[svn] Correctly convert links in <meta http-equiv=Refresh content="...">.

Published in <sxsadx3wp49.fsf@florida.arsdigita.de>.

diff --git a/src/ChangeLog b/src/ChangeLog
index 9d1bc81b359de7aeae81db5a841791a7abdf3c99..45f06eabfc4af58ddfaa2a760728a31b0a18de7e 100644 (file)
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,3 +1,15 @@
+2001-12-01  Hrvoje Niksic  <hniksic@arsdigita.com>
+
+       * url.c (replace_attr_refresh_hack): New function.
+       (convert_links): Call replace_attr_refresh_hack for Refresh
+       links.  It will add the "TMOUT; URL=" junk before the link.
+
+       * html-url.c (collect_tags_mapper): Set ID to the ID of the
+       "content" attribute, not "http-equiv".
+       (collect_tags_mapper): Don't use OFFSET to hack the raw_* values;
+       instead, store the information that this entry belongs to a
+       "refresh" link.
+
 2001-12-01  Hrvoje Niksic  <hniksic@arsdigita.com>
 
        * recur.c (retrieve_tree): Allow -p retrievals to exceed maximum

diff --git a/src/html-url.c b/src/html-url.c
index 5942a49f61e8d8c0358f646ee42ee88ad817b5af..7b9bf29a22cb145457cb6188dbde3697adb75f1f 100644 (file)
--- a/src/html-url.c
+++ b/src/html-url.c
@@ -482,18 +482,21 @@ collect_tags_mapper (struct taginfo *tag, void *arg)
             So we just need to skip past the "NUMBER; URL=" garbage
             to get to the URL.  */
          {
-           int id;
            char *name = find_attr (tag, "name", NULL);
-           char *http_equiv = find_attr (tag, "http-equiv", &id);
+           char *http_equiv = find_attr (tag, "http-equiv", NULL);
            if (http_equiv && !strcasecmp (http_equiv, "refresh"))
              {
-               char *refresh = find_attr (tag, "content", NULL);
-               char *p = refresh;
-               int offset;
-               while (ISDIGIT (*p))
-                 ++p;
+               struct urlpos *entry;
+
+               int id;
+               char *p, *refresh = find_attr (tag, "content", &id);
+               int timeout = 0;
+
+               for (p = refresh; ISDIGIT (*p); p++)
+                 timeout = 10 * timeout + *p - '0';
                if (*p++ != ';')
                  return;
+
                while (ISSPACE (*p))
                  ++p;
                if (!(TOUPPER (*p) == 'U'
@@ -504,10 +507,13 @@ collect_tags_mapper (struct taginfo *tag, void *arg)
                p += 4;
                while (ISSPACE (*p))
                  ++p;
-               offset = p - refresh;
-               tag->attrs[id].value_raw_beginning += offset;
-               tag->attrs[id].value_raw_size -= offset;
-               handle_link (closure, p, tag, id);
+
+               entry = handle_link (closure, p, tag, id);
+               if (entry)
+                 {
+                   entry->link_refresh_p = 1;
+                   entry->refresh_timeout = timeout;
+                 }
              }
            else if (name && !strcasecmp (name, "robots"))
              {

diff --git a/src/url.c b/src/url.c
index 28e41c8cc740ac7881d64290bde75ee86654138b..38469418abb6954081c79528b4ae546b7815688c 100644 (file)
--- a/src/url.c
+++ b/src/url.c
@@ -1698,7 +1698,10 @@ no_proxy_match (const char *host, const char **no_proxy)
 }
 \f
 static void write_backup_file PARAMS ((const char *, downloaded_file_t));
-static const char *replace_attr PARAMS ((const char *, int, FILE *, const char *));
+static const char *replace_attr PARAMS ((const char *, int, FILE *,
+                                        const char *));
+static const char *replace_attr_refresh_hack PARAMS ((const char *, int, FILE *,
+                                                     const char *, int));
 static char *local_quote_string PARAMS ((const char *));
 
 /* Change the links in one HTML file.  LINKS is a list of links in the
@@ -1797,7 +1800,13 @@ convert_links (const char *file, struct urlpos *links)
          {
            char *newname = construct_relative (file, link->local_name);
            char *quoted_newname = local_quote_string (newname);
-           p = replace_attr (p, link->size, fp, quoted_newname);
+
+           if (!link->link_refresh_p)
+             p = replace_attr (p, link->size, fp, quoted_newname);
+           else
+             p = replace_attr_refresh_hack (p, link->size, fp, quoted_newname,
+                                            link->refresh_timeout);
+
            DEBUGP (("TO_RELATIVE: %s to %s at position %d in %s.\n",
                     link->url->url, newname, link->pos, file));
            xfree (newname);
@@ -1810,7 +1819,13 @@ convert_links (const char *file, struct urlpos *links)
          {
            char *newlink = link->url->url;
            char *quoted_newlink = html_quote_string (newlink);
-           p = replace_attr (p, link->size, fp, quoted_newlink);
+
+           if (!link->link_refresh_p)
+             p = replace_attr (p, link->size, fp, quoted_newlink);
+           else
+             p = replace_attr_refresh_hack (p, link->size, fp, quoted_newlink,
+                                            link->refresh_timeout);
+
            DEBUGP (("TO_COMPLETE: <something> to %s at position %d in %s.\n",
                     newlink, link->pos, file));
            xfree (quoted_newlink);
@@ -2014,6 +2029,24 @@ replace_attr (const char *p, int size, FILE *fp, const char *new_text)
   return p;
 }
 
+/* The same as REPLACE_ATTR, but used when replacing
+   <meta http-equiv=refresh content="new_text"> because we need to
+   append "timeout_value; URL=" before the next_text.  */
+
+static const char *
+replace_attr_refresh_hack (const char *p, int size, FILE *fp,
+                          const char *new_text, int timeout)
+{
+  /* "0; URL=..." */
+  char *new_with_timeout = (char *)alloca (numdigit (timeout)
+                                          + 6 /* "; URL=" */
+                                          + strlen (new_text)
+                                          + 1);
+  sprintf (new_with_timeout, "%d; URL=%s", timeout, new_text);
+
+  return replace_attr (p, size, fp, new_with_timeout);
+}
+
 /* Find the first occurrence of '#' in [BEG, BEG+SIZE) that is not
    preceded by '&'.  If the character is not found, return zero.  If
    the character is found, return 1 and set BP and EP to point to the

diff --git a/src/url.h b/src/url.h
index 3c42c5aa1c661199970dd2b1b7f2d450bdc996b4..b99c64cef84a9bfe9297d68dbc3c8a915a05e572 100644 (file)
--- a/src/url.h
+++ b/src/url.h
@@ -91,6 +91,10 @@ struct urlpos {
   unsigned int link_base_p     :1; /* was the link <base href=...> */
   unsigned int link_inline_p   :1; /* needed to render the page. */
 
+  unsigned int link_refresh_p  :1; /* link was received from
+                                      <meta http-equiv=refresh content=...> */
+  int refresh_timeout;         /* for reconstructing the refresh. */
+
   /* Conversion requirements: */
   enum convert_options convert;        /* is conversion required? */