[svn] Convert URLs in <form action=...>.

[wget] / src / url.c

diff --git a/src/url.c b/src/url.c
index ac9eba0f7c4711110b85c7096d6c0fb77f15aa41..6fb2418288f16f4bb4aef28732111a631b76d934 100644 (file)
--- a/src/url.c
+++ b/src/url.c
@@ -111,7 +111,7 @@ const static unsigned char urlchr_table[256] =
  RU,  0,  0,  0,   0,  0,  0,  0,   /* @   A   B   C    D   E   F   G   */
   0,  0,  0,  0,   0,  0,  0,  0,   /* H   I   J   K    L   M   N   O   */
   0,  0,  0,  0,   0,  0,  0,  0,   /* P   Q   R   S    T   U   V   W   */
-  0,  0,  0,  U,   U,  U,  U,  0,   /* X   Y   Z   [    \   ]   ^   _   */
+  0,  0,  0, RU,   U, RU,  U,  0,   /* X   Y   Z   [    \   ]   ^   _   */
   U,  0,  0,  0,   0,  0,  0,  0,   /* `   a   b   c    d   e   f   g   */
   0,  0,  0,  0,   0,  0,  0,  0,   /* h   i   j   k    l   m   n   o   */
   0,  0,  0,  0,   0,  0,  0,  0,   /* p   q   r   s    t   u   v   w   */
@@ -622,16 +622,20 @@ lowercase_str (char *str)
 }
 
 static char *parse_errors[] = {
-#define PE_NO_ERROR            0
+#define PE_NO_ERROR                    0
   "No error",
-#define PE_UNSUPPORTED_SCHEME 1
+#define PE_UNSUPPORTED_SCHEME          1
   "Unsupported scheme",
-#define PE_EMPTY_HOST          2
+#define PE_EMPTY_HOST                  2
   "Empty host",
-#define PE_BAD_PORT_NUMBER     3
+#define PE_BAD_PORT_NUMBER             3
   "Bad port number",
-#define PE_INVALID_USER_NAME   4
-  "Invalid user name"
+#define PE_INVALID_USER_NAME           4
+  "Invalid user name",
+#define PE_UNTERMINATED_IPV6_ADDRESS   5
+  "Unterminated IPv6 numeric address",
+#define PE_INVALID_IPV6_ADDRESS                6
+  "Invalid char in IPv6 numeric address"
 };
 
 #define SETERR(p, v) do {                      \
@@ -693,8 +697,45 @@ url_parse (const char *url, int *error)
   fragment_b = fragment_e = NULL;
 
   host_b = p;
-  p = strpbrk_or_eos (p, ":/;?#");
-  host_e = p;
+
+  if (*p == '[')
+    {
+      /* Support http://[::1]/ used by IPv6. */
+      int invalid = 0;
+      ++p;
+      while (1)
+       {
+         char c = *p++;
+         switch (c)
+           {
+           case ']':
+             goto out;
+           case '\0':
+             SETERR (error, PE_UNTERMINATED_IPV6_ADDRESS);
+             return NULL;
+           case ':': case '.':
+             break;
+           default:
+             if (ISXDIGIT (c))
+               break;
+             invalid = 1;
+           }
+       }
+    out:
+      if (invalid)
+       {
+         SETERR (error, PE_INVALID_IPV6_ADDRESS);
+         return NULL;
+       }
+      /* Don't include brackets in [host_b, host_p). */
+      ++host_b;
+      host_e = p - 1;
+    }
+  else
+    {
+      p = strpbrk_or_eos (p, ":/;?#");
+      host_e = p;
+    }
 
