALIZE_PATH;
658 }
659
660 /* 6.2.2.3 Path Segment Normalization */
661 if (!URI_FUNC(RemoveDotSegmentsEx)(uri, relative,
662 (uri->owner == URI_TRUE)
663 || ((doneMask & URI_NORMALIZE_PATH) != 0)
664 )) {
665 URI_FUNC(PreventLeakage)(uri, doneMask);
666 return URI_ERROR_MALLOC;
667 }
668 URI_FUNC(FixEmptyTrailSegment)(uri);
669 }
670
671 /* Query, fragment */
672 if (outMask != NULL) {
673 const UriBool normalizeQuery = URI_FUNC(ContainsUglyPercentEncoding)(
674 uri->query.first, uri->query.afterLast);
675 const UriBool normalizeFragment = URI_FUNC(ContainsUglyPercentEncoding)(
676 uri->fragment.first, uri->fragment.afterLast);
677 if (normalizeQuery) {
678 *outMask |= URI_NORMALIZE_QUERY;
679 }
680
681 if (normalizeFragment) {
682 *outMask |= URI_NORMALIZE_FRAGMENT;
683 }
684 } else {
685 /* Query */
686 if ((inMask & URI_NORMALIZE_QUERY) && (uri->query.first != NULL)) {
687 if (uri->owner) {
688 URI_FUNC(FixPercentEncodingInplace)(uri->query.first, &(uri->query.afterLast));
689 } else {
690 if (!URI_FUNC(FixPercentEncodingMalloc)(&(uri->query.first),
691 &(uri->query.afterLast))) {
692 URI_FUNC(PreventLeakage)(uri, doneMask);
693 return URI_ERROR_MALLOC;
694 }
695 doneMask |= URI_NORMALIZE_QUERY;
696 }
697 }
698
699 /* Fragment */
700 if ((inMask & URI_NORMALIZE_FRAGMENT) && (uri->fragment.first != NULL)) {
701 if (uri->owner) {
702 URI_FUNC(FixPercentEncodingInplace)(uri->fragment.first, &(uri->fragment.afterLast));
703 } else {
704 if (!URI_FUNC(FixPercentEncodingMalloc)(&(uri->fragment.first),
705 &(uri->fragment.afterLast))) {
706 URI_FUNC(PreventLeakage)(uri, doneMask);
707 return URI_ERROR_MALLOC;
708 }
709 doneMask |= URI_NORMALIZE_FRAGMENT;
710 }
711 }
712 }
713
714 /* Dup all not duped yet */
715 if ((outMask == NULL) && !uri->owner) {
716 if (!URI_FUNC(MakeOwner)(uri, &doneMask)) {
717 URI_FUNC(PreventLeakage)(uri, doneMask);
718 return URI_ERROR_MALLOC;
719 }
720 uri->owner = URI_TRUE;
721 }
722
723 return URI_SUCCESS;
724}
725
726
727
728#endif
diff --git a/plugins/uriparser/UriNormalizeBase.c b/plugins/uriparser/UriNormalizeBase.c
new file mode 100644
index 00000000..bac58886
--- /dev/null
+++ b/plugins/uriparser/UriNormalizeBase.c
@@ -0,0 +1,119 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40#ifndef URI_DOXYGEN
41# include "UriNormalizeBase.h"
42#endif
43
44
45
46UriBool uriIsUnreserved(int code) {
47 switch (code) {
48 case L'a': /* ALPHA */
49 case L'A':
50 case L'b':
51 case L'B':
52 case L'c':
53 case L'C':
54 case L'd':
55 case L'D':
56 case L'e':
57 case L'E':
58 case L'f':
59 case L'F':
60 case L'g':
61 case L'G':
62 case L'h':
63 case L'H':
64 case L'i':
65 case L'I':
66 case L'j':
67 case L'J':
68 case L'k':
69 case L'K':
70 case L'l':
71 case L'L':
72 case L'm':
73 case L'M':
74 case L'n':
75 case L'N':
76 case L'o':
77 case L'O':
78 case L'p':
79 case L'P':
80 case L'q':
81 case L'Q':
82 case L'r':
83 case L'R':
84 case L's':
85 case L'S':
86 case L't':
87 case L'T':
88 case L'u':
89 case L'U':
90 case L'v':
91 case L'V':
92 case L'w':
93 case L'W':
94 case L'x':
95 case L'X':
96 case L'y':
97 case L'Y':
98 case L'z':
99 case L'Z':
100 case L'0': /* DIGIT */
101 case L'1':
102 case L'2':
103 case L'3':
104 case L'4':
105 case L'5':
106 case L'6':
107 case L'7':
108 case L'8':
109 case L'9':
110 case L'-': /* "-" / "." / "_" / "~" */
111 case L'.':
112 case L'_':
113 case L'~':
114 return URI_TRUE;
115
116 default:
117 return URI_FALSE;
118 }
119}
diff --git a/plugins/uriparser/UriNormalizeBase.h b/plugins/uriparser/UriNormalizeBase.h
new file mode 100644
index 00000000..9f3abd55
--- /dev/null
+++ b/plugins/uriparser/UriNormalizeBase.h
@@ -0,0 +1,53 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40#ifndef URI_NORMALIZE_BASE_H
41#define URI_NORMALIZE_BASE_H 1
42
43
44
45#include <uriparser/UriBase.h>
46
47
48
49UriBool uriIsUnreserved(int code);
50
51
52
53#endif /* URI_NORMALIZE_BASE_H */
diff --git a/plugins/uriparser/UriParse.c b/plugins/uriparser/UriParse.c
new file mode 100644
index 00000000..e3cdc68d
--- /dev/null
+++ b/plugins/uriparser/UriParse.c
@@ -0,0 +1,2241 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/**
41 * @file UriParse.c
42 * Holds the RFC 3986 %URI parsing implementation.
43 * NOTE: This source file includes itself twice.
44 */
45
46/* What encodings are enabled? */
47#include <uriparser/UriDefsConfig.h>
48#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
49/* Include SELF twice */
50# ifdef URI_ENABLE_ANSI
51# define URI_PASS_ANSI 1
52# include "UriParse.c"
53# undef URI_PASS_ANSI
54# endif
55# ifdef URI_ENABLE_UNICODE
56# define URI_PASS_UNICODE 1
57# include "UriParse.c"
58# undef URI_PASS_UNICODE
59# endif
60#else
61# ifdef URI_PASS_ANSI
62# include <uriparser/UriDefsAnsi.h>
63# else
64# include <uriparser/UriDefsUnicode.h>
65# include <wchar.h>
66# endif
67
68
69
70#ifndef URI_DOXYGEN
71# include <uriparser/Uri.h>
72# include <uriparser/UriIp4.h>
73# include "UriCommon.h"
74# include "UriParseBase.h"
75#endif
76
77
78
79#define URI_SET_DIGIT \
80 _UT('0'): \
81 case _UT('1'): \
82 case _UT('2'): \
83 case _UT('3'): \
84 case _UT('4'): \
85 case _UT('5'): \
86 case _UT('6'): \
87 case _UT('7'): \
88 case _UT('8'): \
89 case _UT('9')
90
91#define URI_SET_HEX_LETTER_UPPER \
92 _UT('A'): \
93 case _UT('B'): \
94 case _UT('C'): \
95 case _UT('D'): \
96 case _UT('E'): \
97 case _UT('F')
98
99#define URI_SET_HEX_LETTER_LOWER \
100 _UT('a'): \
101 case _UT('b'): \
102 case _UT('c'): \
103 case _UT('d'): \
104 case _UT('e'): \
105 case _UT('f')
106
107#define URI_SET_HEXDIG \
108 URI_SET_DIGIT: \
109 case URI_SET_HEX_LETTER_UPPER: \
110 case URI_SET_HEX_LETTER_LOWER
111
112#define URI_SET_ALPHA \
113 URI_SET_HEX_LETTER_UPPER: \
114 case URI_SET_HEX_LETTER_LOWER: \
115 case _UT('g'): \
116 case _UT('G'): \
117 case _UT('h'): \
118 case _UT('H'): \
119 case _UT('i'): \
120 case _UT('I'): \
121 case _UT('j'): \
122 case _UT('J'): \
123 case _UT('k'): \
124 case _UT('K'): \
125 case _UT('l'): \
126 case _UT('L'): \
127 case _UT('m'): \
128 case _UT('M'): \
129 case _UT('n'): \
130 case _UT('N'): \
131 case _UT('o'): \
132 case _UT('O'): \
133 case _UT('p'): \
134 case _UT('P'): \
135 case _UT('q'): \
136 case _UT('Q'): \
137 case _UT('r'): \
138 case _UT('R'): \
139 case _UT('s'): \
140 case _UT('S'): \
141 case _UT('t'): \
142 case _UT('T'): \
143 case _UT('u'): \
144 case _UT('U'): \
145 case _UT('v'): \
146 case _UT('V'): \
147 case _UT('w'): \
148 case _UT('W'): \
149 case _UT('x'): \
150 case _UT('X'): \
151 case _UT('y'): \
152 case _UT('Y'): \
153 case _UT('z'): \
154 case _UT('Z')
155
156
157
158static const URI_CHAR * URI_FUNC(ParseAuthority)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
159static const URI_CHAR * URI_FUNC(ParseAuthorityTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
160static const URI_CHAR * URI_FUNC(ParseHexZero)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
161static const URI_CHAR * URI_FUNC(ParseHierPart)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
162static const URI_CHAR * URI_FUNC(ParseIpFutLoop)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
163static const URI_CHAR * URI_FUNC(ParseIpFutStopGo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
164static const URI_CHAR * URI_FUNC(ParseIpLit2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
165static const URI_CHAR * URI_FUNC(ParseIPv6address2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
166static const URI_CHAR * URI_FUNC(ParseMustBeSegmentNzNc)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
167static const URI_CHAR * URI_FUNC(ParseOwnHost)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
168static const URI_CHAR * URI_FUNC(ParseOwnHost2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
169static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
170static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfoNz)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
171static const URI_CHAR * URI_FUNC(ParseOwnPortUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
172static const URI_CHAR * URI_FUNC(ParseOwnUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
173static const URI_CHAR * URI_FUNC(ParsePartHelperTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
174static const URI_CHAR * URI_FUNC(ParsePathAbsEmpty)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
175static const URI_CHAR * URI_FUNC(ParsePathAbsNoLeadSlash)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
176static const URI_CHAR * URI_FUNC(ParsePathRootless)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
177static const URI_CHAR * URI_FUNC(ParsePchar)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
178static const URI_CHAR * URI_FUNC(ParsePctEncoded)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
179static const URI_CHAR * URI_FUNC(ParsePctSubUnres)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
180static const URI_CHAR * URI_FUNC(ParsePort)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
181static const URI_CHAR * URI_FUNC(ParseQueryFrag)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
182static const URI_CHAR * URI_FUNC(ParseSegment)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
183static const URI_CHAR * URI_FUNC(ParseSegmentNz)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
184static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
185static const URI_CHAR * URI_FUNC(ParseUriReference)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
186static const URI_CHAR * URI_FUNC(ParseUriTail)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
187static const URI_CHAR * URI_FUNC(ParseUriTailTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
188static const URI_CHAR * URI_FUNC(ParseZeroMoreSlashSegs)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
189
190static UriBool URI_FUNC(OnExitOwnHost2)(URI_TYPE(ParserState) * state, const URI_CHAR * first);
191static UriBool URI_FUNC(OnExitOwnHostUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first);
192static UriBool URI_FUNC(OnExitOwnPortUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first);
193static UriBool URI_FUNC(OnExitSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, const URI_CHAR * first);
194static void URI_FUNC(OnExitPartHelperTwo)(URI_TYPE(ParserState) * state);
195
196static void URI_FUNC(ResetParserState)(URI_TYPE(ParserState) * state);
197
198static UriBool URI_FUNC(PushPathSegment)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast);
199
200static void URI_FUNC(StopSyntax)(URI_TYPE(ParserState) * state, const URI_CHAR * errorPos);
201static void URI_FUNC(StopMalloc)(URI_TYPE(ParserState) * state);
202
203
204
205static URI_INLINE void URI_FUNC(StopSyntax)(URI_TYPE(ParserState) * state,
206 const URI_CHAR * errorPos) {
207 URI_FUNC(FreeUriMembers)(state->uri);
208 state->errorPos = errorPos;
209 state->errorCode = URI_ERROR_SYNTAX;
210}
211
212
213
214static URI_INLINE void URI_FUNC(StopMalloc)(URI_TYPE(ParserState) * state) {
215 URI_FUNC(FreeUriMembers)(state->uri);
216 state->errorPos = NULL;
217 state->errorCode = URI_ERROR_MALLOC;
218}
219
220
221
222/*
223 * [authority]-><[>[ipLit2][authorityTwo]
224 * [authority]->[ownHostUserInfoNz]
225 * [authority]-><NULL>
226 */
227static URI_INLINE const URI_CHAR * URI_FUNC(ParseAuthority)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
228 if (first >= afterLast) {
229 /* "" regname host */
230 state->uri->hostText.first = URI_FUNC(SafeToPointTo);
231 state->uri->hostText.afterLast = URI_FUNC(SafeToPointTo);
232 return afterLast;
233 }
234
235 switch (*first) {
236 case _UT('['):
237 {
238 const URI_CHAR * const afterIpLit2
239 = URI_FUNC(ParseIpLit2)(state, first + 1, afterLast);
240 if (afterIpLit2 == NULL) {
241 return NULL;
242 }
243 state->uri->hostText.first = first + 1; /* HOST BEGIN */
244 return URI_FUNC(ParseAuthorityTwo)(state, afterIpLit2, afterLast);
245 }
246
247 case _UT('!'):
248 case _UT('$'):
249 case _UT('%'):
250 case _UT('&'):
251 case _UT('('):
252 case _UT(')'):
253 case _UT('-'):
254 case _UT('*'):
255 case _UT(','):
256 case _UT('.'):
257 case _UT(':'):
258 case _UT(';'):
259 case _UT('@'):
260 case _UT('\''):
261 case _UT('_'):
262 case _UT('~'):
263 case _UT('+'):
264 case _UT('='):
265 case URI_SET_DIGIT:
266 case URI_SET_ALPHA:
267 state->uri->userInfo.first = first; /* USERINFO BEGIN */
268 return URI_FUNC(ParseOwnHostUserInfoNz)(state, first, afterLast);
269
270 default:
271 /* "" regname host */
272 state->uri->hostText.first = URI_FUNC(SafeToPointTo);
273 state->uri->hostText.afterLast = URI_FUNC(SafeToPointTo);
274 return first;
275 }
276}
277
278
279
280/*
281 * [authorityTwo]-><:>[port]
282 * [authorityTwo]-><NULL>
283 */
284static URI_INLINE const URI_CHAR * URI_FUNC(ParseAuthorityTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
285 if (first >= afterLast) {
286 return afterLast;
287 }
288
289 switch (*first) {
290 case _UT(':'):
291 {
292 const URI_CHAR * const afterPort = URI_FUNC(ParsePort)(state, first + 1, afterLast);
293 if (afterPort == NULL) {
294 return NULL;
295 }
296 state->uri->portText.first = first + 1; /* PORT BEGIN */
297 state->uri->portText.afterLast = afterPort; /* PORT END */
298 return afterPort;
299 }
300
301 default:
302 return first;
303 }
304}
305
306
307
308/*
309 * [hexZero]->[HEXDIG][hexZero]
310 * [hexZero]-><NULL>
311 */
312static const URI_CHAR * URI_FUNC(ParseHexZero)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
313 if (first >= afterLast) {
314 return afterLast;
315 }
316
317 switch (*first) {
318 case URI_SET_HEXDIG:
319 return URI_FUNC(ParseHexZero)(state, first + 1, afterLast);
320
321 default:
322 return first;
323 }
324}
325
326
327
328/*
329 * [hierPart]->[pathRootless]
330 * [hierPart]-></>[partHelperTwo]
331 * [hierPart]-><NULL>
332 */
333static URI_INLINE const URI_CHAR * URI_FUNC(ParseHierPart)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
334 if (first >= afterLast) {
335 return afterLast;
336 }
337
338 switch (*first) {
339 case _UT('!'):
340 case _UT('$'):
341 case _UT('%'):
342 case _UT('&'):
343 case _UT('('):
344 case _UT(')'):
345 case _UT('-'):
346 case _UT('*'):
347 case _UT(','):
348 case _UT('.'):
349 case _UT(':'):
350 case _UT(';'):
351 case _UT('@'):
352 case _UT('\''):
353 case _UT('_'):
354 case _UT('~'):
355 case _UT('+'):
356 case _UT('='):
357 case URI_SET_DIGIT:
358 case URI_SET_ALPHA:
359 return URI_FUNC(ParsePathRootless)(state, first, afterLast);
360
361 case _UT('/'):
362 return URI_FUNC(ParsePartHelperTwo)(state, first + 1, afterLast);
363
364 default:
365 return first;
366 }
367}
368
369
370
371/*
372 * [ipFutLoop]->[subDelims][ipFutStopGo]
373 * [ipFutLoop]->[unreserved][ipFutStopGo]
374 * [ipFutLoop]-><:>[ipFutStopGo]
375 */
376static const URI_CHAR * URI_FUNC(ParseIpFutLoop)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
377 if (first >= afterLast) {
378 URI_FUNC(StopSyntax)(state, first);
379 return NULL;
380 }
381
382 switch (*first) {
383 case _UT('!'):
384 case _UT('$'):
385 case _UT('&'):
386 case _UT('('):
387 case _UT(')'):
388 case _UT('-'):
389 case _UT('*'):
390 case _UT(','):
391 case _UT('.'):
392 case _UT(':'):
393 case _UT(';'):
394 case _UT('\''):
395 case _UT('_'):
396 case _UT('~'):
397 case _UT('+'):
398 case _UT('='):
399 case URI_SET_DIGIT:
400 case URI_SET_ALPHA:
401 return URI_FUNC(ParseIpFutStopGo)(state, first + 1, afterLast);
402
403 default:
404 URI_FUNC(StopSyntax)(state, first);
405 return NULL;
406 }
407}
408
409
410
411/*
412 * [ipFutStopGo]->[ipFutLoop]
413 * [ipFutStopGo]-><NULL>
414 */
415static const URI_CHAR * URI_FUNC(ParseIpFutStopGo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
416 if (first >= afterLast) {
417 return afterLast;
418 }
419
420 switch (*first) {
421 case _UT('!'):
422 case _UT('$'):
423 case _UT('&'):
424 case _UT('('):
425 case _UT(')'):
426 case _UT('-'):
427 case _UT('*'):
428 case _UT(','):
429 case _UT('.'):
430 case _UT(':'):
431 case _UT(';'):
432 case _UT('\''):
433 case _UT('_'):
434 case _UT('~'):
435 case _UT('+'):
436 case _UT('='):
437 case URI_SET_DIGIT:
438 case URI_SET_ALPHA:
439 return URI_FUNC(ParseIpFutLoop)(state, first, afterLast);
440
441 default:
442 return first;
443 }
444}
445
446
447
448/*
449 * [ipFuture]-><v>[HEXDIG][hexZero]<.>[ipFutLoop]
450 */
451static const URI_CHAR * URI_FUNC(ParseIpFuture)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
452 if (first >= afterLast) {
453 URI_FUNC(StopSyntax)(state, first);
454 return NULL;
455 }
456
457 /*
458 First character has already been
459 checked before entering this rule.
