Follow RFC 2616 and httpbis specifications when handling redirects

[wget] / src / html-url.c

diff --git a/src/html-url.c b/src/html-url.c
index 54a0141c4533672c1c8c5de70d60acb6c6fa8d61..f02982173cec0302e0295e68aaecac5abad5942d 100644 (file)
--- a/src/html-url.c
+++ b/src/html-url.c
@@ -1,6 +1,6 @@
 /* Collect URLs from HTML source.
    Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
-   2007, 2008, 2009, 2010 Free Software Foundation, Inc.
+   2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation, Inc.
 
 This file is part of GNU Wget.
 
@@ -36,6 +36,7 @@ as that of the covered work.  */
 #include <errno.h>
 #include <assert.h>
 
+#include "exits.h"
 #include "html-parse.h"
 #include "url.h"
 #include "utils.h"
@@ -164,6 +165,7 @@ static struct {
    to the attributes not mentioned here.  We add them manually.  */
 static const char *additional_attributes[] = {
   "rel",                        /* used by tag_handle_link  */
+  "type",                       /* used by tag_handle_link  */
   "http-equiv",                 /* used by tag_handle_meta  */
   "name",                       /* used by tag_handle_meta  */
   "content",                    /* used by tag_handle_meta  */
@@ -505,8 +507,8 @@ tag_handle_link (int tagid, struct taginfo *tag, struct map_context *ctx)
   /* All <link href="..."> link references are external, except those
      known not to be, such as style sheet and shortcut icon:
 
-       <link rel="stylesheet" href="...">
-       <link rel="shortcut icon" href="...">
+     <link rel="stylesheet" href="...">
+     <link rel="shortcut icon" href="...">
   */
   if (href)
     {
@@ -526,11 +528,18 @@ tag_handle_link (int tagid, struct taginfo *tag, struct map_context *ctx)
                 {
                   up->link_inline_p = 1;
                 }
+              else
+                {
+                  /* The external ones usually point to HTML pages, such as
+                     <link rel="next" href="...">
+                     except when the type attribute says otherwise:
+                     <link rel="alternate" type="application/rss+xml" href=".../?feed=rss2" />
+                  */
+                  char *type = find_attr (tag, "type", NULL);
+                  if (!type || strcasecmp (type, "text/html") == 0)
+                    up->link_expect_html = 1;
+                }
             }
-          else
-            /* The external ones usually point to HTML pages, such as
-               <link rel="next" href="..."> */
-            up->link_expect_html = 1;
         }
     }
 }
@@ -666,8 +675,9 @@ collect_tags_mapper (struct taginfo *tag, void *arg)
 
   check_style_attr (tag, ctx);
 
-  if (tag->end_tag_p && (0 == strcasecmp (tag->name, "style")) &&
-      tag->contents_begin && tag->contents_end)
+  if (tag->end_tag_p && (0 == strcasecmp (tag->name, "style"))
+      && tag->contents_begin && tag->contents_end
+      && tag->contents_begin <= tag->contents_end)
   {
     /* parse contents */
     get_urls_css (ctx, tag->contents_begin - ctx->text,
@@ -794,6 +804,13 @@ get_urls_file (const char *file)
           url_text = merged;
         }
 
+      char *new_url = rewrite_shorthand_url (url_text);
+      if (new_url)
+        {
+          xfree (url_text);
+          url_text = new_url;
+        }
+
       url = url_parse (url_text, &up_error_code, NULL, false);
       if (!url)
         {
@@ -802,6 +819,7 @@ get_urls_file (const char *file)
                      file, url_text, error);
           xfree (url_text);
           xfree (error);
+          inform_exit_status (URLERROR);
           continue;
         }
       xfree (url_text);