   if (host_b == host_e)
     {
@@ -761,6 +802,15 @@ url_parse (const char *url, int *error)
       query_b = p;
       p = strpbrk_or_eos (p, "#");
       query_e = p;
+
+      /* Hack that allows users to use '?' (a wildcard character) in
+        FTP URLs without it being interpreted as a query string
+        delimiter.  */
+      if (scheme == SCHEME_FTP)
+       {
+         query_b = query_e = NULL;
+         path_e = p;
+       }
     }
   if (*p == '#')
     {
@@ -1178,9 +1228,8 @@ count_slashes (const char *s)
 static char *
 mkstruct (const struct url *u)
 {
-  char *dir, *dir_preencoding;
-  char *file, *res, *dirpref;
-  char *query = u->query && *u->query ? u->query : NULL;
+  char *dir, *file;
+  char *res, *dirpref;
   int l;
 
   if (opt.cut_dirs)
@@ -1234,9 +1283,6 @@ mkstruct (const struct url *u)
       dir = newdir;
     }
 
-  dir_preencoding = dir;
-  dir = reencode_string (dir_preencoding);
-
   l = strlen (dir);
   if (l && dir[l - 1] == '/')
     dir[l - 1] = '\0';
@@ -1248,16 +1294,9 @@ mkstruct (const struct url *u)
 
   /* Finally, construct the full name.  */
   res = (char *)xmalloc (strlen (dir) + 1 + strlen (file)
-                        + (query ? (1 + strlen (query)) : 0)
                         + 1);
   sprintf (res, "%s%s%s", dir, *dir ? "/" : "", file);
-  if (query)
-    {
-      strcat (res, "?");
-      strcat (res, query);
-    }
-  if (dir != dir_preencoding)
-    xfree (dir);
+
   return res;
 }
 
@@ -1324,26 +1363,25 @@ char *
 url_filename (const struct url *u)
 {
   char *file, *name;
-  int have_prefix = 0;         /* whether we must prepend opt.dir_prefix */
+
+  char *query = u->query && *u->query ? u->query : NULL;
 
   if (opt.dirstruct)
     {
-      file = mkstruct (u);
-      have_prefix = 1;
+      char *base = mkstruct (u);
+      file = compose_file_name (base, query);
+      xfree (base);
     }
   else
     {
       char *base = *u->file ? u->file : "index.html";
-      char *query = u->query && *u->query ? u->query : NULL;
       file = compose_file_name (base, query);
-    }
 
-  if (!have_prefix)
-    {
       /* Check whether the prefix directory is something other than "."
         before prepending it.  */
       if (!DOTP (opt.dir_prefix))
        {
+         /* #### should just realloc FILE and prepend dir_prefix. */
          char *nfile = (char *)xmalloc (strlen (opt.dir_prefix)
                                         + 1 + strlen (file) + 1);
          sprintf (nfile, "%s/%s", opt.dir_prefix, file);
@@ -1351,6 +1389,7 @@ url_filename (const struct url *u)
          file = nfile;
        }
     }
+
   /* DOS-ish file systems don't like `%' signs in them; we change it
      to `@'.  */
 #ifdef WINDOWS
@@ -1770,6 +1809,8 @@ url_string (const struct url *url, int hide_password)
   char *scheme_str = supported_schemes[url->scheme].leading_string;
   int fplen = full_path_length (url);
 
+  int brackets_around_host = 0;
+
   assert (scheme_str != NULL);
 
   /* Make sure the user name and password are quoted. */
@@ -1785,8 +1826,12 @@ url_string (const struct url *url, int hide_password)
        }
     }
 
+  if (strchr (url->host, ':'))
+    brackets_around_host = 1;
+
   size = (strlen (scheme_str)
          + strlen (url->host)
+         + (brackets_around_host ? 2 : 0)
          + fplen
          + 1);
   if (url->port != scheme_port)
@@ -1812,7 +1857,11 @@ url_string (const struct url *url, int hide_password)
       *p++ = '@';
     }
 
+  if (brackets_around_host)
+    *p++ = '[';
   APPEND (p, url->host);
+  if (brackets_around_host)
+    *p++ = ']';
   if (url->port != scheme_port)
     {
       *p++ = ':';