460
461 switch (*first) {
462 case _UT('v'):
463 */
464 if (first + 1 >= afterLast) {
465 URI_FUNC(StopSyntax)(state, first + 1);
466 return NULL;
467 }
468
469 switch (first[1]) {
470 case URI_SET_HEXDIG:
471 {
472 const URI_CHAR * afterIpFutLoop;
473 const URI_CHAR * const afterHexZero
474 = URI_FUNC(ParseHexZero)(state, first + 2, afterLast);
475 if (afterHexZero == NULL) {
476 return NULL;
477 }
478 if ((afterHexZero >= afterLast)
479 || (*afterHexZero != _UT('.'))) {
480 URI_FUNC(StopSyntax)(state, afterHexZero);
481 return NULL;
482 }
483 state->uri->hostText.first = first; /* HOST BEGIN */
484 state->uri->hostData.ipFuture.first = first; /* IPFUTURE BEGIN */
485 afterIpFutLoop = URI_FUNC(ParseIpFutLoop)(state, afterHexZero + 1, afterLast);
486 if (afterIpFutLoop == NULL) {
487 return NULL;
488 }
489 state->uri->hostText.afterLast = afterIpFutLoop; /* HOST END */
490 state->uri->hostData.ipFuture.afterLast = afterIpFutLoop; /* IPFUTURE END */
491 return afterIpFutLoop;
492 }
493
494 default:
495 URI_FUNC(StopSyntax)(state, first + 1);
496 return NULL;
497 }
498
499 /*
500 default:
501 URI_FUNC(StopSyntax)(state, first);
502 return NULL;
503 }
504 */
505}
506
507
508
509/*
510 * [ipLit2]->[ipFuture]<]>
511 * [ipLit2]->[IPv6address2]
512 */
513static URI_INLINE const URI_CHAR * URI_FUNC(ParseIpLit2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
514 if (first >= afterLast) {
515 URI_FUNC(StopSyntax)(state, first);
516 return NULL;
517 }
518
519 switch (*first) {
520 case _UT('v'):
521 {
522 const URI_CHAR * const afterIpFuture
523 = URI_FUNC(ParseIpFuture)(state, first, afterLast);
524 if (afterIpFuture == NULL) {
525 return NULL;
526 }
527 if ((afterIpFuture >= afterLast)
528 || (*afterIpFuture != _UT(']'))) {
529 URI_FUNC(StopSyntax)(state, first);
530 return NULL;
531 }
532 return afterIpFuture + 1;
533 }
534
535 case _UT(':'):
536 case _UT(']'):
537 case URI_SET_HEXDIG:
538 state->uri->hostData.ip6 = malloc(1 * sizeof(UriIp6)); /* Freed when stopping on parse error */
539 if (state->uri->hostData.ip6 == NULL) {
540 URI_FUNC(StopMalloc)(state);
541 return NULL;
542 }
543 return URI_FUNC(ParseIPv6address2)(state, first, afterLast);
544
545 default:
546 URI_FUNC(StopSyntax)(state, first);
547 return NULL;
548 }
549}
550
551
552
553/*
554 * [IPv6address2]->..<]>
555 */
556static const URI_CHAR * URI_FUNC(ParseIPv6address2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
557 int zipperEver = 0;
558 int quadsDone = 0;
559 int digitCount = 0;
560 unsigned char digitHistory[4];
561 int ip4OctetsDone = 0;
562
563 unsigned char quadsAfterZipper[14];
564 int quadsAfterZipperCount = 0;
565
566
567 for (;;) {
568 if (first >= afterLast) {
569 URI_FUNC(StopSyntax)(state, first);
570 return NULL;
571 }
572
573 /* Inside IPv4 part? */
574 if (ip4OctetsDone > 0) {
575 /* Eat rest of IPv4 address */
576 for (;;) {
577 switch (*first) {
578 case URI_SET_DIGIT:
579 if (digitCount == 4) {
580 URI_FUNC(StopSyntax)(state, first);
581 return NULL;
582 }
583 digitHistory[digitCount++] = (unsigned char)(9 + *first - _UT('9'));
584 break;
585
586 case _UT('.'):
587 if ((ip4OctetsDone == 4) /* NOTE! */
588 || (digitCount == 0)
589 || (digitCount == 4)) {
590 /* Invalid digit or octet count */
591 URI_FUNC(StopSyntax)(state, first);
592 return NULL;
593 } else if ((digitCount > 1)
594 && (digitHistory[0] == 0)) {
595 /* Leading zero */
596 URI_FUNC(StopSyntax)(state, first - digitCount);
597 return NULL;
598 } else if ((digitCount > 2)
599 && (digitHistory[1] == 0)) {
600 /* Leading zero */
601 URI_FUNC(StopSyntax)(state, first - digitCount + 1);
602 return NULL;
603 } else if ((digitCount == 3)
604 && (100 * digitHistory[0]
605 + 10 * digitHistory[1]
606 + digitHistory[2] > 255)) {
607 /* Octet value too large */
608 if (digitHistory[0] > 2) {
609 URI_FUNC(StopSyntax)(state, first - 3);
610 } else if (digitHistory[1] > 5) {
611 URI_FUNC(StopSyntax)(state, first - 2);
612 } else {
613 URI_FUNC(StopSyntax)(state, first - 1);
614 }
615 return NULL;
616 }
617
618 /* Copy IPv4 octet */
619 state->uri->hostData.ip6->data[16 - 4 + ip4OctetsDone] = uriGetOctetValue(digitHistory, digitCount);
620 digitCount = 0;
621 ip4OctetsDone++;
622 break;
623
624 case _UT(']'):
625 if ((ip4OctetsDone != 3) /* NOTE! */
626 || (digitCount == 0)
627 || (digitCount == 4)) {
628 /* Invalid digit or octet count */
629 URI_FUNC(StopSyntax)(state, first);
630 return NULL;
631 } else if ((digitCount > 1)
632 && (digitHistory[0] == 0)) {
633 /* Leading zero */
634 URI_FUNC(StopSyntax)(state, first - digitCount);
635 return NULL;
636 } else if ((digitCount > 2)
637 && (digitHistory[1] == 0)) {
638 /* Leading zero */
639 URI_FUNC(StopSyntax)(state, first - digitCount + 1);
640 return NULL;
641 } else if ((digitCount == 3)
642 && (100 * digitHistory[0]
643 + 10 * digitHistory[1]
644 + digitHistory[2] > 255)) {
645 /* Octet value too large */
646 if (digitHistory[0] > 2) {
647 URI_FUNC(StopSyntax)(state, first - 3);
648 } else if (digitHistory[1] > 5) {
649 URI_FUNC(StopSyntax)(state, first - 2);
650 } else {
651 URI_FUNC(StopSyntax)(state, first - 1);
652 }
653 return NULL;
654 }
655
656 state->uri->hostText.afterLast = first; /* HOST END */
657
658 /* Copy missing quads right before IPv4 */
659 memcpy(state->uri->hostData.ip6->data + 16 - 4 - 2 * quadsAfterZipperCount,
660 quadsAfterZipper, 2 * quadsAfterZipperCount);
661
662 /* Copy last IPv4 octet */
663 state->uri->hostData.ip6->data[16 - 4 + 3] = uriGetOctetValue(digitHistory, digitCount);
664
665 return first + 1;
666
667 default:
668 URI_FUNC(StopSyntax)(state, first);
669 return NULL;
670 }
671 first++;
672 }
673 } else {
674 /* Eat while no dot in sight */
675 int letterAmong = 0;
676 int walking = 1;
677 do {
678 switch (*first) {
679 case URI_SET_HEX_LETTER_LOWER:
680 letterAmong = 1;
681 if (digitCount == 4) {
682 URI_FUNC(StopSyntax)(state, first);
683 return NULL;
684 }
685 digitHistory[digitCount] = (unsigned char)(15 + *first - _UT('f'));
686 digitCount++;
687 break;
688
689 case URI_SET_HEX_LETTER_UPPER:
690 letterAmong = 1;
691 if (digitCount == 4) {
692 URI_FUNC(StopSyntax)(state, first);
693 return NULL;
694 }
695 digitHistory[digitCount] = (unsigned char)(15 + *first - _UT('F'));
696 digitCount++;
697 break;
698
699 case URI_SET_DIGIT:
700 if (digitCount == 4) {
701 URI_FUNC(StopSyntax)(state, first);
702 return NULL;
703 }
704 digitHistory[digitCount] = (unsigned char)(9 + *first - _UT('9'));
705 digitCount++;
706 break;
707
708 case _UT(':'):
709 {
710 int setZipper = 0;
711
712 if (digitCount > 0) {
713 if (zipperEver) {
714 uriWriteQuadToDoubleByte(digitHistory, digitCount, quadsAfterZipper + 2 * quadsAfterZipperCount);
715 quadsAfterZipperCount++;
716 } else {
717 uriWriteQuadToDoubleByte(digitHistory, digitCount, state->uri->hostData.ip6->data + 2 * quadsDone);
718 }
719 quadsDone++;
720 digitCount = 0;
721 }
722 letterAmong = 0;
723
724 /* Too many quads? */
725 if (quadsDone >= 8 - zipperEver) {
726 URI_FUNC(StopSyntax)(state, first);
727 return NULL;
728 }
729
730 /* "::"? */
731 if (first + 1 >= afterLast) {
732 URI_FUNC(StopSyntax)(state, first + 1);
733 return NULL;
734 }
735 if (first[1] == _UT(':')) {
736 const int resetOffset = 2 * (quadsDone + (digitCount > 0));
737
738 first++;
739 if (zipperEver) {
740 URI_FUNC(StopSyntax)(state, first);
741 return NULL; /* "::.+::" */
742 }
743
744 /* Zero everything after zipper */
745 memset(state->uri->hostData.ip6->data + resetOffset, 0, 16 - resetOffset);
746 setZipper = 1;
747
748 /* ":::+"? */
749 if (first + 1 >= afterLast) {
750 URI_FUNC(StopSyntax)(state, first + 1);
751 return NULL; /* No ']' yet */
752 }
753 if (first[1] == _UT(':')) {
754 URI_FUNC(StopSyntax)(state, first + 1);
755 return NULL; /* ":::+ "*/
756 }
757 }
758
759 if (setZipper) {
760 zipperEver = 1;
761 }
762 }
763 break;
764
765 case _UT('.'):
766 if ((quadsDone > 6) /* NOTE */
767 || (!zipperEver && (quadsDone < 6))
768 || letterAmong
769 || (digitCount == 0)
770 || (digitCount == 4)) {
771 /* Invalid octet before */
772 URI_FUNC(StopSyntax)(state, first);
773 return NULL;
774 } else if ((digitCount > 1)
775 && (digitHistory[0] == 0)) {
776 /* Leading zero */
777 URI_FUNC(StopSyntax)(state, first - digitCount);
778 return NULL;
779 } else if ((digitCount > 2)
780 && (digitHistory[1] == 0)) {
781 /* Leading zero */
782 URI_FUNC(StopSyntax)(state, first - digitCount + 1);
783 return NULL;
784 } else if ((digitCount == 3)
785 && (100 * digitHistory[0]
786 + 10 * digitHistory[1]
787 + digitHistory[2] > 255)) {
788 /* Octet value too large */
789 if (digitHistory[0] > 2) {
790 URI_FUNC(StopSyntax)(state, first - 3);
791 } else if (digitHistory[1] > 5) {
792 URI_FUNC(StopSyntax)(state, first - 2);
793 } else {
794 URI_FUNC(StopSyntax)(state, first - 1);
795 }
796 return NULL;
797 }
798
799 /* Copy first IPv4 octet */
800 state->uri->hostData.ip6->data[16 - 4] = uriGetOctetValue(digitHistory, digitCount);
801 digitCount = 0;
802
803 /* Switch over to IPv4 loop */
804 ip4OctetsDone = 1;
805 walking = 0;
806 break;
807
808 case _UT(']'):
809 /* Too little quads? */
810 if (!zipperEver && !((quadsDone == 7) && (digitCount > 0))) {
811 URI_FUNC(StopSyntax)(state, first);
812 return NULL;
813 }
814
815 if (digitCount > 0) {
816 if (zipperEver) {
817 uriWriteQuadToDoubleByte(digitHistory, digitCount, quadsAfterZipper + 2 * quadsAfterZipperCount);
818 quadsAfterZipperCount++;
819 } else {
820 uriWriteQuadToDoubleByte(digitHistory, digitCount, state->uri->hostData.ip6->data + 2 * quadsDone);
821 }
822 /*
823 quadsDone++;
824 digitCount = 0;
825 */
826 }
827
828 /* Copy missing quads to the end */
829 memcpy(state->uri->hostData.ip6->data + 16 - 2 * quadsAfterZipperCount,
830 quadsAfterZipper, 2 * quadsAfterZipperCount);
831
832 state->uri->hostText.afterLast = first; /* HOST END */
833 return first + 1; /* Fine */
834
835 default:
836 URI_FUNC(StopSyntax)(state, first);
837 return NULL;
838 }
839 first++;
840
841 if (first >= afterLast) {
842 URI_FUNC(StopSyntax)(state, first);
843 return NULL; /* No ']' yet */
844 }
845 } while (walking);
846 }
847 }
848}
849
850
851
852/*
853 * [mustBeSegmentNzNc]->[pctEncoded][mustBeSegmentNzNc]
854 * [mustBeSegmentNzNc]->[subDelims][mustBeSegmentNzNc]
855 * [mustBeSegmentNzNc]->[unreserved][mustBeSegmentNzNc]
856 * [mustBeSegmentNzNc]->[uriTail] // can take <NULL>
857 * [mustBeSegmentNzNc]-></>[segment][zeroMoreSlashSegs][uriTail]
858 * [mustBeSegmentNzNc]-><@>[mustBeSegmentNzNc]
859 */
860static const URI_CHAR * URI_FUNC(ParseMustBeSegmentNzNc)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
861 if (first >= afterLast) {
862 if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first)) { /* SEGMENT BOTH */
863 URI_FUNC(StopMalloc)(state);
864 return NULL;
865 }
866 state->uri->scheme.first = NULL; /* Not a scheme, reset */
867 return afterLast;
868 }
869
870 switch (*first) {
871 case _UT('%'):
872 {
873 const URI_CHAR * const afterPctEncoded
874 = URI_FUNC(ParsePctEncoded)(state, first, afterLast);
875 if (afterPctEncoded == NULL) {
876 return NULL;
877 }
878 return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast);
879 }
880
881 case _UT('@'):
882 case _UT('!'):
883 case _UT('$'):
884 case _UT('&'):
885 case _UT('('):
886 case _UT(')'):
887 case _UT('*'):
888 case _UT(','):
889 case _UT(';'):
890 case _UT('\''):
891 case _UT('+'):
892 case _UT('='):
893 case _UT('-'):
894 case _UT('.'):
895 case _UT('_'):
896 case _UT('~'):
897 case URI_SET_DIGIT:
898 case URI_SET_ALPHA:
899 return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast);
900
901 case _UT('/'):
902 {
903 const URI_CHAR * afterZeroMoreSlashSegs;
904 const URI_CHAR * afterSegment;
905 if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first)) { /* SEGMENT BOTH */
906 URI_FUNC(StopMalloc)(state);
907 return NULL;
908 }
909 state->uri->scheme.first = NULL; /* Not a scheme, reset */
910 afterSegment = URI_FUNC(ParseSegment)(state, first + 1, afterLast);
911 if (afterSegment == NULL) {
912 return NULL;
913 }
914 if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment)) { /* SEGMENT BOTH */
915 URI_FUNC(StopMalloc)(state);
916 return NULL;
917 }
918 afterZeroMoreSlashSegs
919 = URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast);
920 if (afterZeroMoreSlashSegs == NULL) {
921 return NULL;
922 }
923 return URI_FUNC(ParseUriTail)(state, afterZeroMoreSlashSegs, afterLast);
924 }
925
926 default:
927 if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first)) { /* SEGMENT BOTH */
928 URI_FUNC(StopMalloc)(state);
929 return NULL;
930 }
931 state->uri->scheme.first = NULL; /* Not a scheme, reset */
932 return URI_FUNC(ParseUriTail)(state, first, afterLast);
933 }
934}
935
936
937
938/*
939 * [ownHost]-><[>[ipLit2][authorityTwo]
940 * [ownHost]->[ownHost2] // can take <NULL>
941 */
942static URI_INLINE const URI_CHAR * URI_FUNC(ParseOwnHost)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
943 if (first >= afterLast) {
944 return afterLast;
945 }
946
947 switch (*first) {
948 case _UT('['):
949 {
950 const URI_CHAR * const afterIpLit2
951 = URI_FUNC(ParseIpLit2)(state, first + 1, afterLast);
952 if (afterIpLit2 == NULL) {
953 return NULL;
954 }
955 state->uri->hostText.first = first + 1; /* HOST BEGIN */
956 return URI_FUNC(ParseAuthorityTwo)(state, afterIpLit2, afterLast);
957 }
958
959 default:
960 return URI_FUNC(ParseOwnHost2)(state, first, afterLast);
961 }
962}
963
964
965
966static URI_INLINE UriBool URI_FUNC(OnExitOwnHost2)(URI_TYPE(ParserState) * state, const URI_CHAR * first) {
967 state->uri->hostText.afterLast = first; /* HOST END */
968
969 /* Valid IPv4 or just a regname? */
970 state->uri->hostData.ip4 = malloc(1 * sizeof(UriIp4)); /* Freed when stopping on parse error */
971 if (state->uri->hostData.ip4 == NULL) {
972 return URI_FALSE; /* Raises malloc error */
973 }
974 if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data,
975 state->uri->hostText.first, state->uri->hostText.afterLast)) {
976 /* Not IPv4 */
977 free(state->uri->hostData.ip4);
978 state->uri->hostData.ip4 = NULL;
979 }
980 return URI_TRUE; /* Success */
981}
982
983
984
985/*
986 * [ownHost2]->[authorityTwo] // can take <NULL>
987 * [ownHost2]->[pctSubUnres][ownHost2]
988 */
989static const URI_CHAR * URI_FUNC(ParseOwnHost2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
990 if (first >= afterLast) {
991 if (!URI_FUNC(OnExitOwnHost2)(state, first)) {
