shall include the source code for the parts of OpenSSL used as well
as that of the covered work. */
-#define USE_GNULIB_ALLOC
-
#include "wget.h"
#include <stdio.h>
#include "hash.h"
#include "ptimer.h"
#include "res.h"
+#include "html-url.h"
+#include "css-url.h"
static struct hash_table *dl_file_url_map;
struct hash_table *dl_url_file_map;
-/* Set of HTML files downloaded in this Wget run, used for link
+/* Set of HTML/CSS files downloaded in this Wget run, used for link
conversion after Wget is done. */
struct hash_table *downloaded_html_set;
+struct hash_table *downloaded_css_set;
static void convert_links (const char *, struct urlpos *);
-/* This function is called when the retrieval is done to convert the
- links that have been downloaded. It has to be called at the end of
- the retrieval, because only then does Wget know conclusively which
- URLs have been downloaded, and which not, so it can tell which
- direction to convert to.
-
- The "direction" means that the URLs to the files that have been
- downloaded get converted to the relative URL which will point to
- that file. And the other URLs get converted to the remote URL on
- the server.
-
- All the downloaded HTMLs are kept in downloaded_html_files, and
- downloaded URLs in urls_downloaded. All the information is
- extracted from these two lists. */
void
-convert_all_links (void)
+convert_links_in_hashtable (struct hash_table *downloaded_set,
+ int is_css,
+ int *file_count)
{
int i;
- double secs;
- int file_count = 0;
-
- struct ptimer *timer = ptimer_new ();
int cnt;
char **file_array;
cnt = 0;
- if (downloaded_html_set)
- cnt = hash_table_count (downloaded_html_set);
+ if (downloaded_set)
+ cnt = hash_table_count (downloaded_set);
if (cnt == 0)
- goto cleanup;
- file_array = xcalloc (cnt, sizeof (char *));
- string_set_to_array (downloaded_html_set, file_array);
+ return;
+ file_array = alloca_array (char *, cnt);
+ string_set_to_array (downloaded_set, file_array);
for (i = 0; i < cnt; i++)
{
char *url;
char *file = file_array[i];
- /* Determine the URL of the HTML file. get_urls_html will need
+ /* Determine the URL of the file. get_urls_{html,css} will need
it. */
url = hash_table_get (dl_file_url_map, file);
if (!url)
DEBUGP (("Scanning %s (from %s)\n", file, url));
- /* Parse the HTML file... */
- urls = get_urls_html (file, url, NULL);
+ /* Parse the file... */
+ urls = is_css ? get_urls_css_file (file, url) :
+ get_urls_html (file, url, NULL, NULL);
/* We don't respect meta_disallow_follow here because, even if
the file is not followed, we might still want to convert the
/* Convert the links in the file. */
convert_links (file, urls);
- ++file_count;
+ ++*file_count;
/* Free the data. */
free_urlpos (urls);
}
+}
+
+/* This function is called when the retrieval is done to convert the
+ links that have been downloaded. It has to be called at the end of
+ the retrieval, because only then does Wget know conclusively which
+ URLs have been downloaded, and which not, so it can tell which
+ direction to convert to.
+
+ The "direction" means that the URLs to the files that have been
+ downloaded get converted to the relative URL which will point to
+ that file. And the other URLs get converted to the remote URL on
+ the server.
