/* Handling of recursive HTTP retrieving.
- Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003,
- 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
+ Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
+ 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation,
+ Inc.
This file is part of GNU Wget.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
-#ifdef HAVE_UNISTD_H
-# include <unistd.h>
-#endif /* HAVE_UNISTD_H */
+#include <unistd.h>
#include <errno.h>
#include <assert.h>
#include "html-url.h"
#include "css-url.h"
#include "spider.h"
-
+\f
/* Functions for maintaining the URL queue. */
struct queue_element {
int depth; /* the depth */
bool html_allowed; /* whether the document is allowed to
be treated as HTML. */
+ struct iri *iri; /* sXXXav */
bool css_allowed; /* whether the document is allowed to
be treated as CSS. */
struct queue_element *next; /* next element in queue */
into it. */
static void
-url_enqueue (struct url_queue *queue,
+url_enqueue (struct url_queue *queue, struct iri *i,
const char *url, const char *referer, int depth,
bool html_allowed, bool css_allowed)
{
struct queue_element *qel = xnew (struct queue_element);
+ qel->iri = i;
qel->url = url;
qel->referer = referer;
qel->depth = depth;
if (queue->count > queue->maxcount)
queue->maxcount = queue->count;
- DEBUGP (("Enqueuing %s at depth %d\n", url, depth));
+ DEBUGP (("Enqueuing %s at depth %d\n",
+ quotearg_n_style (0, escape_quoting_style, url), depth));
DEBUGP (("Queue count %d, maxcount %d.\n", queue->count, queue->maxcount));
+ if (i)
+ DEBUGP (("[IRI Enqueuing %s with %s\n", quote_n (0, url),
+ i->uri_encoding ? quote_n (1, i->uri_encoding) : "None"));
+
if (queue->tail)
queue->tail->next = qel;
queue->tail = qel;
succeeded, or false if the queue is empty. */
static bool
-url_dequeue (struct url_queue *queue,
+url_dequeue (struct url_queue *queue, struct iri **i,
const char **url, const char **referer, int *depth,
bool *html_allowed, bool *css_allowed)
{
if (!queue->head)
queue->tail = NULL;
+ *i = qel->iri;
*url = qel->url;
*referer = qel->referer;
*depth = qel->depth;
--queue->count;
- DEBUGP (("Dequeuing %s at depth %d\n", qel->url, qel->depth));
+ DEBUGP (("Dequeuing %s at depth %d\n",
+ quotearg_n_style (0, escape_quoting_style, qel->url), qel->depth));
DEBUGP (("Queue count %d, maxcount %d.\n", queue->count, queue->maxcount));
xfree (qel);
}
\f
static bool download_child_p (const struct urlpos *, struct url *, int,
- struct url *, struct hash_table *);
+ struct url *, struct hash_table *, struct iri *);
static bool descend_redirect_p (const char *, struct url *, int,
- struct url *, struct hash_table *);
+ struct url *, struct hash_table *, struct iri *);
/* Retrieve a part of the web beginning with START_URL. This used to
options, add it to the queue. */
uerr_t
-retrieve_tree (struct url *start_url_parsed)
+retrieve_tree (struct url *start_url_parsed, struct iri *pi)
{
uerr_t status = RETROK;
the queue, but haven't been downloaded yet. */
struct hash_table *blacklist;
+ struct iri *i = iri_new ();
+
+#define COPYSTR(x) (x) ? xstrdup(x) : NULL;
+ /* Duplicate pi struct if not NULL */
+ if (pi)
+ {
+ i->uri_encoding = COPYSTR (pi->uri_encoding);
+ i->content_encoding = COPYSTR (pi->content_encoding);
+ i->utf8_encode = pi->utf8_encode;
+ }
+ else
+ set_uri_encoding (i, opt.locale, true);
+#undef COPYSTR
+
queue = url_queue_new ();
blacklist = make_string_hash_table (0);
/* Enqueue the starting URL. Use start_url_parsed->url rather than
just URL so we enqueue the canonical form of the URL. */
- url_enqueue (queue, xstrdup (start_url_parsed->url), NULL, 0, true, false);
+ url_enqueue (queue, i, xstrdup (start_url_parsed->url), NULL, 0, true,
+ false);