992 URI_FUNC(StopMalloc)(state);
993 return NULL;
994 }
995 return afterLast;
996 }
997
998 switch (*first) {
999 case _UT('!'):
1000 case _UT('$'):
1001 case _UT('%'):
1002 case _UT('&'):
1003 case _UT('('):
1004 case _UT(')'):
1005 case _UT('-'):
1006 case _UT('*'):
1007 case _UT(','):
1008 case _UT('.'):
1009 case _UT(';'):
1010 case _UT('\''):
1011 case _UT('_'):
1012 case _UT('~'):
1013 case _UT('+'):
1014 case _UT('='):
1015 case URI_SET_DIGIT:
1016 case URI_SET_ALPHA:
1017 {
1018 const URI_CHAR * const afterPctSubUnres
1019 = URI_FUNC(ParsePctSubUnres)(state, first, afterLast);
1020 if (afterPctSubUnres == NULL) {
1021 return NULL;
1022 }
1023 return URI_FUNC(ParseOwnHost2)(state, afterPctSubUnres, afterLast);
1024 }
1025
1026 default:
1027 if (!URI_FUNC(OnExitOwnHost2)(state, first)) {
1028 URI_FUNC(StopMalloc)(state);
1029 return NULL;
1030 }
1031 return URI_FUNC(ParseAuthorityTwo)(state, first, afterLast);
1032 }
1033}
1034
1035
1036
1037static URI_INLINE UriBool URI_FUNC(OnExitOwnHostUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first) {
1038 state->uri->hostText.first = state->uri->userInfo.first; /* Host instead of userInfo, update */
1039 state->uri->userInfo.first = NULL; /* Not a userInfo, reset */
1040 state->uri->hostText.afterLast = first; /* HOST END */
1041
1042 /* Valid IPv4 or just a regname? */
1043 state->uri->hostData.ip4 = malloc(1 * sizeof(UriIp4)); /* Freed when stopping on parse error */
1044 if (state->uri->hostData.ip4 == NULL) {
1045 return URI_FALSE; /* Raises malloc error */
1046 }
1047 if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data,
1048 state->uri->hostText.first, state->uri->hostText.afterLast)) {
1049 /* Not IPv4 */
1050 free(state->uri->hostData.ip4);
1051 state->uri->hostData.ip4 = NULL;
1052 }
1053 return URI_TRUE; /* Success */
1054}
1055
1056
1057
1058/*
1059 * [ownHostUserInfo]->[ownHostUserInfoNz]
1060 * [ownHostUserInfo]-><NULL>
1061 */
1062static URI_INLINE const URI_CHAR * URI_FUNC(ParseOwnHostUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1063 if (first >= afterLast) {
1064 if (!URI_FUNC(OnExitOwnHostUserInfo)(state, first)) {
1065 URI_FUNC(StopMalloc)(state);
1066 return NULL;
1067 }
1068 return afterLast;
1069 }
1070
1071 switch (*first) {
1072 case _UT('!'):
1073 case _UT('$'):
1074 case _UT('%'):
1075 case _UT('&'):
1076 case _UT('('):
1077 case _UT(')'):
1078 case _UT('-'):
1079 case _UT('*'):
1080 case _UT(','):
1081 case _UT('.'):
1082 case _UT(':'):
1083 case _UT(';'):
1084 case _UT('@'):
1085 case _UT('\''):
1086 case _UT('_'):
1087 case _UT('~'):
1088 case _UT('+'):
1089 case _UT('='):
1090 case URI_SET_DIGIT:
1091 case URI_SET_ALPHA:
1092 return URI_FUNC(ParseOwnHostUserInfoNz)(state, first, afterLast);
1093
1094 default:
1095 if (!URI_FUNC(OnExitOwnHostUserInfo)(state, first)) {
1096 URI_FUNC(StopMalloc)(state);
1097 return NULL;
1098 }
1099 return first;
1100 }
1101}
1102
1103
1104
1105/*
1106 * [ownHostUserInfoNz]->[pctSubUnres][ownHostUserInfo]
1107 * [ownHostUserInfoNz]-><:>[ownPortUserInfo]
1108 * [ownHostUserInfoNz]-><@>[ownHost]
1109 */
1110static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfoNz)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1111 if (first >= afterLast) {
1112 URI_FUNC(StopSyntax)(state, first);
1113 return NULL;
1114 }
1115
1116 switch (*first) {
1117 case _UT('!'):
1118 case _UT('$'):
1119 case _UT('%'):
1120 case _UT('&'):
1121 case _UT('('):
1122 case _UT(')'):
1123 case _UT('-'):
1124 case _UT('*'):
1125 case _UT(','):
1126 case _UT('.'):
1127 case _UT(';'):
1128 case _UT('\''):
1129 case _UT('_'):
1130 case _UT('~'):
1131 case _UT('+'):
1132 case _UT('='):
1133 case URI_SET_DIGIT:
1134 case URI_SET_ALPHA:
1135 {
1136 const URI_CHAR * const afterPctSubUnres
1137 = URI_FUNC(ParsePctSubUnres)(state, first, afterLast);
1138 if (afterPctSubUnres == NULL) {
1139 return NULL;
1140 }
1141 return URI_FUNC(ParseOwnHostUserInfo)(state, afterPctSubUnres, afterLast);
1142 }
1143
1144 case _UT(':'):
1145 state->uri->hostText.afterLast = first; /* HOST END */
1146 state->uri->portText.first = first + 1; /* PORT BEGIN */
1147 return URI_FUNC(ParseOwnPortUserInfo)(state, first + 1, afterLast);
1148
1149 case _UT('@'):
1150 state->uri->userInfo.afterLast = first; /* USERINFO END */
1151 state->uri->hostText.first = first + 1; /* HOST BEGIN */
1152 return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast);
1153
1154 default:
1155 URI_FUNC(StopSyntax)(state, first);
1156 return NULL;
1157 }
1158}
1159
1160
1161
1162static URI_INLINE UriBool URI_FUNC(OnExitOwnPortUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first) {
1163 state->uri->hostText.first = state->uri->userInfo.first; /* Host instead of userInfo, update */
1164 state->uri->userInfo.first = NULL; /* Not a userInfo, reset */
1165 state->uri->portText.afterLast = first; /* PORT END */
1166
1167 /* Valid IPv4 or just a regname? */
1168 state->uri->hostData.ip4 = malloc(1 * sizeof(UriIp4)); /* Freed when stopping on parse error */
1169 if (state->uri->hostData.ip4 == NULL) {
1170 return URI_FALSE; /* Raises malloc error */
1171 }
1172 if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data,
1173 state->uri->hostText.first, state->uri->hostText.afterLast)) {
1174 /* Not IPv4 */
1175 free(state->uri->hostData.ip4);
1176 state->uri->hostData.ip4 = NULL;
1177 }
1178 return URI_TRUE; /* Success */
1179}
1180
1181
1182
1183/*
1184 * [ownPortUserInfo]->[ALPHA][ownUserInfo]
1185 * [ownPortUserInfo]->[DIGIT][ownPortUserInfo]
1186 * [ownPortUserInfo]-><.>[ownUserInfo]
1187 * [ownPortUserInfo]-><_>[ownUserInfo]
1188 * [ownPortUserInfo]-><~>[ownUserInfo]
1189 * [ownPortUserInfo]-><->[ownUserInfo]
1190 * [ownPortUserInfo]->[subDelims][ownUserInfo]
1191 * [ownPortUserInfo]->[pctEncoded][ownUserInfo]
1192 * [ownPortUserInfo]-><:>[ownUserInfo]
1193 * [ownPortUserInfo]-><@>[ownHost]
1194 * [ownPortUserInfo]-><NULL>
1195 */
1196static const URI_CHAR * URI_FUNC(ParseOwnPortUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1197 if (first >= afterLast) {
1198 if (!URI_FUNC(OnExitOwnPortUserInfo)(state, first)) {
1199 URI_FUNC(StopMalloc)(state);
1200 return NULL;
1201 }
1202 return afterLast;
1203 }
1204
1205 switch (*first) {
1206 /* begin sub-delims */
1207 case _UT('!'):
1208 case _UT('$'):
1209 case _UT('&'):
1210 case _UT('\''):
1211 case _UT('('):
1212 case _UT(')'):
1213 case _UT('*'):
1214 case _UT('+'):
1215 case _UT(','):
1216 case _UT(';'):
1217 case _UT('='):
1218 /* end sub-delims */
1219 /* begin unreserved (except alpha and digit) */
1220 case _UT('-'):
1221 case _UT('.'):
1222 case _UT('_'):
1223 case _UT('~'):
1224 /* end unreserved (except alpha and digit) */
1225 case _UT(':'):
1226 case URI_SET_ALPHA:
1227 state->uri->hostText.afterLast = NULL; /* Not a host, reset */
1228 state->uri->portText.first = NULL; /* Not a port, reset */
1229 return URI_FUNC(ParseOwnUserInfo)(state, first + 1, afterLast);
1230
1231 case URI_SET_DIGIT:
1232 return URI_FUNC(ParseOwnPortUserInfo)(state, first + 1, afterLast);
1233
1234 case _UT('%'):
1235 state->uri->portText.first = NULL; /* Not a port, reset */
1236 {
1237 const URI_CHAR * const afterPct
1238 = URI_FUNC(ParsePctEncoded)(state, first, afterLast);
1239 if (afterPct == NULL) {
1240 return NULL;
1241 }
1242 return URI_FUNC(ParseOwnUserInfo)(state, afterPct, afterLast);
1243 }
1244
1245 case _UT('@'):
1246 state->uri->hostText.afterLast = NULL; /* Not a host, reset */
1247 state->uri->portText.first = NULL; /* Not a port, reset */
1248 state->uri->userInfo.afterLast = first; /* USERINFO END */
1249 state->uri->hostText.first = first + 1; /* HOST BEGIN */
1250 return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast);
1251
1252 default:
1253 if (!URI_FUNC(OnExitOwnPortUserInfo)(state, first)) {
1254 URI_FUNC(StopMalloc)(state);
1255 return NULL;
1256 }
1257 return first;
1258 }
1259}
1260
1261
1262
1263/*
1264 * [ownUserInfo]->[pctSubUnres][ownUserInfo]
1265 * [ownUserInfo]-><:>[ownUserInfo]
1266 * [ownUserInfo]-><@>[ownHost]
1267 */
1268static const URI_CHAR * URI_FUNC(ParseOwnUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1269 if (first >= afterLast) {
1270 URI_FUNC(StopSyntax)(state, first);
1271 return NULL;
1272 }
1273
1274 switch (*first) {
1275 case _UT('!'):
1276 case _UT('$'):
1277 case _UT('%'):
1278 case _UT('&'):
1279 case _UT('('):
1280 case _UT(')'):
1281 case _UT('-'):
1282 case _UT('*'):
1283 case _UT(','):
1284 case _UT('.'):
1285 case _UT(';'):
1286 case _UT('\''):
1287 case _UT('_'):
1288 case _UT('~'):
1289 case _UT('+'):
1290 case _UT('='):
1291 case URI_SET_DIGIT:
1292 case URI_SET_ALPHA:
1293 {
1294 const URI_CHAR * const afterPctSubUnres
1295 = URI_FUNC(ParsePctSubUnres)(state, first, afterLast);
1296 if (afterPctSubUnres == NULL) {
1297 return NULL;
1298 }
1299 return URI_FUNC(ParseOwnUserInfo)(state, afterPctSubUnres, afterLast);
1300 }
1301
1302 case _UT(':'):
1303 return URI_FUNC(ParseOwnUserInfo)(state, first + 1, afterLast);
1304
1305 case _UT('@'):
1306 /* SURE */
1307 state->uri->userInfo.afterLast = first; /* USERINFO END */
1308 state->uri->hostText.first = first + 1; /* HOST BEGIN */
1309 return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast);
1310
1311 default:
1312 URI_FUNC(StopSyntax)(state, first);
1313 return NULL;
1314 }
1315}
1316
1317
1318
1319static URI_INLINE void URI_FUNC(OnExitPartHelperTwo)(URI_TYPE(ParserState) * state) {
1320 state->uri->absolutePath = URI_TRUE;
1321}
1322
1323
1324
1325/*
1326 * [partHelperTwo]->[pathAbsNoLeadSlash] // can take <NULL>
1327 * [partHelperTwo]-></>[authority][pathAbsEmpty]
1328 */
1329static URI_INLINE const URI_CHAR * URI_FUNC(ParsePartHelperTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1330 if (first >= afterLast) {
1331 URI_FUNC(OnExitPartHelperTwo)(state);
1332 return afterLast;
1333 }
1334
1335 switch (*first) {
1336 case _UT('/'):
1337 {
1338 const URI_CHAR * const afterAuthority
1339 = URI_FUNC(ParseAuthority)(state, first + 1, afterLast);
1340 const URI_CHAR * afterPathAbsEmpty;
1341 if (afterAuthority == NULL) {
1342 return NULL;
1343 }
1344 afterPathAbsEmpty = URI_FUNC(ParsePathAbsEmpty)(state, afterAuthority, afterLast);
1345
1346 URI_FUNC(FixEmptyTrailSegment)(state->uri);
1347
1348 return afterPathAbsEmpty;
1349 }
1350
1351 default:
1352 URI_FUNC(OnExitPartHelperTwo)(state);
1353 return URI_FUNC(ParsePathAbsNoLeadSlash)(state, first, afterLast);
1354 }
1355}
1356
1357
1358
1359/*
1360 * [pathAbsEmpty]-></>[segment][pathAbsEmpty]
1361 * [pathAbsEmpty]-><NULL>
1362 */
1363static const URI_CHAR * URI_FUNC(ParsePathAbsEmpty)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1364 if (first >= afterLast) {
1365 return afterLast;
1366 }
1367
1368 switch (*first) {
1369 case _UT('/'):
1370 {
1371 const URI_CHAR * const afterSegment
1372 = URI_FUNC(ParseSegment)(state, first + 1, afterLast);
1373 if (afterSegment == NULL) {
1374 return NULL;
1375 }
1376 if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment)) { /* SEGMENT BOTH */
1377 URI_FUNC(StopMalloc)(state);
1378 return NULL;
1379 }
1380 return URI_FUNC(ParsePathAbsEmpty)(state, afterSegment, afterLast);
1381 }
1382
1383 default:
1384 return first;
1385 }
1386}
1387
1388
1389
1390/*
1391 * [pathAbsNoLeadSlash]->[segmentNz][zeroMoreSlashSegs]
1392 * [pathAbsNoLeadSlash]-><NULL>
1393 */
1394static URI_INLINE const URI_CHAR * URI_FUNC(ParsePathAbsNoLeadSlash)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1395 if (first >= afterLast) {
1396 return afterLast;
1397 }
1398
1399 switch (*first) {
1400 case _UT('!'):
1401 case _UT('$'):
1402 case _UT('%'):
1403 case _UT('&'):
1404 case _UT('('):
1405 case _UT(')'):
1406 case _UT('-'):
1407 case _UT('*'):
1408 case _UT(','):
1409 case _UT('.'):
1410 case _UT(':'):
1411 case _UT(';'):
1412 case _UT('@'):
1413 case _UT('\''):
1414 case _UT('_'):
1415 case _UT('~'):
1416 case _UT('+'):
1417 case _UT('='):
1418 case URI_SET_DIGIT:
1419 case URI_SET_ALPHA:
1420 {
1421 const URI_CHAR * const afterSegmentNz
1422 = URI_FUNC(ParseSegmentNz)(state, first, afterLast);
1423 if (afterSegmentNz == NULL) {
1424 return NULL;
1425 }
1426 if (!URI_FUNC(PushPathSegment)(state, first, afterSegmentNz)) { /* SEGMENT BOTH */
1427 URI_FUNC(StopMalloc)(state);
1428 return NULL;
1429 }
1430 return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegmentNz, afterLast);
1431 }
1432
1433 default:
1434 return first;
1435 }
1436}
1437
1438
1439
1440/*
1441 * [pathRootless]->[segmentNz][zeroMoreSlashSegs]
1442 */
1443static URI_INLINE const URI_CHAR * URI_FUNC(ParsePathRootless)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1444 const URI_CHAR * const afterSegmentNz
1445 = URI_FUNC(ParseSegmentNz)(state, first, afterLast);
1446 if (afterSegmentNz == NULL) {
1447 return NULL;
1448 } else {
1449 if (!URI_FUNC(PushPathSegment)(state, first, afterSegmentNz)) { /* SEGMENT BOTH */
1450 URI_FUNC(StopMalloc)(state);
1451 return NULL;
1452 }
1453 }
1454 return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegmentNz, afterLast);
1455}
1456
1457
1458
1459/*
1460 * [pchar]->[pctEncoded]
1461 * [pchar]->[subDelims]
1462 * [pchar]->[unreserved]
1463 * [pchar]-><:>
1464 * [pchar]-><@>
1465 */
1466static const URI_CHAR * URI_FUNC(ParsePchar)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1467 if (first >= afterLast) {
1468 URI_FUNC(StopSyntax)(state, first);
1469 return NULL;
1470 }
1471
1472 switch (*first) {
1473 case _UT('%'):
1474 return URI_FUNC(ParsePctEncoded)(state, first, afterLast);
1475
1476 case _UT(':'):
1477 case _UT('@'):
1478 case _UT('!'):
1479 case _UT('$'):
1480 case _UT('&'):
1481 case _UT('('):
1482 case _UT(')'):
1483 case _UT('*'):
1484 case _UT(','):
1485 case _UT(';'):
1486 case _UT('\''):
1487 case _UT('+'):
1488 case _UT('='):
1489 case _UT('-'):
1490 case _UT('.'):
1491 case _UT('_'):
1492 case _UT('~'):
1493 case URI_SET_DIGIT:
1494 case URI_SET_ALPHA:
1495 return first + 1;
1496
1497 default:
1498 URI_FUNC(StopSyntax)(state, first);
1499 return NULL;
1500 }
1501}
1502
1503
1504
1505/*
1506 * [pctEncoded]-><%>[HEXDIG][HEXDIG]
1507 */
1508static const URI_CHAR * URI_FUNC(ParsePctEncoded)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1509 if (first >= afterLast) {
1510 URI_FUNC(StopSyntax)(state, first);
1511 return NULL;
1512 }
1513
1514 /*
1515 First character has already been
1516 checked before entering this rule.
