X-Git-Url: http://mj.ucw.cz/gitweb/?a=blobdiff_plain;ds=sidebyside;f=ucw%2Furl.c;h=8bdf581d5442596de8b465813af610929e637734;hb=0053f1a862cd0c2f359a4d12f0fdd35b17d18906;hp=75b96b063fc2b890a87a3d93fafdc0b3c6cafedb;hpb=c6c6c66f35c76a1e7dfec03c30af6b6c501357ff;p=libucw.git

diff --git a/ucw/url.c b/ucw/url.c
index 75b96b06..8bdf581d 100644
--- a/ucw/url.c
+++ b/ucw/url.c
@@ -7,18 +7,14 @@
  *	This software may be freely distributed and used according to the terms
  *	of the GNU Lesser General Public License.
  *
- *	The URL syntax corresponds to RFC 2396 with several exceptions:
- *
- *	   o  Escaping of special characters still follows RFC 1738.
- *	   o  Interpretation of path parameters follows RFC 1808.
- *
  *	XXX: The buffer handling in this module is really horrible, but it works.
  */
 
-#include "ucw/lib.h"
-#include "ucw/url.h"
-#include "ucw/chartype.h"
-#include "ucw/conf.h"
+#include <ucw/lib.h>
+#include <ucw/url.h>
+#include <ucw/chartype.h>
+#include <ucw/conf.h>
+#include <ucw/prime.h>
 
 #include <string.h>
 #include <stdlib.h>
@@ -27,21 +23,22 @@
 
 /* Configuration */
 
-static uns url_ignore_spaces;
-static uns url_ignore_underflow;
+static uint url_ignore_spaces;
+static uint url_ignore_underflow;
 static char *url_component_separators = "";
-static uns url_min_repeat_count = 0x7fffffff;
-static uns url_max_repeat_length = 0;
-static uns url_max_occurences = ~0U;
+static uint url_min_repeat_count = 0x7fffffff;
+static uint url_max_repeat_length = 0;
+static uint url_max_occurences = ~0U;
 
+#ifndef TEST
 static struct cf_section url_config = {
   CF_ITEMS {
-    CF_UNS("IgnoreSpaces", &url_ignore_spaces),
-    CF_UNS("IgnoreUnderflow", &url_ignore_underflow),
+    CF_UINT("IgnoreSpaces", &url_ignore_spaces),
+    CF_UINT("IgnoreUnderflow", &url_ignore_underflow),
     CF_STRING("ComponentSeparators", &url_component_separators),
-    CF_UNS("MinRepeatCount", &url_min_repeat_count),
-    CF_UNS("MaxRepeatLength", &url_max_repeat_length),
-    CF_UNS("MaxOccurences", &url_max_occurences),
+    CF_UINT("MinRepeatCount", &url_min_repeat_count),
+    CF_UINT("MaxRepeatLength", &url_max_repeat_length),
+    CF_UINT("MaxOccurences", &url_max_occurences),
     CF_END
   }
 };
@@ -50,11 +47,12 @@ static void CONSTRUCTOR url_init_config(void)
 {
   cf_declare_section("URL", &url_config, 0);
 }
+#endif
 
 /* Escaping and de-escaping */
 
-static uns
-enhex(uns x)
+static uint
+enhex(uint x)
 {
   return (x<10) ? (x + '0') : (x - 10 + 'A');
 }
@@ -70,7 +68,7 @@ url_deescape(const char *s, char *d)
 	return URL_ERR_TOO_LONG;
       if (*s == '%')
 	{
-	  unsigned int val;
+	  uint val;
 	  if (!Cxdigit(s[1]) || !Cxdigit(s[2]))
 	    return URL_ERR_INVALID_ESCAPE;
 	  val = Cxvalue(s[1])*16 + Cxvalue(s[2]);
@@ -94,6 +92,12 @@ url_deescape(const char *s, char *d)
 	      val = NCC_AND; break;
 	    case '#':
 	      val = NCC_HASH; break;
+	    case '$':
+	      val = NCC_DOLLAR; break;
+	    case '+':
+	      val = NCC_PLUS; break;
+	    case ',':
+	      val = NCC_COMMA; break;
 	    }
 	  *d++ = val;
 	  s += 3;
@@ -126,22 +130,22 @@ int
 url_enescape(const char *s, char *d)
 {
   char *end = d + MAX_URL_SIZE - 10;
-  unsigned int c;
+  uint c;
 