+
+ All the downloaded HTMLs are kept in downloaded_html_files, and
+ downloaded URLs in urls_downloaded. All the information is
+ extracted from these two lists. */
+
+void
+convert_all_links (void)
+{
+ double secs;
+ int file_count = 0;
+
+ struct ptimer *timer = ptimer_new ();
+
+ convert_links_in_hashtable (downloaded_html_set, 0, &file_count);
+ convert_links_in_hashtable (downloaded_css_set, 1, &file_count);
secs = ptimer_measure (timer);
logprintf (LOG_VERBOSE, _("Converted %d files in %s seconds.\n"),
file_count, print_decimal (secs));
-cleanup:
+
ptimer_destroy (timer);
}
static void write_backup_file (const char *, downloaded_file_t);
+static const char *replace_plain (const char*, int, FILE*, const char *);
static const char *replace_attr (const char *, int, FILE *, const char *);
static const char *replace_attr_refresh_hack (const char *, int, FILE *,
const char *, int);
static char *local_quote_string (const char *);
static char *construct_relative (const char *, const char *);
-/* Change the links in one HTML file. LINKS is a list of links in the
+/* Change the links in one file. LINKS is a list of links in the
document, along with their positions and the desired direction of
the conversion. */
static void
char *newname = construct_relative (file, link->local_name);
char *quoted_newname = local_quote_string (newname);
- if (!link->link_refresh_p)
+ if (link->link_css_p)
+ p = replace_plain (p, link->size, fp, quoted_newname);
+ else 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,
DEBUGP (("TO_RELATIVE: %s to %s at position %d in %s.\n",
link->url->url, newname, link->pos, file));
- free (newname);
- free (quoted_newname);
+ xfree (newname);
+ xfree (quoted_newname);
++to_file_count;
break;
}
char *newlink = link->url->url;
char *quoted_newlink = html_quote_string (newlink);
- if (!link->link_refresh_p)
+ if (link->link_css_p)
+ p = replace_plain (p, link->size, fp, quoted_newlink);
+ else 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,
DEBUGP (("TO_COMPLETE: <something> to %s at position %d in %s.\n",
newlink, link->pos, file));
- free (quoted_newlink);
+ xfree (quoted_newlink);
++to_url_count;
break;
}
/* Rather than just writing over the original .html file with the
converted version, save the former to *.orig. Note we only do
this for files we've _successfully_ downloaded, so we don't
- clobber .orig files sitting around from previous invocations. */
+ clobber .orig files sitting around from previous invocations.
+ On VMS, use "_orig" instead of ".orig". See "wget.h". */
/* Construct the backup filename as the original name plus ".orig". */
size_t filename_len = strlen (file);
char* filename_plus_orig_suffix;
+ /* TODO: hack this to work with css files */
if (downloaded_file_return == FILE_DOWNLOADED_AND_HTML_EXTENSION_ADDED)
{
/* Just write "orig" over "html". We need to do it this way
else /* downloaded_file_return == FILE_DOWNLOADED_NORMALLY */
{
/* Append ".orig" to the name. */
- filename_plus_orig_suffix = alloca (filename_len + sizeof (".orig"));
+ filename_plus_orig_suffix = alloca (filename_len + sizeof ("ORIG_SFX"));
strcpy (filename_plus_orig_suffix, file);
- strcpy (filename_plus_orig_suffix + filename_len, ".orig");
+ strcpy (filename_plus_orig_suffix + filename_len, "ORIG_SFX");
}
if (!converted_files)
static bool find_fragment (const char *, int, const char **, const char **);
+/* Replace a string with NEW_TEXT. Ignore quoting. */
+static const char *
+replace_plain (const char *p, int size, FILE *fp, const char *new_text)
+{
+ fputs (new_text, fp);
+ p += size;
+ return p;
+}
+
/* Replace an attribute's original text with NEW_TEXT. */
static const char *
if (0 == strcmp (mapping_file, file))
{
hash_table_remove (dl_url_file_map, mapping_url);
- free (mapping_url);
- free (mapping_file);
+ xfree (mapping_url);
+ xfree (mapping_file);
}
/* Continue mapping. */
goto url_only;
hash_table_remove (dl_file_url_map, file);
- free (old_file);
- free (old_url);
+ xfree (old_file);
+ xfree (old_url);
/* Remove all the URLs that point to this file. Yes, there can
be more than one such URL, because we store redirections as
if (hash_table_get_pair (dl_url_file_map, url, &old_url, &old_file))
{
hash_table_remove (dl_url_file_map, url);
- free (old_url);
- free (old_file);
+ xfree (old_url);
+ xfree (old_file);
}
hash_table_put (dl_url_file_map, xstrdup (url), xstrdup (file));
return;
hash_table_remove (dl_file_url_map, file);
- free (old_file);
- free (old_url);
+ xfree (old_file);
+ xfree (old_url);
dissociate_urls_from_file (file);
}
string_set_add (downloaded_html_set, file);
}
+/* Register that FILE is a CSS file that has been downloaded. */
+
+void
+register_css (const char *url, const char *file)
+{
+ if (!downloaded_css_set)
+ downloaded_css_set = make_string_hash_table (0);
+ string_set_add (downloaded_css_set, file);
+}
+
static void downloaded_files_free (void);
/* Cleanup the data structures associated with this file. */
for (hash_table_iterate (downloaded_files_hash, &iter);
hash_table_iter_next (&iter);
)
- free (iter.key);
+ xfree (iter.key);
hash_table_destroy (downloaded_files_hash);
downloaded_files_hash = NULL;
}
projects / wget / blobdiff