1517
1518 switch (*first) {
1519 case _UT('%'):
1520 */
1521 if (first + 1 >= afterLast) {
1522 URI_FUNC(StopSyntax)(state, first + 1);
1523 return NULL;
1524 }
1525
1526 switch (first[1]) {
1527 case URI_SET_HEXDIG:
1528 if (first + 2 >= afterLast) {
1529 URI_FUNC(StopSyntax)(state, first + 2);
1530 return NULL;
1531 }
1532
1533 switch (first[2]) {
1534 case URI_SET_HEXDIG:
1535 return first + 3;
1536
1537 default:
1538 URI_FUNC(StopSyntax)(state, first + 2);
1539 return NULL;
1540 }
1541
1542 default:
1543 URI_FUNC(StopSyntax)(state, first + 1);
1544 return NULL;
1545 }
1546
1547 /*
1548 default:
1549 URI_FUNC(StopSyntax)(state, first);
1550 return NULL;
1551 }
1552 */
1553}
1554
1555
1556
1557/*
1558 * [pctSubUnres]->[pctEncoded]
1559 * [pctSubUnres]->[subDelims]
1560 * [pctSubUnres]->[unreserved]
1561 */
1562static const URI_CHAR * URI_FUNC(ParsePctSubUnres)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1563 if (first >= afterLast) {
1564 URI_FUNC(StopSyntax)(state, first);
1565 return NULL;
1566 }
1567
1568 switch (*first) {
1569 case _UT('%'):
1570 return URI_FUNC(ParsePctEncoded)(state, first, afterLast);
1571
1572 case _UT('!'):
1573 case _UT('$'):
1574 case _UT('&'):
1575 case _UT('('):
1576 case _UT(')'):
1577 case _UT('*'):
1578 case _UT(','):
1579 case _UT(';'):
1580 case _UT('\''):
1581 case _UT('+'):
1582 case _UT('='):
1583 case _UT('-'):
1584 case _UT('.'):
1585 case _UT('_'):
1586 case _UT('~'):
1587 case URI_SET_DIGIT:
1588 case URI_SET_ALPHA:
1589 return first + 1;
1590
1591 default:
1592 URI_FUNC(StopSyntax)(state, first);
1593 return NULL;
1594 }
1595}
1596
1597
1598
1599/*
1600 * [port]->[DIGIT][port]
1601 * [port]-><NULL>
1602 */
1603static const URI_CHAR * URI_FUNC(ParsePort)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1604 if (first >= afterLast) {
1605 return afterLast;
1606 }
1607
1608 switch (*first) {
1609 case URI_SET_DIGIT:
1610 return URI_FUNC(ParsePort)(state, first + 1, afterLast);
1611
1612 default:
1613 return first;
1614 }
1615}
1616
1617
1618
1619/*
1620 * [queryFrag]->[pchar][queryFrag]
1621 * [queryFrag]-></>[queryFrag]
1622 * [queryFrag]-><?>[queryFrag]
1623 * [queryFrag]-><NULL>
1624 */
1625static const URI_CHAR * URI_FUNC(ParseQueryFrag)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1626 if (first >= afterLast) {
1627 return afterLast;
1628 }
1629
1630 switch (*first) {
1631 case _UT('!'):
1632 case _UT('$'):
1633 case _UT('%'):
1634 case _UT('&'):
1635 case _UT('('):
1636 case _UT(')'):
1637 case _UT('-'):
1638 case _UT('*'):
1639 case _UT(','):
1640 case _UT('.'):
1641 case _UT(':'):
1642 case _UT(';'):
1643 case _UT('@'):
1644 case _UT('\''):
1645 case _UT('_'):
1646 case _UT('~'):
1647 case _UT('+'):
1648 case _UT('='):
1649 case URI_SET_DIGIT:
1650 case URI_SET_ALPHA:
1651 {
1652 const URI_CHAR * const afterPchar
1653 = URI_FUNC(ParsePchar)(state, first, afterLast);
1654 if (afterPchar == NULL) {
1655 return NULL;
1656 }
1657 return URI_FUNC(ParseQueryFrag)(state, afterPchar, afterLast);
1658 }
1659
1660 case _UT('/'):
1661 case _UT('?'):
1662 return URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast);
1663
1664 default:
1665 return first;
1666 }
1667}
1668
1669
1670
1671/*
1672 * [segment]->[pchar][segment]
1673 * [segment]-><NULL>
1674 */
1675static const URI_CHAR * URI_FUNC(ParseSegment)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1676 if (first >= afterLast) {
1677 return afterLast;
1678 }
1679
1680 switch (*first) {
1681 case _UT('!'):
1682 case _UT('$'):
1683 case _UT('%'):
1684 case _UT('&'):
1685 case _UT('('):
1686 case _UT(')'):
1687 case _UT('-'):
1688 case _UT('*'):
1689 case _UT(','):
1690 case _UT('.'):
1691 case _UT(':'):
1692 case _UT(';'):
1693 case _UT('@'):
1694 case _UT('\''):
1695 case _UT('_'):
1696 case _UT('~'):
1697 case _UT('+'):
1698 case _UT('='):
1699 case URI_SET_DIGIT:
1700 case URI_SET_ALPHA:
1701 {
1702 const URI_CHAR * const afterPchar
1703 = URI_FUNC(ParsePchar)(state, first, afterLast);
1704 if (afterPchar == NULL) {
1705 return NULL;
1706 }
1707 return URI_FUNC(ParseSegment)(state, afterPchar, afterLast);
1708 }
1709
1710 default:
1711 return first;
1712 }
1713}
1714
1715
1716
1717/*
1718 * [segmentNz]->[pchar][segment]
1719 */
1720static URI_INLINE const URI_CHAR * URI_FUNC(ParseSegmentNz)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1721 const URI_CHAR * const afterPchar
1722 = URI_FUNC(ParsePchar)(state, first, afterLast);
1723 if (afterPchar == NULL) {
1724 return NULL;
1725 }
1726 return URI_FUNC(ParseSegment)(state, afterPchar, afterLast);
1727}
1728
1729
1730
1731static URI_INLINE UriBool URI_FUNC(OnExitSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, const URI_CHAR * first) {
1732 if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first)) { /* SEGMENT BOTH */
1733 return URI_FALSE; /* Raises malloc error*/
1734 }
1735 state->uri->scheme.first = NULL; /* Not a scheme, reset */
1736 return URI_TRUE; /* Success */
1737}
1738
1739
1740
1741/*
1742 * [segmentNzNcOrScheme2]->[ALPHA][segmentNzNcOrScheme2]
1743 * [segmentNzNcOrScheme2]->[DIGIT][segmentNzNcOrScheme2]
1744 * [segmentNzNcOrScheme2]->[pctEncoded][mustBeSegmentNzNc]
1745 * [segmentNzNcOrScheme2]->[uriTail] // can take <NULL>
1746 * [segmentNzNcOrScheme2]-><!>[mustBeSegmentNzNc]
1747 * [segmentNzNcOrScheme2]-><$>[mustBeSegmentNzNc]
1748 * [segmentNzNcOrScheme2]-><&>[mustBeSegmentNzNc]
1749 * [segmentNzNcOrScheme2]-><(>[mustBeSegmentNzNc]
1750 * [segmentNzNcOrScheme2]-><)>[mustBeSegmentNzNc]
1751 * [segmentNzNcOrScheme2]-><*>[mustBeSegmentNzNc]
1752 * [segmentNzNcOrScheme2]-><,>[mustBeSegmentNzNc]
1753 * [segmentNzNcOrScheme2]-><.>[segmentNzNcOrScheme2]
1754 * [segmentNzNcOrScheme2]-></>[segment][zeroMoreSlashSegs][uriTail]
1755 * [segmentNzNcOrScheme2]-><:>[hierPart][uriTail]
1756 * [segmentNzNcOrScheme2]-><;>[mustBeSegmentNzNc]
1757 * [segmentNzNcOrScheme2]-><@>[mustBeSegmentNzNc]
1758 * [segmentNzNcOrScheme2]-><_>[mustBeSegmentNzNc]
1759 * [segmentNzNcOrScheme2]-><~>[mustBeSegmentNzNc]
1760 * [segmentNzNcOrScheme2]-><+>[segmentNzNcOrScheme2]
1761 * [segmentNzNcOrScheme2]-><=>[mustBeSegmentNzNc]
1762 * [segmentNzNcOrScheme2]-><'>[mustBeSegmentNzNc]
1763 * [segmentNzNcOrScheme2]-><->[segmentNzNcOrScheme2]
1764 */
1765static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1766 if (first >= afterLast) {
1767 if (!URI_FUNC(OnExitSegmentNzNcOrScheme2)(state, first)) {
1768 URI_FUNC(StopMalloc)(state);
1769 return NULL;
1770 }
1771 return afterLast;
1772 }
1773
1774 switch (*first) {
1775 case _UT('.'):
1776 case _UT('+'):
1777 case _UT('-'):
1778 case URI_SET_ALPHA:
1779 case URI_SET_DIGIT:
1780 return URI_FUNC(ParseSegmentNzNcOrScheme2)(state, first + 1, afterLast);
1781
1782 case _UT('%'):
1783 {
1784 const URI_CHAR * const afterPctEncoded
1785 = URI_FUNC(ParsePctEncoded)(state, first, afterLast);
1786 if (afterPctEncoded == NULL) {
1787 return NULL;
1788 }
1789 return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast);
1790 }
1791
1792 case _UT('!'):
1793 case _UT('$'):
1794 case _UT('&'):
1795 case _UT('('):
1796 case _UT(')'):
1797 case _UT('*'):
1798 case _UT(','):
1799 case _UT(';'):
1800 case _UT('@'):
1801 case _UT('_'):
1802 case _UT('~'):
1803 case _UT('='):
1804 case _UT('\''):
1805 return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast);
1806
1807 case _UT('/'):
1808 {
1809 const URI_CHAR * afterZeroMoreSlashSegs;
1810 const URI_CHAR * const afterSegment
1811 = URI_FUNC(ParseSegment)(state, first + 1, afterLast);
1812 if (afterSegment == NULL) {
1813 return NULL;
1814 }
1815 if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first)) { /* SEGMENT BOTH */
1816 URI_FUNC(StopMalloc)(state);
1817 return NULL;
1818 }
1819 state->uri->scheme.first = NULL; /* Not a scheme, reset */
1820 if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment)) { /* SEGMENT BOTH */
1821 URI_FUNC(StopMalloc)(state);
1822 return NULL;
1823 }
1824 afterZeroMoreSlashSegs
1825 = URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast);
1826 if (afterZeroMoreSlashSegs == NULL) {
1827 return NULL;
1828 }
1829 return URI_FUNC(ParseUriTail)(state, afterZeroMoreSlashSegs, afterLast);
1830 }
1831
1832 case _UT(':'):
1833 {
1834 const URI_CHAR * const afterHierPart
1835 = URI_FUNC(ParseHierPart)(state, first + 1, afterLast);
1836 state->uri->scheme.afterLast = first; /* SCHEME END */
1837 if (afterHierPart == NULL) {
1838 return NULL;
1839 }
1840 return URI_FUNC(ParseUriTail)(state, afterHierPart, afterLast);
1841 }
1842
1843 default:
1844 if (!URI_FUNC(OnExitSegmentNzNcOrScheme2)(state, first)) {
1845 URI_FUNC(StopMalloc)(state);
1846 return NULL;
1847 }
1848 return URI_FUNC(ParseUriTail)(state, first, afterLast);
1849 }
1850}
1851
1852
1853
1854/*
1855 * [uriReference]->[ALPHA][segmentNzNcOrScheme2]
1856 * [uriReference]->[DIGIT][mustBeSegmentNzNc]
1857 * [uriReference]->[pctEncoded][mustBeSegmentNzNc]
1858 * [uriReference]->[subDelims][mustBeSegmentNzNc]
1859 * [uriReference]->[uriTail] // can take <NULL>
1860 * [uriReference]-><.>[mustBeSegmentNzNc]
1861 * [uriReference]-></>[partHelperTwo][uriTail]
1862 * [uriReference]-><@>[mustBeSegmentNzNc]
1863 * [uriReference]-><_>[mustBeSegmentNzNc]
1864 * [uriReference]-><~>[mustBeSegmentNzNc]
1865 * [uriReference]-><->[mustBeSegmentNzNc]
1866 */
1867static const URI_CHAR * URI_FUNC(ParseUriReference)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1868 if (first >= afterLast) {
1869 return afterLast;
1870 }
1871
1872 switch (*first) {
1873 case URI_SET_ALPHA:
1874 state->uri->scheme.first = first; /* SCHEME BEGIN */
1875 return URI_FUNC(ParseSegmentNzNcOrScheme2)(state, first + 1, afterLast);
1876
1877 case URI_SET_DIGIT:
1878 case _UT('!'):
1879 case _UT('$'):
1880 case _UT('&'):
1881 case _UT('('):
1882 case _UT(')'):
1883 case _UT('*'):
1884 case _UT(','):
1885 case _UT(';'):
1886 case _UT('\''):
1887 case _UT('+'):
1888 case _UT('='):
1889 case _UT('.'):
1890 case _UT('_'):
1891 case _UT('~'):
1892 case _UT('-'):
1893 case _UT('@'):
1894 state->uri->scheme.first = first; /* SEGMENT BEGIN, ABUSE SCHEME POINTER */
1895 return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast);
1896
1897 case _UT('%'):
1898 {
1899 const URI_CHAR * const afterPctEncoded
1900 = URI_FUNC(ParsePctEncoded)(state, first, afterLast);
1901 if (afterPctEncoded == NULL) {
1902 return NULL;
1903 }
1904 state->uri->scheme.first = first; /* SEGMENT BEGIN, ABUSE SCHEME POINTER */
1905 return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast);
1906 }
1907
1908 case _UT('/'):
1909 {
1910 const URI_CHAR * const afterPartHelperTwo
1911 = URI_FUNC(ParsePartHelperTwo)(state, first + 1, afterLast);
1912 if (afterPartHelperTwo == NULL) {
1913 return NULL;
1914 }
1915 return URI_FUNC(ParseUriTail)(state, afterPartHelperTwo, afterLast);
1916 }
1917
1918 default:
1919 return URI_FUNC(ParseUriTail)(state, first, afterLast);
1920 }
1921}
1922
1923
1924
1925/*
1926 * [uriTail]-><#>[queryFrag]
1927 * [uriTail]-><?>[queryFrag][uriTailTwo]
1928 * [uriTail]-><NULL>
1929 */
1930static URI_INLINE const URI_CHAR * URI_FUNC(ParseUriTail)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1931 if (first >= afterLast) {
1932 return afterLast;
1933 }
1934
1935 switch (*first) {
1936 case _UT('#'):
1937 {
1938 const URI_CHAR * const afterQueryFrag = URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast);
1939 if (afterQueryFrag == NULL) {
1940 return NULL;
1941 }
1942 state->uri->fragment.first = first + 1; /* FRAGMENT BEGIN */
1943 state->uri->fragment.afterLast = afterQueryFrag; /* FRAGMENT END */
1944 return afterQueryFrag;
1945 }
1946
1947 case _UT('?'):
1948 {
1949 const URI_CHAR * const afterQueryFrag
1950 = URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast);
1951 if (afterQueryFrag == NULL) {
1952 return NULL;
1953 }
1954 state->uri->query.first = first + 1; /* QUERY BEGIN */
1955 state->uri->query.afterLast = afterQueryFrag; /* QUERY END */
1956 return URI_FUNC(ParseUriTailTwo)(state, afterQueryFrag, afterLast);
1957 }
1958
1959 default:
1960 return first;
1961 }
1962}
1963
1964
1965
1966/*
1967 * [uriTailTwo]-><#>[queryFrag]
1968 * [uriTailTwo]-><NULL>
1969 */
1970static URI_INLINE const URI_CHAR * URI_FUNC(ParseUriTailTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1971 if (first >= afterLast) {
1972 return afterLast;
1973 }
1974
1975 switch (*first) {
1976 case _UT('#'):
1977 {
1978 const URI_CHAR * const afterQueryFrag = URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast);
1979 if (afterQueryFrag == NULL) {
1980 return NULL;
1981 }
1982 state->uri->fragment.first = first + 1; /* FRAGMENT BEGIN */