   while (c = *s)
     {
       if (d >= end)
 	return URL_ERR_TOO_LONG;
-      if (Calnum(c) ||							/* RFC 1738(2.2): Only alphanumerics ... */
-	  c == '$' || c == '-' || c == '_' || c == '.' || c == '+' ||	/* ... and several other exceptions ... */
-	  c == '!' || c == '*' || c == '\'' || c == '(' || c == ')' ||
-	  c == ',' ||
-	  c == '/' || c == '?' || c == ':' || c == '@' ||		/* ... and reserved chars used for reserved purpose */
-	  c == '=' || c == '&' || c == '#' || c == ';')
+      if (Calnum(c) ||							/* RFC 2396 (2.1-2.3): Only alphanumerics ... */
+	  c == '!' || c == '*' || c == '\'' || c == '(' || c == ')' ||	/* ... and some exceptions and reserved chars */
+	  c == '$' || c == '-' || c == '_' || c == '.' || c == '+' ||
+	  c == ',' || c == '=' || c == '&' || c == '#' || c == ';' ||
+	  c == '/' || c == '?' || c == ':' || c == '@' || c == '~'
+	)
 	*d++ = *s++;
       else
 	{
-	  uns val = ((byte)*s < NCC_MAX) ? NCC_CHARS[(byte)*s] : *s;
+	  uint val = (byte)(((byte)*s < NCC_MAX) ? NCC_CHARS[(byte)*s] : *s);
 	  *d++ = '%';
 	  *d++ = enhex(val >> 4);
 	  *d++ = enhex(val & 0x0f);
@@ -161,14 +165,14 @@ url_enescape_friendly(const char *src, char *dest)
     {
       if (dest >= end)
 	return URL_ERR_TOO_LONG;
-      if (*srcb < NCC_MAX)
+      if ((byte)*srcb < NCC_MAX)
 	*dest++ = NCC_CHARS[*srcb++];
       else if (*srcb >= 0x20 && *srcb < 0x7f)
 	*dest++ = *srcb++;
       else
 	{
 	  *dest++ = '%';
-	  *dest++ = enhex(*srcb >> 4);
+	  *dest++ = enhex((byte)*srcb >> 4);
 	  *dest++ = enhex(*srcb++ & 0x0f);
 	}
     }
@@ -181,10 +185,10 @@ url_enescape_friendly(const char *src, char *dest)
 char *url_proto_names[URL_PROTO_MAX] = URL_PNAMES;
 static int url_proto_path_flags[URL_PROTO_MAX] = URL_PATH_FLAGS;
 
-uns
-identify_protocol(const char *p)
+uint
+url_identify_protocol(const char *p)
 {
-  uns i;
+  uint i;
 
   for(i=1; i<URL_PROTO_MAX; i++)
     if (!strcasecmp(p, url_proto_names[i]))
@@ -210,7 +214,7 @@ url_split(char *s, struct url *u, char *d)
 	  while (s < p)
 	    *d++ = *s++;
 	  *d++ = 0;
-	  u->protoid = identify_protocol(u->protocol);
+	  u->protoid = url_identify_protocol(u->protocol);
 	  s++;
 	  if (url_proto_path_flags[u->protoid] && (s[0] != '/' || s[1] != '/'))
 	    {
@@ -262,7 +266,7 @@ url_split(char *s, struct url *u, char *d)
 	  e = strchr(at, ':');
 	  if (e)			/* host:port present */
 	    {
-	      uns p;
+	      uint p;
 	      *e++ = 0;
 	      p = strtoul(e, &ep, 10);
 	      if (ep && *ep || p > 65535)
@@ -281,7 +285,7 @@ url_split(char *s, struct url *u, char *d)
 
 /* Normalization according to given base URL */
 
-static uns std_ports[] = URL_DEFPORTS;	/* Default port numbers */
+static uint std_ports[] = URL_DEFPORTS;	/* Default port numbers */
 
 static int
 relpath_merge(struct url *u, struct url *b)
@@ -317,15 +321,9 @@ relpath_merge(struct url *u, struct url *b)
 	;
       goto copy;
     }
-  if (a[0] == ';')			/* Change parameters */
-    {
-      for(p=o; *p && *p != ';' && *p != '?' && *p != '#'; p++)
-	;
-      goto copy;
-    }
 
   p = NULL;				/* Copy original path and find the last slash */
-  while (*o && *o != ';' && *o != '?' && *o != '#')
+  while (*o && *o != '?' && *o != '#')
     {
       if (d >= e)
 	return URL_ERR_TOO_LONG;
@@ -516,7 +514,7 @@ url_pack(struct url *u, char *d)
     {
       d = append(d, u->protocol, e);
       d = append(d, ":", e);
-      u->protoid = identify_protocol(u->protocol);
+      u->protoid = url_identify_protocol(u->protocol);
     }
   if (u->host)
     {
@@ -564,7 +562,7 @@ static char *errmsg[] = {
 };
 
 char *
-url_error(uns err)
+url_error(uint err)
 {
   if (err >= sizeof(errmsg) / sizeof(char *))
     err = 0;
@@ -609,7 +607,7 @@ int main(int argc, char **argv)
   char buf1[MAX_URL_SIZE], buf2[MAX_URL_SIZE], buf3[MAX_URL_SIZE], buf4[MAX_URL_SIZE];
   int err;
   struct url url, url0;
-  char *base = "http://mj@www.hell.org/123/sub_dir/index.html;param?query&zzz/subquery#fragment";
+  char *base = "http://mj@www.hell.org/123/sub_dir;param/index.html;param?query&zzz/sub;query+#fragment?";
 
   if (argc != 2 && argc != 3)
     return 1;
@@ -670,7 +668,7 @@ int main(int argc, char **argv)
 struct component {
 	const char *start;
 	int length;
-	uns count;
+	uint count;
 	u32 hash;
 };
 
@@ -683,14 +681,14 @@ hashf(const char *start, int length)
 	return hf;
 }
 
-static inline uns
-repeat_count(struct component *comp, uns count, uns len)
+static inline uint
+repeat_count(struct component *comp, uint count, uint len)
 {
 	struct component *orig_comp = comp;
-	uns found = 0;
+	uint found = 0;
 	while (1)
 	{
-		uns i;
+		uint i;
 		comp += len;
 		count -= len;
 		found++;
@@ -708,9 +706,9 @@ int
 url_has_repeated_component(const char *url)
 {
 	struct component *comp;
-	uns comps, comp_len, rep_prefix, hash_size, *hash, *next;
+	uint comps, comp_len, rep_prefix, hash_size, *hash, *next;
 	const char *c;
-	uns i, j, k;
+	uint i, j, k;
 
 	for (comps=0, c=url; c; comps++)
 	{