string_set_add (blacklist, start_url_parsed->url);
while (1)
/* Get the next URL from the queue... */
- if (!url_dequeue (queue,
+ if (!url_dequeue (queue, (struct iri **) &i,
(const char **)&url, (const char **)&referer,
&depth, &html_allowed, &css_allowed))
break;
the second time. */
if (dl_url_file_map && hash_table_contains (dl_url_file_map, url))
{
+ bool is_css_bool;
+
file = xstrdup (hash_table_get (dl_url_file_map, url));
DEBUGP (("Already downloaded \"%s\", reusing it from \"%s\".\n",
url, file));
- /* this sucks, needs to be combined! */
- if (html_allowed
- && downloaded_html_set
- && string_set_contains (downloaded_html_set, file))
- {
- descend = true;
- is_css = false;
- }
- if (css_allowed
- && downloaded_css_set
- && string_set_contains (downloaded_css_set, file))
+ if ((is_css_bool = (css_allowed
+ && downloaded_css_set
+ && string_set_contains (downloaded_css_set, file)))
+ || (html_allowed
+ && downloaded_html_set
+ && string_set_contains (downloaded_html_set, file)))
{
descend = true;
- is_css = true;
+ is_css = is_css_bool;
}
}
else
{
int dt = 0, url_err;
char *redirected = NULL;
- struct url *url_parsed = url_parse (url, &url_err);
+ struct url *url_parsed = url_parse (url, &url_err, i, true);
- if (!url_parsed)
- {
- char *error = url_error (url, url_err);
- logprintf (LOG_NOTQUIET, "%s: %s.\n", url, error);
- xfree (error);
- status = URLERROR;
- }
- else
- {
- status = retrieve_url (url_parsed, url, &file, &redirected,
- referer, &dt, false);
- }
+ status = retrieve_url (url_parsed, url, &file, &redirected, referer,
+ &dt, false, i, true);
if (html_allowed && file && status == RETROK
&& (dt & RETROKF) && (dt & TEXTHTML))
if (descend)
{
if (!descend_redirect_p (redirected, url_parsed, depth,
- start_url_parsed, blacklist))
+ start_url_parsed, blacklist, i))
descend = false;
else
/* Make sure that the old pre-redirect form gets
xfree (url);
url = redirected;
}
+ else
+ {
+ xfree (url);
+ url = xstrdup (url_parsed->url);
+ }
url_free(url_parsed);
}
bool meta_disallow_follow = false;
struct urlpos *children
= is_css ? get_urls_css_file (file, url) :
- get_urls_html (file, url, &meta_disallow_follow);
+ get_urls_html (file, url, &meta_disallow_follow, i);
if (opt.use_robots && meta_disallow_follow)
{
if (children)
{
struct urlpos *child = children;
- struct url *url_parsed = url_parsed = url_parse (url, NULL);
+ struct url *url_parsed = url_parse (url, NULL, i, true);
+ struct iri *ci;
char *referer_url = url;
bool strip_auth = (url_parsed != NULL
&& url_parsed->user != NULL);
if (dash_p_leaf_HTML && !child->link_inline_p)
continue;
if (download_child_p (child, url_parsed, depth, start_url_parsed,
- blacklist))
+ blacklist, i))
{
- url_enqueue (queue, xstrdup (child->url->url),
+ ci = iri_new ();
+ set_uri_encoding (ci, i->content_encoding, false);
+ url_enqueue (queue, ci, xstrdup (child->url->url),
xstrdup (referer_url), depth + 1,
child->link_expect_html,
child->link_expect_css);
}
}
- if (file
- && (opt.delete_after
+ if (file
+ && (opt.delete_after
|| opt.spider /* opt.recursive is implicitely true */
|| !acceptable (file)))
{
/* Either --delete-after was specified, or we loaded this
- (otherwise unneeded because of --spider or rejected by -R)
- HTML file just to harvest its hyperlinks -- in either case,
+ (otherwise unneeded because of --spider or rejected by -R)
+ HTML file just to harvest its hyperlinks -- in either case,
delete the local file. */
DEBUGP (("Removing file due to %s in recursive_retrieve():\n",
opt.delete_after ? "--delete-after" :
- (opt.spider ? "--spider" :
+ (opt.spider ? "--spider" :
"recursive rejection criteria")));
logprintf (LOG_VERBOSE,
(opt.delete_after || opt.spider
xfree (url);
xfree_null (referer);
xfree_null (file);