1983 state->uri->fragment.afterLast = afterQueryFrag; /* FRAGMENT END */
1984 return afterQueryFrag;
1985 }
1986
1987 default:
1988 return first;
1989 }
1990}
1991
1992
1993
1994/*
1995 * [zeroMoreSlashSegs]-></>[segment][zeroMoreSlashSegs]
1996 * [zeroMoreSlashSegs]-><NULL>
1997 */
1998static const URI_CHAR * URI_FUNC(ParseZeroMoreSlashSegs)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
1999 if (first >= afterLast) {
2000 return afterLast;
2001 }
2002
2003 switch (*first) {
2004 case _UT('/'):
2005 {
2006 const URI_CHAR * const afterSegment
2007 = URI_FUNC(ParseSegment)(state, first + 1, afterLast);
2008 if (afterSegment == NULL) {
2009 return NULL;
2010 }
2011 if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment)) { /* SEGMENT BOTH */
2012 URI_FUNC(StopMalloc)(state);
2013 return NULL;
2014 }
2015 return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast);
2016 }
2017
2018 default:
2019 return first;
2020 }
2021}
2022
2023
2024
2025static URI_INLINE void URI_FUNC(ResetParserState)(URI_TYPE(ParserState) * state) {
2026 URI_TYPE(Uri) * const uriBackup = state->uri;
2027 memset(state, 0, sizeof(URI_TYPE(ParserState)));
2028 state->uri = uriBackup;
2029}
2030
2031
2032
2033static URI_INLINE UriBool URI_FUNC(PushPathSegment)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
2034 URI_TYPE(PathSegment) * segment = malloc(1 * sizeof(URI_TYPE(PathSegment)));
2035 if (segment == NULL) {
2036 return URI_FALSE; /* Raises malloc error */
2037 }
2038 memset(segment, 0, sizeof(URI_TYPE(PathSegment)));
2039 if (first == afterLast) {
2040 segment->text.first = URI_FUNC(SafeToPointTo);
2041 segment->text.afterLast = URI_FUNC(SafeToPointTo);
2042 } else {
2043 segment->text.first = first;
2044 segment->text.afterLast = afterLast;
2045 }
2046
2047 /* First segment ever? */
2048 if (state->uri->pathHead == NULL) {
2049 /* First segment ever, set head and tail */
2050 state->uri->pathHead = segment;
2051 state->uri->pathTail = segment;
2052 } else {
2053 /* Append, update tail */
2054 state->uri->pathTail->next = segment;
2055 state->uri->pathTail = segment;
2056 }
2057
2058 return URI_TRUE; /* Success */
2059}
2060
2061
2062
2063int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) {
2064 const URI_CHAR * afterUriReference;
2065 URI_TYPE(Uri) * uri;
2066
2067 /* Check params */
2068 if ((state == NULL) || (first == NULL) || (afterLast == NULL)) {
2069 return URI_ERROR_NULL;
2070 }
2071 uri = state->uri;
2072
2073 /* Init parser */
2074 URI_FUNC(ResetParserState)(state);
2075 URI_FUNC(ResetUri)(uri);
2076
2077 /* Parse */
2078 afterUriReference = URI_FUNC(ParseUriReference)(state, first, afterLast);
2079 if (afterUriReference == NULL) {
2080 return state->errorCode;
2081 }
2082 if (afterUriReference != afterLast) {
2083 URI_FUNC(StopSyntax)(state, afterUriReference);
2084 return state->errorCode;
2085 }
2086 return URI_SUCCESS;
2087}
2088
2089
2090
2091int URI_FUNC(ParseUri)(URI_TYPE(ParserState) * state, const URI_CHAR * text) {
2092 if ((state == NULL) || (text == NULL)) {
2093 return URI_ERROR_NULL;
2094 }
2095 return URI_FUNC(ParseUriEx)(state, text, text + URI_STRLEN(text));
2096}
2097
2098
2099
2100void URI_FUNC(FreeUriMembers)(URI_TYPE(Uri) * uri) {
2101 if (uri == NULL) {
2102 return;
2103 }
2104
2105 if (uri->owner) {
2106 /* Scheme */
2107 if (uri->scheme.first != NULL) {
2108 if (uri->scheme.first != uri->scheme.afterLast) {
2109 free((URI_CHAR *)uri->scheme.first);
2110 }
2111 uri->scheme.first = NULL;
2112 uri->scheme.afterLast = NULL;
2113 }
2114
2115 /* User info */
2116 if (uri->userInfo.first != NULL) {
2117 if (uri->userInfo.first != uri->userInfo.afterLast) {
2118 free((URI_CHAR *)uri->userInfo.first);
2119 }
2120 uri->userInfo.first = NULL;
2121 uri->userInfo.afterLast = NULL;
2122 }
2123
2124 /* Host data - IPvFuture */
2125 if (uri->hostData.ipFuture.first != NULL) {
2126 if (uri->hostData.ipFuture.first != uri->hostData.ipFuture.afterLast) {
2127 free((URI_CHAR *)uri->hostData.ipFuture.first);
2128 }
2129 uri->hostData.ipFuture.first = NULL;
2130 uri->hostData.ipFuture.afterLast = NULL;
2131 uri->hostText.first = NULL;
2132 uri->hostText.afterLast = NULL;
2133 }
2134
2135 /* Host text (if regname, after IPvFuture!) */
2136 if ((uri->hostText.first != NULL)
2137 && (uri->hostData.ip4 == NULL)
2138 && (uri->hostData.ip6 == NULL)) {
2139 /* Real regname */
2140 if (uri->hostText.first != uri->hostText.afterLast) {
2141 free((URI_CHAR *)uri->hostText.first);
2142 }
2143 uri->hostText.first = NULL;
2144 uri->hostText.afterLast = NULL;
2145 }
2146 }
2147
2148 /* Host data - IPv4 */
2149 if (uri->hostData.ip4 != NULL) {
2150 free(uri->hostData.ip4);
2151 uri->hostData.ip4 = NULL;
2152 }
2153
2154 /* Host data - IPv6 */
2155 if (uri->hostData.ip6 != NULL) {
2156 free(uri->hostData.ip6);
2157 uri->hostData.ip6 = NULL;
2158 }
2159
2160 /* Port text */
2161 if (uri->owner && (uri->portText.first != NULL)) {
2162 if (uri->portText.first != uri->portText.afterLast) {
2163 free((URI_CHAR *)uri->portText.first);
2164 }
2165 uri->portText.first = NULL;
2166 uri->portText.afterLast = NULL;
2167 }
2168
2169 /* Path */
2170 if (uri->pathHead != NULL) {
2171 URI_TYPE(PathSegment) * segWalk = uri->pathHead;
2172 while (segWalk != NULL) {
2173 URI_TYPE(PathSegment) * const next = segWalk->next;
2174 if (uri->owner && (segWalk->text.first != NULL)
2175 && (segWalk->text.first < segWalk->text.afterLast)) {
2176 free((URI_CHAR *)segWalk->text.first);
2177 }
2178 free(segWalk);
2179 segWalk = next;
2180 }
2181 uri->pathHead = NULL;
2182 uri->pathTail = NULL;
2183 }
2184
2185 if (uri->owner) {
2186 /* Query */
2187 if (uri->query.first != NULL) {
2188 if (uri->query.first != uri->query.afterLast) {
2189 free((URI_CHAR *)uri->query.first);
2190 }
2191 uri->query.first = NULL;
2192 uri->query.afterLast = NULL;
2193 }
2194
2195 /* Fragment */
2196 if (uri->fragment.first != NULL) {
2197 if (uri->fragment.first != uri->fragment.afterLast) {
2198 free((URI_CHAR *)uri->fragment.first);
2199 }
2200 uri->fragment.first = NULL;
2201 uri->fragment.afterLast = NULL;
2202 }
2203 }
2204}
2205
2206
2207
2208UriBool URI_FUNC(_TESTING_ONLY_ParseIpSix)(const URI_CHAR * text) {
2209 URI_TYPE(Uri) uri;
2210 URI_TYPE(ParserState) parser;
2211 const URI_CHAR * const afterIpSix = text + URI_STRLEN(text);
2212 const URI_CHAR * res;
2213
2214 URI_FUNC(ResetParserState)(&parser);
2215 URI_FUNC(ResetUri)(&uri);
2216 parser.uri = &uri;
2217 parser.uri->hostData.ip6 = malloc(1 * sizeof(UriIp6));
2218 res = URI_FUNC(ParseIPv6address2)(&parser, text, afterIpSix);
2219 URI_FUNC(FreeUriMembers)(&uri);
2220 return res == afterIpSix ? URI_TRUE : URI_FALSE;
2221}
2222
2223
2224
2225UriBool URI_FUNC(_TESTING_ONLY_ParseIpFour)(const URI_CHAR * text) {
2226 unsigned char octets[4];
2227 int res = URI_FUNC(ParseIpFourAddress)(octets, text, text + URI_STRLEN(text));
2228 return (res == URI_SUCCESS) ? URI_TRUE : URI_FALSE;
2229}
2230
2231
2232
2233#undef URI_SET_DIGIT
2234#undef URI_SET_HEX_LETTER_UPPER
2235#undef URI_SET_HEX_LETTER_LOWER
2236#undef URI_SET_HEXDIG
2237#undef URI_SET_ALPHA
2238
2239
2240
2241#endif
diff --git a/plugins/uriparser/UriParseBase.c b/plugins/uriparser/UriParseBase.c
new file mode 100644
index 00000000..75e47613
--- /dev/null
+++ b/plugins/uriparser/UriParseBase.c
@@ -0,0 +1,90 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40#ifndef URI_DOXYGEN
41# include "UriParseBase.h"
42#endif
43
44
45
46void uriWriteQuadToDoubleByte(const unsigned char * hexDigits, int digitCount, unsigned char * output) {
47 switch (digitCount) {
48 case 1:
49 /* 0x___? -> \x00 \x0? */
50 output[0] = 0;
51 output[1] = hexDigits[0];
52 break;
53
54 case 2:
55 /* 0x__?? -> \0xx \x?? */
56 output[0] = 0;
57 output[1] = 16 * hexDigits[0] + hexDigits[1];
58 break;
59
60 case 3:
61 /* 0x_??? -> \0x? \x?? */
62 output[0] = hexDigits[0];
63 output[1] = 16 * hexDigits[1] + hexDigits[2];
64 break;
65
66 case 4:
67 /* 0x???? -> \0?? \x?? */
68 output[0] = 16 * hexDigits[0] + hexDigits[1];
69 output[1] = 16 * hexDigits[2] + hexDigits[3];
70 break;
71
72 }
73}
74
75
76
77unsigned char uriGetOctetValue(const unsigned char * digits, int digitCount) {
78 switch (digitCount) {
79 case 1:
80 return digits[0];
81
82 case 2:
83 return 10 * digits[0] + digits[1];
84
85 case 3:
86 default:
87 return 100 * digits[0] + 10 * digits[1] + digits[2];
88
89 }
90}
diff --git a/plugins/uriparser/UriParseBase.h b/plugins/uriparser/UriParseBase.h
new file mode 100644
index 00000000..dea5c81c
--- /dev/null
+++ b/plugins/uriparser/UriParseBase.h
@@ -0,0 +1,55 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40#ifndef URI_PARSE_BASE_H
41#define URI_PARSE_BASE_H 1
42
43
44
45#include <uriparser/UriBase.h>
46
47
48
49void uriWriteQuadToDoubleByte(const unsigned char * hexDigits, int digitCount,
50 unsigned char * output);
51unsigned char uriGetOctetValue(const unsigned char * digits, int digitCount);
52
53
54
55#endif /* URI_PARSE_BASE_H */
diff --git a/plugins/uriparser/UriQuery.c b/plugins/uriparser/UriQuery.c
new file mode 100644
index 00000000..7adb0736
--- /dev/null
+++ b/plugins/uriparser/UriQuery.c
@@ -0,0 +1,460 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriQuery.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriQuery.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include "UriCommon.h"
67#endif
68
69
70
71static int URI_FUNC(ComposeQueryEngine)(URI_CHAR * dest,
72 const URI_TYPE(QueryList) * queryList,
73 int maxChars, int * charsWritten, int * charsRequired,
74 UriBool spaceToPlus, UriBool normalizeBreaks);
75
76static UriBool URI_FUNC(AppendQueryItem)(URI_TYPE(QueryList) ** prevNext,
77 int * itemCount, const URI_CHAR * keyFirst, const URI_CHAR * keyAfter,
78 const URI_CHAR * valueFirst, const URI_CHAR * valueAfter,
79 UriBool plusToSpace, UriBreakConversion breakConversion);
80
81
82
83int URI_FUNC(ComposeQueryCharsRequired)(const URI_TYPE(QueryList) * queryList,
84 int * charsRequired) {
85 const UriBool spaceToPlus = URI_TRUE;
86 const UriBool normalizeBreaks = URI_TRUE;
87
88 return URI_FUNC(ComposeQueryCharsRequiredEx)(queryList, charsRequired,
89 spaceToPlus, normalizeBreaks);
90}
91
92
93
94int URI_FUNC(ComposeQueryCharsRequiredEx)(const URI_TYPE(QueryList) * queryList,
95 int * charsRequired, UriBool spaceToPlus, UriBool normalizeBreaks) {
96 if ((queryList == NULL) || (charsRequired == NULL)) {
97 return URI_ERROR_NULL;
98 }
99
100 return URI_FUNC(ComposeQueryEngine)(NULL, queryList, 0, NULL,
101 charsRequired, spaceToPlus, normalizeBreaks);
102}
103
104
105
106int URI_FUNC(ComposeQuery)(URI_CHAR * dest,
107 const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten) {
108 const UriBool spaceToPlus = URI_TRUE;
109 const UriBool normalizeBreaks = URI_TRUE;
110
111 return URI_FUNC(ComposeQueryEx)(dest, queryList, maxChars, charsWritten,
112 spaceToPlus, normalizeBreaks);
113}
114
115
116
117int URI_FUNC(ComposeQueryEx)(URI_CHAR * dest,
118 const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten,
119 UriBool spaceToPlus, UriBool normalizeBreaks) {
120 if ((dest == NULL) || (queryList == NULL)) {
121 return URI_ERROR_NULL;
122 }
123
124 if (maxChars < 1) {
125 return URI_ERROR_OUTPUT_TOO_LARGE;
126 }
127
128 return URI_FUNC(ComposeQueryEngine)(dest, queryList, maxChars,
129 charsWritten, NULL, spaceToPlus, normalizeBreaks);
130}
131
132
133
134int URI_FUNC(ComposeQueryMalloc)(URI_CHAR ** dest,
135 const URI_TYPE(QueryList) * queryList) {
136 const UriBool spaceToPlus = URI_TRUE;
137 const UriBool normalizeBreaks = URI_TRUE;
138
139 return URI_FUNC(ComposeQueryMallocEx)(dest, queryList,
140 spaceToPlus, normalizeBreaks);
141}
142
143
144
145int URI_FUNC(ComposeQueryMallocEx)(URI_CHAR ** dest,
146 const URI_TYPE(QueryList) * queryList,
147 UriBool spaceToPlus, UriBool normalizeBreaks) {
148 int charsRequired;
149 int res;
150 URI_CHAR * queryString;
151
152 if (dest == NULL) {
153 return URI_ERROR_NULL;
154 }
155
156 /* Calculate space */
157 res = URI_FUNC(ComposeQueryCharsRequiredEx)(queryList, &charsRequired,
158 spaceToPlus, normalizeBreaks);
159 if (res != URI_SUCCESS) {
160 return res;
161 }
162 charsRequired++;
163
164 /* Allocate space */
165 queryString = malloc(charsRequired * sizeof(URI_CHAR));
166 if (queryString == NULL) {
167 return URI_ERROR_MALLOC;
168 }
169
170 /* Put query in */
171 res = URI_FUNC(ComposeQueryEx)(queryString, queryList, charsRequired,
172 NULL, spaceToPlus, normalizeBreaks);
173 if (res != URI_SUCCESS) {
174 free(queryString);
175 return res;
176 }
177
178 *dest = queryString;
179 return URI_SUCCESS;
180}
181
182
183
184int URI_FUNC(ComposeQueryEngine)(URI_CHAR * dest,
185 const URI_TYPE(QueryList) * queryList,
