moparisthebest/wget
1
0
Fork 0
mirror of https://github.com/moparisthebest/wget synced 2024-07-03 16:38:41 -04:00
Code Issues Releases Wiki Activity

[svn] Removed the visited_url function.

Browse Source
This commit is contained in:
micah 2007-08-29 21:52:16 -07:00
parent f4771e5299
commit fa8344734f
3 changed files with 8 additions and 71 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/ChangeLog
View File

@ -2,6 +2,12 @@
* openssl.c (ssl_init): Re un-const-ified the meth local
variable, to match current versions of openssl.
* spider.c: Removed visited_url function, as it may be very
inefficient.
(print_broken_links): Removed traversal of referrers, until such
time as a more efficient implementation can be written.
* spider.h: Replaced declaration of visited_url with an
empty-bodied, function-like macro.
2007-08-27 Gisle Vanem <giva@bgnett.no>

71
src/spider.c
View File

@ -40,7 +40,6 @@ so, delete this exception statement from your version. */
#include "res.h"
static struct hash_table *visited_urls_hash;
static struct hash_table *nonexisting_urls_set;
/* Cleanup the data structures associated with this file. */
@ -48,72 +47,10 @@ static struct hash_table *nonexisting_urls_set;
void
spider_cleanup (void)
{
if (visited_urls_hash)
{
free_keys_and_values (visited_urls_hash);
hash_table_destroy (visited_urls_hash);
visited_urls_hash = NULL;
}
if (nonexisting_urls_set)
string_set_free (nonexisting_urls_set);
}
/* Remembers visited files. */
struct url_list
{
char *url;
struct url_list *next;
};
static bool
in_url_list_p (const struct url_list *list, const char *url)
{
const struct url_list *ptr;
for (ptr = list; ptr; ptr = ptr->next)
{
/* str[case]cmp is inadequate for URL comparison */
if (ptr->url != NULL && are_urls_equal (url, ptr->url))
return true;
}
return false;
}
void
visited_url (const char *url, const char *referrer)
{
struct url_list *list;
/* Ignore robots.txt URLs */
if (is_robots_txt_url (url))
return;
if (!visited_urls_hash)
visited_urls_hash = make_string_hash_table (0);
list = hash_table_get (visited_urls_hash, url);
if (!list)
{
list = (struct url_list *) xnew0 (struct url_list);
list->url = referrer ? xstrdup (referrer) : NULL;
hash_table_put (visited_urls_hash, xstrdup (url), list);
}
else if (referrer && !in_url_list_p (list, referrer))
{
/* Append referrer at the end of the list */
struct url_list *newnode;
while (list->next)
list = list->next;
newnode = (struct url_list *) xnew0 (struct url_list);
newnode->url = xstrdup (referrer);
list->next = newnode;
}
}
/* Remembers broken links. */
void
nonexisting_url (const char *url)
@ -151,13 +88,7 @@ print_broken_links (void)
struct url_list *list;
const char *url = (const char *) iter.key;
logprintf (LOG_NOTQUIET, _("%s referred by:\n"), url);
for (list = (struct url_list *) hash_table_get (visited_urls_hash, url);
list; list = list->next)
{
logprintf (LOG_NOTQUIET, _(" %s\n"), list->url);
}
logprintf (LOG_NOTQUIET, _("%s\n"), url);
}
logputs (LOG_NOTQUIET, "\n");
}

2
src/spider.h
View File

@ -29,7 +29,7 @@ so, delete this exception statement from your version. */
#ifndef SPIDER_H
#define SPIDER_H
void visited_url (const char *, const char *);
#define visited_url(a,b)
void nonexisting_url (const char *);
void print_broken_links (void);