+ iri_free (i);
}
/* If anything is left of the queue due to a premature exit, free it
char *d1, *d2;
int d3;
bool d4, d5;
- while (url_dequeue (queue,
+ struct iri *d6;
+ while (url_dequeue (queue, (struct iri **)&d6,
(const char **)&d1, (const char **)&d2, &d3, &d4, &d5))
{
+ iri_free (d6);
xfree (d1);
xfree_null (d2);
}
static bool
download_child_p (const struct urlpos *upos, struct url *parent, int depth,
- struct url *start_url_parsed, struct hash_table *blacklist)
+ struct url *start_url_parsed, struct hash_table *blacklist,
+ struct iri *iri)
{
struct url *u = upos->url;
const char *url = u->url;
if (string_set_contains (blacklist, url))
{
- if (opt.spider)
+ if (opt.spider)
{
char *referrer = url_string (parent, URL_AUTH_HIDE_PASSWD);
DEBUGP (("download_child_p: parent->url is: %s\n", quote (parent->url)));
}
/* Several things to check for:
- 1. if scheme is not http, and we don't load it
- 2. check for relative links (if relative_only is set)
- 3. check for domain
- 4. check for no-parent
- 5. check for excludes && includes
- 6. check for suffix
- 7. check for same host (if spanhost is unset), with possible
+ 1. if scheme is not https and https_only requested
+ 2. if scheme is not http, and we don't load it
+ 3. check for relative links (if relative_only is set)
+ 4. check for domain
+ 5. check for no-parent
+ 6. check for excludes && includes
+ 7. check for suffix
+ 8. check for same host (if spanhost is unset), with possible
gethostbyname baggage
- 8. check for robots.txt
+ 9. check for robots.txt
Addendum: If the URL is FTP, and it is to be loaded, only the
domain and suffix settings are "stronger".
More time- and memory- consuming tests should be put later on
the list. */
+#ifdef HAVE_SSL
+ if (opt.https_only && u->scheme != SCHEME_HTTPS)
+ {
+ DEBUGP (("Not following non-HTTPS links.\n"));
+ goto out;
+ }
+#endif
+
/* Determine whether URL under consideration has a HTTP-like scheme. */
u_scheme_like_http = schemes_are_similar_p (u->scheme, SCHEME_HTTP);
if (opt.no_parent
&& schemes_are_similar_p (u->scheme, start_url_parsed->scheme)
&& 0 == strcasecmp (u->host, start_url_parsed->host)
- && u->port == start_url_parsed->port
+ && (u->scheme != start_url_parsed->scheme
+ || u->port == start_url_parsed->port)
&& !(opt.page_requisites && upos->link_inline_p))
{
if (!subdir_p (start_url_parsed->dir, u->dir))
goto out;
}
}
+ if (!accept_url (url))
+ {
+ DEBUGP (("%s is excluded/not-included through regex.\n", url));
+ goto out;
+ }
/* 6. Check for acceptance/rejection rules. We ignore these rules
for directories (no file name to match) and for non-leaf HTMLs,
if (!specs)
{
char *rfile;
- if (res_retrieve_file (url, &rfile))
+ if (res_retrieve_file (url, &rfile, iri))
{
specs = res_parse_from_file (rfile);
files after downloading or we're just running a spider. */
if (opt.delete_after || opt.spider)
{
- logprintf (LOG_VERBOSE, "Removing %s.\n", rfile);
+ logprintf (LOG_VERBOSE, _("Removing %s.\n"), rfile);
if (unlink (rfile))
logprintf (LOG_NOTQUIET, "unlink: %s\n",
strerror (errno));
static bool
descend_redirect_p (const char *redirected, struct url *orig_parsed, int depth,
- struct url *start_url_parsed, struct hash_table *blacklist)
+ struct url *start_url_parsed, struct hash_table *blacklist,
+ struct iri *iri)
{
struct url *new_parsed;
struct urlpos *upos;
assert (orig_parsed != NULL);
- new_parsed = url_parse (redirected, NULL);
+ new_parsed = url_parse (redirected, NULL, NULL, false);
assert (new_parsed != NULL);
upos = xnew0 (struct urlpos);
upos->url = new_parsed;
success = download_child_p (upos, orig_parsed, depth,
- start_url_parsed, blacklist);
+ start_url_parsed, blacklist, iri);
url_free (new_parsed);
xfree (upos);
projects / wget / blobdiff