186 int maxChars, int * charsWritten, int * charsRequired,
187 UriBool spaceToPlus, UriBool normalizeBreaks) {
188 UriBool firstItem = URI_TRUE;
189 int ampersandLen = 0;
190 URI_CHAR * write = dest;
191
192 /* Subtract terminator */
193 if (dest == NULL) {
194 *charsRequired = 0;
195 } else {
196 maxChars--;
197 }
198
199 while (queryList != NULL) {
200 const URI_CHAR * const key = queryList->key;
201 const URI_CHAR * const value = queryList->value;
202 const int worstCase = (normalizeBreaks == URI_TRUE ? 6 : 3);
203 const int keyLen = (key == NULL) ? 0 : (int)URI_STRLEN(key);
204 const int keyRequiredChars = worstCase * keyLen;
205 const int valueLen = (value == NULL) ? 0 : (int)URI_STRLEN(value);
206 const int valueRequiredChars = worstCase * valueLen;
207
208 if (dest == NULL) {
209 if (firstItem == URI_TRUE) {
210 ampersandLen = 1;
211 firstItem = URI_FALSE;
212 }
213
214 (*charsRequired) += ampersandLen + keyRequiredChars + ((value == NULL)
215 ? 0
216 : 1 + valueRequiredChars);
217 } else {
218 URI_CHAR * afterKey;
219
220 if ((write - dest) + ampersandLen + keyRequiredChars > maxChars) {
221 return URI_ERROR_OUTPUT_TOO_LARGE;
222 }
223
224 /* Copy key */
225 if (firstItem == URI_TRUE) {
226 firstItem = URI_FALSE;
227 } else {
228 write[0] = _UT('&');
229 write++;
230 }
231 afterKey = URI_FUNC(EscapeEx)(key, key + keyLen,
232 write, spaceToPlus, normalizeBreaks);
233 write += (afterKey - write);
234
235 if (value != NULL) {
236 URI_CHAR * afterValue;
237
238 if ((write - dest) + 1 + valueRequiredChars > maxChars) {
239 return URI_ERROR_OUTPUT_TOO_LARGE;
240 }
241
242 /* Copy value */
243 write[0] = _UT('=');
244 write++;
245 afterValue = URI_FUNC(EscapeEx)(value, value + valueLen,
246 write, spaceToPlus, normalizeBreaks);
247 write += (afterValue - write);
248 }
249 }
250
251 queryList = queryList->next;
252 }
253
254 if (dest != NULL) {
255 write[0] = _UT('\0');
256 if (charsWritten != NULL) {
257 *charsWritten = (int)(write - dest) + 1; /* .. for terminator */
258 }
259 }
260
261 return URI_SUCCESS;
262}
263
264
265
266UriBool URI_FUNC(AppendQueryItem)(URI_TYPE(QueryList) ** prevNext,
267 int * itemCount, const URI_CHAR * keyFirst, const URI_CHAR * keyAfter,
268 const URI_CHAR * valueFirst, const URI_CHAR * valueAfter,
269 UriBool plusToSpace, UriBreakConversion breakConversion) {
270 const int keyLen = (int)(keyAfter - keyFirst);
271 const int valueLen = (int)(valueAfter - valueFirst);
272 URI_CHAR * key;
273 URI_CHAR * value;
274
275 if ((prevNext == NULL) || (itemCount == NULL)
276 || (keyFirst == NULL) || (keyAfter == NULL)
277 || (keyFirst > keyAfter) || (valueFirst > valueAfter)
278 || ((keyFirst == keyAfter)
279 && (valueFirst == NULL) && (valueAfter == NULL))) {
280 return URI_TRUE;
281 }
282
283 /* Append new empty item */
284 *prevNext = malloc(1 * sizeof(URI_TYPE(QueryList)));
285 if (*prevNext == NULL) {
286 return URI_FALSE; /* Raises malloc error */
287 }
288 (*prevNext)->next = NULL;
289
290
291 /* Fill key */
292 key = malloc((keyLen + 1) * sizeof(URI_CHAR));
293 if (key == NULL) {
294 free(*prevNext);
295 *prevNext = NULL;
296 return URI_FALSE; /* Raises malloc error */
297 }
298
299 key[keyLen] = _UT('\0');
300 if (keyLen > 0) {
301 /* Copy 1:1 */
302 memcpy(key, keyFirst, keyLen * sizeof(URI_CHAR));
303
304 /* Unescape */
305 URI_FUNC(UnescapeInPlaceEx)(key, plusToSpace, breakConversion);
306 }
307 (*prevNext)->key = key;
308
309
310 /* Fill value */
311 if (valueFirst != NULL) {
312 value = malloc((valueLen + 1) * sizeof(URI_CHAR));
313 if (value == NULL) {
314 free(key);
315 free(*prevNext);
316 *prevNext = NULL;
317 return URI_FALSE; /* Raises malloc error */
318 }
319
320 value[valueLen] = _UT('\0');
321 if (valueLen > 0) {
322 /* Copy 1:1 */
323 memcpy(value, valueFirst, valueLen * sizeof(URI_CHAR));
324
325 /* Unescape */
326 URI_FUNC(UnescapeInPlaceEx)(value, plusToSpace, breakConversion);
327 }
328 (*prevNext)->value = value;
329 } else {
330 value = NULL;
331 }
332 (*prevNext)->value = value;
333
334 (*itemCount)++;
335 return URI_TRUE;
336}
337
338
339
340void URI_FUNC(FreeQueryList)(URI_TYPE(QueryList) * queryList) {
341 while (queryList != NULL) {
342 URI_TYPE(QueryList) * nextBackup = queryList->next;
343 free((URI_CHAR *)queryList->key); /* const cast */
344 free((URI_CHAR *)queryList->value); /* const cast */
345 free(queryList);
346 queryList = nextBackup;
347 }
348}
349
350
351
352int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) ** dest, int * itemCount,
353 const URI_CHAR * first, const URI_CHAR * afterLast) {
354 const UriBool plusToSpace = URI_TRUE;
355 const UriBreakConversion breakConversion = URI_BR_DONT_TOUCH;
356
357 return URI_FUNC(DissectQueryMallocEx)(dest, itemCount, first, afterLast,
358 plusToSpace, breakConversion);
359}
360
361
362
363int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) ** dest, int * itemCount,
364 const URI_CHAR * first, const URI_CHAR * afterLast,
365 UriBool plusToSpace, UriBreakConversion breakConversion) {
366 const URI_CHAR * walk = first;
367 const URI_CHAR * keyFirst = first;
368 const URI_CHAR * keyAfter = NULL;
369 const URI_CHAR * valueFirst = NULL;
370 const URI_CHAR * valueAfter = NULL;
371 URI_TYPE(QueryList) ** prevNext = dest;
372 int nullCounter;
373 int * itemsAppended = (itemCount == NULL) ? &nullCounter : itemCount;
374
375 if ((dest == NULL) || (first == NULL) || (afterLast == NULL)) {
376 return URI_ERROR_NULL;
377 }
378
379 if (first > afterLast) {
380 return URI_ERROR_RANGE_INVALID;
381 }
382
383 *dest = NULL;
384 *itemsAppended = 0;
385
386 /* Parse query string */
387 for (; walk < afterLast; walk++) {
388 switch (*walk) {
389 case _UT('&'):
390 if (valueFirst != NULL) {
391 valueAfter = walk;
392 } else {
393 keyAfter = walk;
394 }
395
396 if (URI_FUNC(AppendQueryItem)(prevNext, itemsAppended,
397 keyFirst, keyAfter, valueFirst, valueAfter,
398 plusToSpace, breakConversion)
399 == URI_FALSE) {
400 /* Free list we built */
401 *itemsAppended = 0;
402 URI_FUNC(FreeQueryList)(*dest);
403 return URI_ERROR_MALLOC;
404 }
405
406 /* Make future items children of the current */
407 if ((prevNext != NULL) && (*prevNext != NULL)) {
408 prevNext = &((*prevNext)->next);
409 }
410
411 if (walk + 1 < afterLast) {
412 keyFirst = walk + 1;
413 } else {
414 keyFirst = NULL;
415 }
416 keyAfter = NULL;
417 valueFirst = NULL;
418 valueAfter = NULL;
419 break;
420
421 case _UT('='):
422 /* NOTE: WE treat the first '=' as a separator, */
423 /* all following go into the value part */
424 if (keyAfter == NULL) {
425 keyAfter = walk;
426 if (walk + 1 <= afterLast) {
427 valueFirst = walk + 1;
428 valueAfter = walk + 1;
429 }
430 }
431 break;
432
433 default:
434 break;
435 }
436 }
437
438 if (valueFirst != NULL) {
439 /* Must be key/value pair */
440 valueAfter = walk;
441 } else {
442 /* Must be key only */
443 keyAfter = walk;
444 }
445
446 if (URI_FUNC(AppendQueryItem)(prevNext, itemsAppended, keyFirst, keyAfter,
447 valueFirst, valueAfter, plusToSpace, breakConversion)
448 == URI_FALSE) {
449 /* Free list we built */
450 *itemsAppended = 0;
451 URI_FUNC(FreeQueryList)(*dest);
452 return URI_ERROR_MALLOC;
453 }
454
455 return URI_SUCCESS;
456}
457
458
459
460#endif
diff --git a/plugins/uriparser/UriRecompose.c b/plugins/uriparser/UriRecompose.c
new file mode 100644
index 00000000..09daee07
--- /dev/null
+++ b/plugins/uriparser/UriRecompose.c
@@ -0,0 +1,577 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriRecompose.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriRecompose.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include "UriCommon.h"
67#endif
68
69
70
71static int URI_FUNC(ToStringEngine)(URI_CHAR * dest, const URI_TYPE(Uri) * uri,
72 int maxChars, int * charsWritten, int * charsRequired);
73
74
75
76int URI_FUNC(ToStringCharsRequired)(const URI_TYPE(Uri) * uri,
77 int * charsRequired) {
78 const int MAX_CHARS = ((unsigned int)-1) >> 1;
79 return URI_FUNC(ToStringEngine)(NULL, uri, MAX_CHARS, NULL, charsRequired);
80}
81
82
83
84int URI_FUNC(ToString)(URI_CHAR * dest, const URI_TYPE(Uri) * uri,
85 int maxChars, int * charsWritten) {
86 return URI_FUNC(ToStringEngine)(dest, uri, maxChars, charsWritten, NULL);
87}
88
89
90
91static URI_INLINE int URI_FUNC(ToStringEngine)(URI_CHAR * dest,
92 const URI_TYPE(Uri) * uri, int maxChars, int * charsWritten,
93 int * charsRequired) {
94 int written = 0;
95 if ((uri == NULL) || ((dest == NULL) && (charsRequired == NULL))) {
96 if (charsWritten != NULL) {
97 *charsWritten = 0;
98 }
99 return URI_ERROR_NULL;
100 }
101
102 if (maxChars < 1) {
103 if (charsWritten != NULL) {
104 *charsWritten = 0;
105 }
106 return URI_ERROR_TOSTRING_TOO_LONG;
107 }
108 maxChars--; /* So we don't have to substract 1 for '\0' all the time */
109
110 /* [01/19] result = "" */
111 if (dest != NULL) {
112 dest[0] = _UT('\0');
113 } else {
114 (*charsRequired) = 0;
115 }
116 /* [02/19] if defined(scheme) then */
117 if (uri->scheme.first != NULL) {
118 /* [03/19] append scheme to result; */
119 const int charsToWrite
120 = (int)(uri->scheme.afterLast - uri->scheme.first);
121 if (dest != NULL) {
122 if (written + charsToWrite <= maxChars) {
123 memcpy(dest + written, uri->scheme.first,
124 charsToWrite * sizeof(URI_CHAR));
125 written += charsToWrite;
126 } else {
127 dest[0] = _UT('\0');
128 if (charsWritten != NULL) {
129 *charsWritten = 0;
130 }
131 return URI_ERROR_TOSTRING_TOO_LONG;
132 }
133 } else {
134 (*charsRequired) += charsToWrite;
135 }
136 /* [04/19] append ":" to result; */
137 if (dest != NULL) {
138 if (written + 1 <= maxChars) {
139 memcpy(dest + written, _UT(":"),
140 1 * sizeof(URI_CHAR));
141 written += 1;
142 } else {
143 dest[0] = _UT('\0');
144 if (charsWritten != NULL) {
145 *charsWritten = 0;
146 }
147 return URI_ERROR_TOSTRING_TOO_LONG;
148 }
149 } else {
150 (*charsRequired) += 1;
151 }
152 /* [05/19] endif; */
153 }
154 /* [06/19] if defined(authority) then */
155 if (URI_FUNC(IsHostSet)(uri)) {
156 /* [07/19] append "//" to result; */
157 if (dest != NULL) {
158 if (written + 2 <= maxChars) {
159 memcpy(dest + written, _UT("//"),
160 2 * sizeof(URI_CHAR));
161 written += 2;
162 } else {
163 dest[0] = _UT('\0');
164 if (charsWritten != NULL) {
165 *charsWritten = 0;
166 }
167 return URI_ERROR_TOSTRING_TOO_LONG;
168 }
169 } else {
170 (*charsRequired) += 2;
171 }
172 /* [08/19] append authority to result; */
173 /* UserInfo */
174 if (uri->userInfo.first != NULL) {
175 const int charsToWrite = (int)(uri->userInfo.afterLast - uri->userInfo.first);
176 if (dest != NULL) {
177 if (written + charsToWrite <= maxChars) {
178 memcpy(dest + written, uri->userInfo.first,
179 charsToWrite * sizeof(URI_CHAR));
180 written += charsToWrite;
181 } else {
182 dest[0] = _UT('\0');
183 if (charsWritten != NULL) {
184 *charsWritten = 0;
185 }
186 return URI_ERROR_TOSTRING_TOO_LONG;
187 }
188
189 if (written + 1 <= maxChars) {
190 memcpy(dest + written, _UT("@"),
191 1 * sizeof(URI_CHAR));
192 written += 1;
193 } else {
194 dest[0] = _UT('\0');
195 if (charsWritten != NULL) {
196 *charsWritten = 0;
197 }
198 return URI_ERROR_TOSTRING_TOO_LONG;
199 }
200 } else {
201 (*charsRequired) += charsToWrite + 1;
202 }
203 }
204
205 /* Host */
206 if (uri->hostData.ip4 != NULL) {
207 /* IPv4 */
208 int i = 0;
209 for (; i < 4; i++) {
210 const unsigned char value = uri->hostData.ip4->data[i];
211 const int charsToWrite = (value > 99) ? 3 : ((value > 9) ? 2 : 1);
212 if (dest != NULL) {
213 if (written + charsToWrite <= maxChars) {
214 URI_CHAR text[4];
215 if (value > 99) {
216 text[0] = _UT('0') + (value / 100);
217 text[1] = _UT('0') + ((value % 100) / 10);
218 text[2] = _UT('0') + (value % 10);
219 } else if (value > 9) {
220 text[0] = _UT('0') + (value / 10);
221 text[1] = _UT('0') + (value % 10);
222 } else {
223 text[0] = _UT('0') + value;
224 }
225 text[charsToWrite] = _UT('\0');
226 memcpy(dest + written, text, charsToWrite * sizeof(URI_CHAR));
227 written += charsToWrite;
228 } else {
229 dest[0] = _UT('\0');
230 if (charsWritten != NULL) {
231 *charsWritten = 0;
232 }
233 return URI_ERROR_TOSTRING_TOO_LONG;
234 }
235 if (i < 3) {
236 if (written + 1 <= maxChars) {
237 memcpy(dest + written, _UT("."),
238 1 * sizeof(URI_CHAR));
239 written += 1;
240 } else {
241 dest[0] = _UT('\0');
242 if (charsWritten != NULL) {
243 *charsWritten = 0;
244 }
245 return URI_ERROR_TOSTRING_TOO_LONG;
246 }
247 }
248 } else {
249 (*charsRequired) += charsToWrite + 1;
250 }
251 }
252 } else if (uri->hostData.ip6 != NULL) {
253 /* IPv6 */
254 int i = 0;
255 if (dest != NULL) {
256 if (written + 1 <= maxChars) {
257 memcpy(dest + written, _UT("["),
258 1 * sizeof(URI_CHAR));
259 written += 1;
260 } else {
261 dest[0] = _UT('\0');
262 if (charsWritten != NULL) {
263 *charsWritten = 0;
264 }
265 return URI_ERROR_TOSTRING_TOO_LONG;
266 }
267 } else {
268 (*charsRequired) += 1;
269 }
270
271 for (; i < 16; i++) {
272 const unsigned char value = uri->hostData.ip6->data[i];
273 if (dest != NULL) {
274 if (written + 2 <= maxChars) {
275 URI_CHAR text[3];
276 text[0] = URI_FUNC(HexToLetterEx)(value / 16, URI_FALSE);
277 text[1] = URI_FUNC(HexToLetterEx)(value % 16, URI_FALSE);
278 text[2] = _UT('\0');
279 memcpy(dest + written, text, 2 * sizeof(URI_CHAR));
280 written += 2;
281 } else {
282 dest[0] = _UT('\0');
283 if (charsWritten != NULL) {
284 *charsWritten = 0;
285 }
286 return URI_ERROR_TOSTRING_TOO_LONG;
287 }
288 } else {
289 (*charsRequired) += 2;
290 }
291 if (((i & 1) == 1) && (i < 15)) {
292 if (dest != NULL) {
293 if (written + 1 <= maxChars) {
294 memcpy(dest + written, _UT(":"),
295 1 * sizeof(URI_CHAR));
296 written += 1;
297 } else {
298 dest[0] = _UT('\0');
299 if (charsWritten != NULL) {
300 *charsWritten = 0;
301 }
302 return URI_ERROR_TOSTRING_TOO_LONG;
303 }
304 } else {
305 (*charsRequired) += 1;
306 }
307 }
308 }
309
310 if (dest != NULL) {
311 if (written + 1 <= maxChars) {
312 memcpy(dest + written, _UT("]"),
313 1 * sizeof(URI_CHAR));
314 written += 1;
315 } else {
316 dest[0] = _UT('\0');
317 if (charsWritten != NULL) {
318 *charsWritten = 0;
319 }
320 return URI_ERROR_TOSTRING_TOO_LONG;
321 }
322 } else {
323 (*charsRequired) += 1;
324 }
325 } else if (uri->hostData.ipFuture.first != NULL) {
326 /* IPvFuture */
327 const int charsToWrite = (int)(uri->hostData.ipFuture.afterLast
328 - uri->hostData.ipFuture.first);
329 if (dest != NULL) {
330 if (written + 1 <= maxChars) {
331 memcpy(dest + written, _UT("["),
332 1 * sizeof(URI_CHAR));
333 written += 1;
334 } else {
335 dest[0] = _UT('\0');
336 if (charsWritten != NULL) {
337 *charsWritten = 0;
338 }
339 return URI_ERROR_TOSTRING_TOO_LONG;
340 }
341
342 if (written + charsToWrite <= maxChars) {
343 memcpy(dest + written, uri->hostData.ipFuture.first, charsToWrite * sizeof(URI_CHAR));
344 written += charsToWrite;
345 } else {
346 dest[0] = _UT('\0');
347 if (charsWritten != NULL) {
348 *charsWritten = 0;
349 }
350 return URI_ERROR_TOSTRING_TOO_LONG;
351 }
352
353 if (written + 1 <= maxChars) {
354 memcpy(dest + written, _UT("]"),
355 1 * sizeof(URI_CHAR));
356 written += 1;
357 } else {
358 dest[0] = _UT('\0');
359 if (charsWritten != NULL) {
360 *charsWritten = 0;
361 }
362 return URI_ERROR_TOSTRING_TOO_LONG;
363 }
364 } else {
365 (*charsRequired) += 1 + charsToWrite + 1;
366 }
367 } else if (uri->hostText.first != NULL) {
368 /* Regname */
369 const int charsToWrite = (int)(uri->hostText.afterLast - uri->hostText.first);
370 if (dest != NULL) {
371 if (written + charsToWrite <= maxChars) {
372 memcpy(dest + written, uri->hostText.first,
373 charsToWrite * sizeof(URI_CHAR));
374 written += charsToWrite;
375 } else {
376 dest[0] = _UT('\0');
377 if (charsWritten != NULL) {
378 *charsWritten = 0;
379 }
380 return URI_ERROR_TOSTRING_TOO_LONG;
381 }
382 } else {
383 (*charsRequired) += charsToWrite;
384 }
385 }
386
387 /* Port */
388 if (uri->portText.first != NULL) {
389 const int charsToWrite = (int)(uri->portText.afterLast - uri->portText.first);
390 if (dest != NULL) {
391 /* Leading ':' */
392 if (written + 1 <= maxChars) {
393 memcpy(dest + written, _UT(":"),
394 1 * sizeof(URI_CHAR));
395 written += 1;
396 } else {
397 dest[0] = _UT('\0');
398 if (charsWritten != NULL) {
399 *charsWritten = 0;
400 }
401 return URI_ERROR_TOSTRING_TOO_LONG;
402 }
403
404 /* Port number */
405 if (written + charsToWrite <= maxChars) {
406 memcpy(dest + written, uri->portText.first,
407 charsToWrite * sizeof(URI_CHAR));
408 written += charsToWrite;
409 } else {
410 dest[0] = _UT('\0');
411 if (charsWritten != NULL) {
412 *charsWritten = 0;
413 }
414 return URI_ERROR_TOSTRING_TOO_LONG;
415 }
416 } else {
417 (*charsRequired) += 1 + charsToWrite;
418 }
419 }
420 /* [09/19] endif; */
421 }
422 /* [10/19] append path to result; */
423 /* Slash needed here? */
424 if (uri->absolutePath || ((uri->pathHead != NULL)
425 && URI_FUNC(IsHostSet)(uri))) {
426 if (dest != NULL) {
427 if (written + 1 <= maxChars) {
428 memcpy(dest + written, _UT("/"),
429 1 * sizeof(URI_CHAR));
430 written += 1;
431 } else {
432 dest[0] = _UT('\0');
433 if (charsWritten != NULL) {
434 *charsWritten = 0;
435 }
436 return URI_ERROR_TOSTRING_TOO_LONG;
437 }
438 } else {
439 (*charsRequired) += 1;
440 }
441 }
442
443 if (uri->pathHead != NULL) {
444 URI_TYPE(PathSegment) * walker = uri->pathHead;
445 do {
446 const int charsToWrite = (int)(walker->text.afterLast - walker->text.first);
447 if (dest != NULL) {
448 if (written + charsToWrite <= maxChars) {
449 memcpy(dest + written, walker->text.first,
450 charsToWrite * sizeof(URI_CHAR));
451 written += charsToWrite;
452 } else {
453 dest[0] = _UT('\0');
454 if (charsWritten != NULL) {
455 *charsWritten = 0;
456 }
457 return URI_ERROR_TOSTRING_TOO_LONG;
458 }
459 } else {
460 (*charsRequired) += charsToWrite;
461 }
462
463 /* Not last segment -> append slash */
464 if (walker->next != NULL) {
465 if (dest != NULL) {
466 if (written + 1 <= maxChars) {
467 memcpy(dest + written, _UT("/"),
468 1 * sizeof(URI_CHAR));
469 written += 1;
470 } else {
471 dest[0] = _UT('\0');
472 if (charsWritten != NULL) {
473 *charsWritten = 0;
474 }
475 return URI_ERROR_TOSTRING_TOO_LONG;
476 }
477 } else {
478 (*charsRequired) += 1;
479 }
480 }
481
482 walker = walker->next;
483 } while (walker != NULL);
484 }
485 /* [11/19] if defined(query) then */
486 if (uri->query.first != NULL) {
487 /* [12/19] append "?" to result; */
488 if (dest != NULL) {
489 if (written + 1 <= maxChars) {
490 memcpy(dest + written, _UT("?"),
491 1 * sizeof(URI_CHAR));
492 written += 1;
493 } else {
494 dest[0] = _UT('\0');
495 if (charsWritten != NULL) {
496 *charsWritten = 0;
497 }
498 return URI_ERROR_TOSTRING_TOO_LONG;
499 }
500 } else {
501 (*charsRequired) += 1;
502 }
503 /* [13/19] append query to result; */
504 {
505 const int charsToWrite
506 = (int)(uri->query.afterLast - uri->query.first);
507 if (dest != NULL) {
508 if (written + charsToWrite <= maxChars) {
509 memcpy(dest + written, uri->query.first,
510 charsToWrite * sizeof(URI_CHAR));
511 written += charsToWrite;
512 } else {
513 dest[0] = _UT('\0');
514 if (charsWritten != NULL) {
515 *charsWritten = 0;
516 }
517 return URI_ERROR_TOSTRING_TOO_LONG;
518 }
519 } else {
520 (*charsRequired) += charsToWrite;
521 }
522 }
523 /* [14/19] endif; */
524 }
525 /* [15/19] if defined(fragment) then */
526 if (uri->fragment.first != NULL) {
527 /* [16/19] append "#" to result; */
528 if (dest != NULL) {
529 if (written + 1 <= maxChars) {
530 memcpy(dest + written, _UT("#"),
531 1 * sizeof(URI_CHAR));
532 written += 1;
533 } else {
534 dest[0] = _UT('\0');
535 if (charsWritten != NULL) {
536 *charsWritten = 0;
537 }
538 return URI_ERROR_TOSTRING_TOO_LONG;
539 }
540 } else {
541 (*charsRequired) += 1;
542 }
543 /* [17/19] append fragment to result; */
544 {
545 const int charsToWrite
546 = (int)(uri->fragment.afterLast - uri->fragment.first);
547 if (dest != NULL) {
548 if (written + charsToWrite <= maxChars) {
549 memcpy(dest + written, uri->fragment.first,
550 charsToWrite * sizeof(URI_CHAR));
551 written += charsToWrite;
552 } else {
553 dest[0] = _UT('\0');
554 if (charsWritten != NULL) {
555 *charsWritten = 0;
556 }
557 return URI_ERROR_TOSTRING_TOO_LONG;
558 }
559 } else {
560 (*charsRequired) += charsToWrite;
561 }
562 }
563 /* [18/19] endif; */
564 }
565 /* [19/19] return result; */
566 if (dest != NULL) {
567 dest[written++] = _UT('\0');
568 if (charsWritten != NULL) {
569 *charsWritten = written;
570 }
571 }
572 return URI_SUCCESS;
573}
574
575
576
577#endif
diff --git a/plugins/uriparser/UriResolve.c b/plugins/uriparser/UriResolve.c
new file mode 100644
index 00000000..3660b6b2
--- /dev/null
+++ b/plugins/uriparser/UriResolve.c
@@ -0,0 +1,316 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriResolve.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriResolve.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include "UriCommon.h"
67#endif
68
69
70
71/* Appends a relative URI to an absolute. The last path segment of
72 * the absolute URI is replaced. */
73static URI_INLINE UriBool URI_FUNC(MergePath)(URI_TYPE(Uri) * absWork,
74 const URI_TYPE(Uri) * relAppend) {
75 URI_TYPE(PathSegment) * sourceWalker;
76 URI_TYPE(PathSegment) * destPrev;
77 if (relAppend->pathHead == NULL) {
78 return URI_TRUE;
79 }
80
81 /* Replace last segment ("" if trailing slash) with first of append chain */
82 if (absWork->pathHead == NULL) {
83 URI_TYPE(PathSegment) * const dup = malloc(sizeof(URI_TYPE(PathSegment)));
84 if (dup == NULL) {
85 return URI_FALSE; /* Raises malloc error */
86 }
87 dup->next = NULL;
88 absWork->pathHead = dup;
89 absWork->pathTail = dup;
90 }
91 absWork->pathTail->text.first = relAppend->pathHead->text.first;
92 absWork->pathTail->text.afterLast = relAppend->pathHead->text.afterLast;
93
94 /* Append all the others */
95 sourceWalker = relAppend->pathHead->next;
96 if (sourceWalker == NULL) {
97 return URI_TRUE;
98 }
99 destPrev = absWork->pathTail;
100
101 for (;;) {
102 URI_TYPE(PathSegment) * const dup = malloc(sizeof(URI_TYPE(PathSegment)));
103 if (dup == NULL) {
104 destPrev->next = NULL;
105 absWork->pathTail = destPrev;
106 return URI_FALSE; /* Raises malloc error */
107 }
108 dup->text = sourceWalker->text;
109 destPrev->next = dup;
110
111 if (sourceWalker->next == NULL) {
112 absWork->pathTail = dup;
113 absWork->pathTail->next = NULL;
114 break;
115 }
116 destPrev = dup;
117 sourceWalker = sourceWalker->next;
118 }
119
120 return URI_TRUE;
121}
122
123
124static int URI_FUNC(ResolveAbsolutePathFlag)(URI_TYPE(Uri) * absWork) {
125 if (absWork == NULL) {
126 return URI_ERROR_NULL;
127 }
128
129 if (URI_FUNC(IsHostSet)(absWork) && absWork->absolutePath) {
130 /* Empty segment needed, instead? */
131 if (absWork->pathHead == NULL) {
132 URI_TYPE(PathSegment) * const segment = malloc(sizeof(URI_TYPE(PathSegment)));
133 if (segment == NULL) {
134 return URI_ERROR_MALLOC;
135 }
136 segment->text.first = URI_FUNC(SafeToPointTo);
137 segment->text.afterLast = URI_FUNC(SafeToPointTo);
138 segment->next = NULL;
139
140 absWork->pathHead = segment;
141 absWork->pathTail = segment;
142 }
143
144 absWork->absolutePath = URI_FALSE;
145 }
146
147 return URI_SUCCESS;
148}
149
150
151static int URI_FUNC(AddBaseUriImpl)(URI_TYPE(Uri) * absDest,
152 const URI_TYPE(Uri) * relSource,
153 const URI_TYPE(Uri) * absBase,
154 UriResolutionOptions options) {
155 if (absDest == NULL) {
156 return URI_ERROR_NULL;
157 }
158 URI_FUNC(ResetUri)(absDest);
159
160 if ((relSource == NULL) || (absBase == NULL)) {
161 return URI_ERROR_NULL;
162 }
163
164 /* absBase absolute? */
165 if (absBase->scheme.first == NULL) {
166 return URI_ERROR_ADDBASE_REL_BASE;
167 }
168
169 /* [00/32] -- A non-strict parser may ignore a scheme in the reference */
170 /* [00/32] -- if it is identical to the base URI's scheme. */
171 /* [00/32] if ((not strict) and (R.scheme == Base.scheme)) then */
172 UriBool relSourceHasScheme = (relSource->scheme.first != NULL) ? URI_TRUE : URI_FALSE;
173 if ((options & URI_RESOLVE_IDENTICAL_SCHEME_COMPAT)
174 && (absBase->scheme.first != NULL)
175 && (relSource->scheme.first != NULL)
176 && (0 == URI_FUNC(CompareRange)(&(absBase->scheme), &(relSource->scheme)))) {
177 /* [00/32] undefine(R.scheme); */
178 relSourceHasScheme = URI_FALSE;
179 /* [00/32] endif; */
180 }
181
182 /* [01/32] if defined(R.scheme) then */
183 if (relSourceHasScheme) {
184 /* [02/32] T.scheme = R.scheme; */
185 absDest->scheme = relSource->scheme;
186 /* [03/32] T.authority = R.authority; */
187 if (!URI_FUNC(CopyAuthority)(absDest, relSource)) {
188 return URI_ERROR_MALLOC;
189 }
190 /* [04/32] T.path = remove_dot_segments(R.path); */
191 if (!URI_FUNC(CopyPath)(absDest, relSource)) {
192 return URI_ERROR_MALLOC;
193 }
194 if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest)) {
195 return URI_ERROR_MALLOC;
196 }
197 /* [05/32] T.query = R.query; */
198 absDest->query = relSource->query;
199 /* [06/32] else */
200 } else {
201 /* [07/32] if defined(R.authority) then */
202 if (URI_FUNC(IsHostSet)(relSource)) {
203 /* [08/32] T.authority = R.authority; */
204 if (!URI_FUNC(CopyAuthority)(absDest, relSource)) {
205 return URI_ERROR_MALLOC;
206 }
207 /* [09/32] T.path = remove_dot_segments(R.path); */
208 if (!URI_FUNC(CopyPath)(absDest, relSource)) {
209 return URI_ERROR_MALLOC;
210 }
211 if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest)) {
212 return URI_ERROR_MALLOC;
213 }
214 /* [10/32] T.query = R.query; */
215 absDest->query = relSource->query;
216 /* [11/32] else */
217 } else {
218 /* [28/32] T.authority = Base.authority; */
219 if (!URI_FUNC(CopyAuthority)(absDest, absBase)) {
220 return URI_ERROR_MALLOC;
221 }
222 /* [12/32] if (R.path == "") then */
223 if (relSource->pathHead == NULL && !relSource->absolutePath) {
224 /* [13/32] T.path = Base.path; */
225 if (!URI_FUNC(CopyPath)(absDest, absBase)) {
226 return URI_ERROR_MALLOC;
227 }
228 /* [14/32] if defined(R.query) then */
229 if (relSource->query.first != NULL) {
230 /* [15/32] T.query = R.query; */
231 absDest->query = relSource->query;
232 /* [16/32] else */
233 } else {
234 /* [17/32] T.query = Base.query; */
235 absDest->query = absBase->query;
236 /* [18/32] endif; */
237 }
238 /* [19/32] else */
239 } else {
240 /* [20/32] if (R.path starts-with "/") then */
241 if (relSource->absolutePath) {
242 int res;
243 /* [21/32] T.path = remove_dot_segments(R.path); */
244 if (!URI_FUNC(CopyPath)(absDest, relSource)) {
245 return URI_ERROR_MALLOC;
246 }
247 res = URI_FUNC(ResolveAbsolutePathFlag)(absDest);
248 if (res != URI_SUCCESS) {
249 return res;
250 }
251 if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest)) {
252 return URI_ERROR_MALLOC;
253 }
254 /* [22/32] else */
255 } else {
256 /* [23/32] T.path = merge(Base.path, R.path); */
257 if (!URI_FUNC(CopyPath)(absDest, absBase)) {
258 return URI_ERROR_MALLOC;
259 }
260 if (!URI_FUNC(MergePath)(absDest, relSource)) {
261 return URI_ERROR_MALLOC;
262 }
263 /* [24/32] T.path = remove_dot_segments(T.path); */
264 if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest)) {
265 return URI_ERROR_MALLOC;
266 }
267
268 if (!URI_FUNC(FixAmbiguity)(absDest)) {
269 return URI_ERROR_MALLOC;
270 }
271 /* [25/32] endif; */
272 }
273 /* [26/32] T.query = R.query; */
274 absDest->query = relSource->query;
275 /* [27/32] endif; */
276 }
277 URI_FUNC(FixEmptyTrailSegment)(absDest);
278 /* [29/32] endif; */
279 }
280 /* [30/32] T.scheme = Base.scheme; */
281 absDest->scheme = absBase->scheme;
282 /* [31/32] endif; */
283 }
284 /* [32/32] T.fragment = R.fragment; */
285 absDest->fragment = relSource->fragment;
286
287 return URI_SUCCESS;
288
289}
290
291
292
293int URI_FUNC(AddBaseUri)(URI_TYPE(Uri) * absDest,
294 const URI_TYPE(Uri) * relSource, const URI_TYPE(Uri) * absBase) {
295 const int res = URI_FUNC(AddBaseUriImpl)(absDest, relSource, absBase, URI_RESOLVE_STRICTLY);
296 if ((res != URI_SUCCESS) && (absDest != NULL)) {
297 URI_FUNC(FreeUriMembers)(absDest);
298 }
299 return res;
300}
301
302
303
304int URI_FUNC(AddBaseUriEx)(URI_TYPE(Uri) * absDest,
305 const URI_TYPE(Uri) * relSource, const URI_TYPE(Uri) * absBase,
306 UriResolutionOptions options) {
307 const int res = URI_FUNC(AddBaseUriImpl)(absDest, relSource, absBase, options);
308 if ((res != URI_SUCCESS) && (absDest != NULL)) {
309 URI_FUNC(FreeUriMembers)(absDest);
310 }
311 return res;
312}
313
314
315
316#endif
diff --git a/plugins/uriparser/UriShorten.c b/plugins/uriparser/UriShorten.c
new file mode 100644
index 00000000..c839ae53
--- /dev/null
+++ b/plugins/uriparser/UriShorten.c
@@ -0,0 +1,320 @@
1/*
2 * uriparser - RFC 3986 URI parsing library
3 *
4 * Copyright (C) 2007, Weijia Song <songweijia@gmail.com>
5 * Copyright (C) 2007, Sebastian Pipping <webmaster@hartwork.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * * Redistributions of source code must retain the above
13 * copyright notice, this list of conditions and the following
14 * disclaimer.
15 *
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials
19 * provided with the distribution.
20 *
21 * * Neither the name of the <ORGANIZATION> nor the names of its
22 * contributors may be used to endorse or promote products
23 * derived from this software without specific prior written
24 * permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
27 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
28 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
29 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
30 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
31 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
32 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
33 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
35 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
37 * OF THE POSSIBILITY OF SUCH DAMAGE.
38 */
39
40/* What encodings are enabled? */
41#include <uriparser/UriDefsConfig.h>
42#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE))
43/* Include SELF twice */
44# ifdef URI_ENABLE_ANSI
45# define URI_PASS_ANSI 1
46# include "UriShorten.c"
47# undef URI_PASS_ANSI
48# endif
49# ifdef URI_ENABLE_UNICODE
50# define URI_PASS_UNICODE 1
51# include "UriShorten.c"
52# undef URI_PASS_UNICODE
53# endif
54#else
55# ifdef URI_PASS_ANSI
56# include <uriparser/UriDefsAnsi.h>
57# else
58# include <uriparser/UriDefsUnicode.h>
59# include <wchar.h>
60# endif
61
62
63
64#ifndef URI_DOXYGEN
65# include <uriparser/Uri.h>
66# include "UriCommon.h"
67#endif
68
69
70
71static URI_INLINE UriBool URI_FUNC(AppendSegment)(URI_TYPE(Uri) * uri,
72 const URI_CHAR * first, const URI_CHAR * afterLast) {
73 /* Create segment */
74 URI_TYPE(PathSegment) * segment = malloc(1 * sizeof(URI_TYPE(PathSegment)));
75 if (segment == NULL) {
76 return URI_FALSE; /* Raises malloc error */
77 }
78 segment->next = NULL;
79 segment->text.first = first;
80 segment->text.afterLast = afterLast;
81
82 /* Put into chain */
83 if (uri->pathTail == NULL) {
84 uri->pathHead = segment;
85 } else {
86 uri->pathTail->next = segment;
87 }
88 uri->pathTail = segment;
89
90 return URI_TRUE;
91}
92
93
94
95static URI_INLINE UriBool URI_FUNC(EqualsAuthority)(const URI_TYPE(Uri) * first,
96 const URI_TYPE(Uri) * second) {
97 /* IPv4 */
98 if (first->hostData.ip4 != NULL) {
99 return ((second->hostData.ip4 != NULL)
100 && !memcmp(first->hostData.ip4->data,
101 second->hostData.ip4->data, 4)) ? URI_TRUE : URI_FALSE;
102 }
103
104 /* IPv6 */
105 if (first->hostData.ip6 != NULL) {
106 return ((second->hostData.ip6 != NULL)
107 && !memcmp(first->hostData.ip6->data,
108 second->hostData.ip6->data, 16)) ? URI_TRUE : URI_FALSE;
109 }
110
111 /* IPvFuture */
112 if (first->hostData.ipFuture.first != NULL) {
113 return ((second->hostData.ipFuture.first != NULL)
114 && !URI_STRNCMP(first->hostData.ipFuture.first,
115 second->hostData.ipFuture.first,
116 first->hostData.ipFuture.afterLast
117 - first->hostData.ipFuture.first))
118 ? URI_TRUE : URI_FALSE;
119 }
120
121 if (first->hostText.first != NULL) {
122 return ((second->hostText.first != NULL)
123 && !URI_STRNCMP(first->hostText.first,
124 second->hostText.first,
125 first->hostText.afterLast
126 - first->hostText.first)) ? URI_TRUE : URI_FALSE;
127 }
128
129 return (second->hostText.first == NULL);
130}
131
132
133
134int URI_FUNC(RemoveBaseUriImpl)(URI_TYPE(Uri) * dest,
135 const URI_TYPE(Uri) * absSource,
136 const URI_TYPE(Uri) * absBase,
137 UriBool domainRootMode) {
138 if (dest == NULL) {
139 return URI_ERROR_NULL;
140 }
141 URI_FUNC(ResetUri)(dest);
142
143 if ((absSource == NULL) || (absBase == NULL)) {
144 return URI_ERROR_NULL;
145 }
146
147 /* absBase absolute? */
148 if (absBase->scheme.first == NULL) {
149 return URI_ERROR_REMOVEBASE_REL_BASE;
150 }
151
152 /* absSource absolute? */
153 if (absSource->scheme.first == NULL) {
154 return URI_ERROR_REMOVEBASE_REL_SOURCE;
155 }
156
157 /* [01/50] if (A.scheme != Base.scheme) then */
158 if (URI_STRNCMP(absSource->scheme.first, absBase->scheme.first,
159 absSource->scheme.afterLast - absSource->scheme.first)) {
160 /* [02/50] T.scheme = A.scheme; */
161 dest->scheme = absSource->scheme;
162 /* [03/50] T.authority = A.authority; */
163 if (!URI_FUNC(CopyAuthority)(dest, absSource)) {
164 return URI_ERROR_MALLOC;
165 }
166 /* [04/50] T.path = A.path; */
167 if (!URI_FUNC(CopyPath)(dest, absSource)) {
168 return URI_ERROR_MALLOC;
169 }
170 /* [05/50] else */
171 } else {
172 /* [06/50] undef(T.scheme); */
173 /* NOOP */
174 /* [07/50] if (A.authority != Base.authority) then */
175 if (!URI_FUNC(EqualsAuthority)(absSource, absBase)) {
176 /* [08/50] T.authority = A.authority; */
177 if (!URI_FUNC(CopyAuthority)(dest, absSource)) {
178 return URI_ERROR_MALLOC;
179 }
180 /* [09/50] T.path = A.path; */
181 if (!URI_FUNC(CopyPath)(dest, absSource)) {
182 return URI_ERROR_MALLOC;
183 }
184 /* [10/50] else */
185 } else {
186 /* [11/50] if domainRootMode then */
187 if (domainRootMode == URI_TRUE) {
188 /* [12/50] undef(T.authority); */
189 /* NOOP */
190 /* [13/50] if (first(A.path) == "") then */
191 /* GROUPED */
192 /* [14/50] T.path = "/." + A.path; */
193 /* GROUPED */
194 /* [15/50] else */
195 /* GROUPED */
196 /* [16/50] T.path = A.path; */
197 /* GROUPED */
198 /* [17/50] endif; */
199 if (!URI_FUNC(CopyPath)(dest, absSource)) {
200 return URI_ERROR_MALLOC;
201 }
202 dest->absolutePath = URI_TRUE;
203
204 if (!URI_FUNC(FixAmbiguity)(dest)) {
205 return URI_ERROR_MALLOC;
206 }
207 /* [18/50] else */
208 } else {
209 const URI_TYPE(PathSegment) * sourceSeg = absSource->pathHead;
210 const URI_TYPE(PathSegment) * baseSeg = absBase->pathHead;
211 /* [19/50] bool pathNaked = true; */
212 UriBool pathNaked = URI_TRUE;
213 /* [20/50] undef(last(Base.path)); */
214 /* NOOP */
215 /* [21/50] T.path = ""; */
216 dest->absolutePath = URI_FALSE;
217 /* [22/50] while (first(A.path) == first(Base.path)) do */
218 while ((sourceSeg != NULL) && (baseSeg != NULL)
219 && !URI_STRNCMP(sourceSeg->text.first, baseSeg->text.first,
220 sourceSeg->text.afterLast - sourceSeg->text.first)
221 && !((sourceSeg->text.first == sourceSeg->text.afterLast)
222 && ((sourceSeg->next == NULL) != (baseSeg->next == NULL)))) {
223 /* [23/50] A.path++; */
224 sourceSeg = sourceSeg->next;
225 /* [24/50] Base.path++; */
226 baseSeg = baseSeg->next;
227 /* [25/50] endwhile; */
228 }
229 /* [26/50] while defined(first(Base.path)) do */
230 while ((baseSeg != NULL) && (baseSeg->next != NULL)) {
231 /* [27/50] Base.path++; */
232 baseSeg = baseSeg->next;
233 /* [28/50] T.path += "../"; */
234 if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstParent),
235 URI_FUNC(ConstParent) + 2)) {
236 return URI_ERROR_MALLOC;
237 }
238 /* [29/50] pathNaked = false; */
239 pathNaked = URI_FALSE;
240 /* [30/50] endwhile; */
241 }
242 /* [31/50] while defined(first(A.path)) do */
243 while (sourceSeg != NULL) {
244 /* [32/50] if pathNaked then */
245 if (pathNaked == URI_TRUE) {
246 /* [33/50] if (first(A.path) contains ":") then */
247 UriBool containsColon = URI_FALSE;
248 const URI_CHAR * ch = sourceSeg->text.first;
249 for (; ch < sourceSeg->text.afterLast; ch++) {
250 if (*ch == _UT(':')) {
251 containsColon = URI_TRUE;
252 break;
253 }
254 }
255
256 if (containsColon) {
257 /* [34/50] T.path += "./"; */
258 if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstPwd),
259 URI_FUNC(ConstPwd) + 1)) {
260 return URI_ERROR_MALLOC;
261 }
262 /* [35/50] elseif (first(A.path) == "") then */
263 } else if (sourceSeg->text.first == sourceSeg->text.afterLast) {
264 /* [36/50] T.path += "/."; */
265 if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstPwd),
266 URI_FUNC(ConstPwd) + 1)) {
267 return URI_ERROR_MALLOC;
268 }
269 /* [37/50] endif; */
270 }
271 /* [38/50] endif; */
272 }
273 /* [39/50] T.path += first(A.path); */
274 if (!URI_FUNC(AppendSegment)(dest, sourceSeg->text.first,
275 sourceSeg->text.afterLast)) {
276 return URI_ERROR_MALLOC;
277 }
278 /* [40/50] pathNaked = false; */
279 pathNaked = URI_FALSE;
280 /* [41/50] A.path++; */
281 sourceSeg = sourceSeg->next;
282 /* [42/50] if defined(first(A.path)) then */
283 /* NOOP */
284 /* [43/50] T.path += + "/"; */
285 /* NOOP */
286 /* [44/50] endif; */
287 /* NOOP */
288 /* [45/50] endwhile; */
289 }
290 /* [46/50] endif; */
291 }
292 /* [47/50] endif; */
293 }
294 /* [48/50] endif; */
295 }
296 /* [49/50] T.query = A.query; */
297 dest->query = absSource->query;
298 /* [50/50] T.fragment = A.fragment; */
299 dest->fragment = absSource->fragment;
300
301 return URI_SUCCESS;
302}
303
304
305
306int URI_FUNC(RemoveBaseUri)(URI_TYPE(Uri) * dest,
307 const URI_TYPE(Uri) * absSource,
308 const URI_TYPE(Uri) * absBase,
309 UriBool domainRootMode) {
310 const int res = URI_FUNC(RemoveBaseUriImpl)(dest, absSource,
311 absBase, domainRootMode);
312 if ((res != URI_SUCCESS) && (dest != NULL)) {
313 URI_FUNC(FreeUriMembers)(dest);
314 }
315 return res;
316}
317
318
319
320#endif