recur.c

一个从网络上自动下载文件的自由工具

/* Handling of recursive HTTP retrieving.
   Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003,
   2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.

This file is part of GNU Wget.

GNU Wget is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
 (at your option) any later version.

GNU Wget is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with Wget.  If not, see <http://www.gnu.org/licenses/>.

Additional permission under GNU GPL version 3 section 7

If you modify this program, or any covered work, by linking or
combining it with the OpenSSL project's OpenSSL library (or a
modified version of that library), containing parts covered by the
terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
grants you additional permission to convey the resulting work.
Corresponding Source for a non-source form of such a combination
shall include the source code for the parts of OpenSSL used as well
as that of the covered work.  */

#include <config.h>

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#ifdef HAVE_UNISTD_H
# include <unistd.h>
#endif /* HAVE_UNISTD_H */
#include <errno.h>
#include <assert.h>

#include "wget.h"
#include "url.h"
#include "recur.h"
#include "utils.h"
#include "retr.h"
#include "ftp.h"
#include "host.h"
#include "hash.h"
#include "res.h"
#include "convert.h"
#include "spider.h"

/* Functions for maintaining the URL queue.  */

struct queue_element {
  const char *url;              /* the URL to download */
  const char *referer;          /* the referring document */
  int depth;                    /* the depth */
  bool html_allowed;            /* whether the document is allowed to
                                   be treated as HTML. */

  struct queue_element *next;   /* next element in queue */
};

struct url_queue {
  struct queue_element *head;
  struct queue_element *tail;
  int count, maxcount;
};

/* Create a URL queue. */

static struct url_queue *
url_queue_new (void)
{
  struct url_queue *queue = xnew0 (struct url_queue);
  return queue;
}

/* Delete a URL queue. */

static void
url_queue_delete (struct url_queue *queue)
{
  xfree (queue);
}

/* Enqueue a URL in the queue.  The queue is FIFO: the items will be
   retrieved ("dequeued") from the queue in the order they were placed
   into it.  */

static void
url_enqueue (struct url_queue *queue,
             const char *url, const char *referer, int depth, bool html_allowed)
{
  struct queue_element *qel = xnew (struct queue_element);
  qel->url = url;
  qel->referer = referer;
  qel->depth = depth;
  qel->html_allowed = html_allowed;
  qel->next = NULL;

  ++queue->count;
  if (queue->count > queue->maxcount)
    queue->maxcount = queue->count;

  DEBUGP (("Enqueuing %s at depth %d\n", url, depth));
  DEBUGP (("Queue count %d, maxcount %d.\n", queue->count, queue->maxcount));

  if (queue->tail)
    queue->tail->next = qel;
  queue->tail = qel;

  if (!queue->head)
    queue->head = queue->tail;
}

/* Take a URL out of the queue.  Return true if this operation
   succeeded, or false if the queue is empty.  */

static bool
url_dequeue (struct url_queue *queue,
             const char **url, const char **referer, int *depth,
             bool *html_allowed)
{
  struct queue_element *qel = queue->head;

  if (!qel)
    return false;

  queue->head = queue->head->next;
  if (!queue->head)
    queue->tail = NULL;

  *url = qel->url;
  *referer = qel->referer;
  *depth = qel->depth;
  *html_allowed = qel->html_allowed;

  --queue->count;

  DEBUGP (("Dequeuing %s at depth %d\n", qel->url, qel->depth));
  DEBUGP (("Queue count %d, maxcount %d.\n", queue->count, queue->maxcount));

  xfree (qel);
  return true;
}

static bool download_child_p (const struct urlpos *, struct url *, int,
                              struct url *, struct hash_table *);
static bool descend_redirect_p (const char *, const char *, int,
                                struct url *, struct hash_table *);


/* Retrieve a part of the web beginning with START_URL.  This used to
   be called "recursive retrieval", because the old function was
   recursive and implemented depth-first search.  retrieve_tree on the
   other hand implements breadth-search traversal of the tree, which
   results in much nicer ordering of downloads.

   The algorithm this function uses is simple:

   1. put START_URL in the queue.
   2. while there are URLs in the queue:

     3. get next URL from the queue.
     4. download it.
     5. if the URL is HTML and its depth does not exceed maximum depth,
        get the list of URLs embedded therein.
     6. for each of those URLs do the following:

       7. if the URL is not one of those downloaded before, and if it
          satisfies the criteria specified by the various command-line
          options, add it to the queue. */

uerr_t
retrieve_tree (const char *start_url)
{
  uerr_t status = RETROK;

  /* The queue of URLs we need to load. */
  struct url_queue *queue;

  /* The URLs we do not wish to enqueue, because they are already in
     the queue, but haven't been downloaded yet.  */
  struct hash_table *blacklist;

  int up_error_code;
  struct url *start_url_parsed = url_parse (start_url, &up_error_code);

  if (!start_url_parsed)
    {
      logprintf (LOG_NOTQUIET, "%s: %s.\n", start_url,
                 url_error (up_error_code));
      return URLERROR;
    }

  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);
  string_set_add (blacklist, start_url_parsed->url);

  while (1)
    {
      bool descend = false;
      char *url, *referer, *file = NULL;
      int depth;
      bool html_allowed;
      bool dash_p_leaf_HTML = false;

      if (opt.quota && total_downloaded_bytes > opt.quota)
        break;
      if (status == FWRITEERR)
        break;

      /* Get the next URL from the queue... */

      if (!url_dequeue (queue,
                        (const char **)&url, (const char **)&referer,
                        &depth, &html_allowed))
        break;

      /* ...and download it.  Note that this download is in most cases
         unconditional, as download_child_p already makes sure a file
         doesn't get enqueued twice -- and yet this check is here, and
         not in download_child_p.  This is so that if you run `wget -r
         URL1 URL2', and a random URL is encountered once under URL1
         and again under URL2, but at a different (possibly smaller)
         depth, we want the URL's children to be taken into account
         the second time.  */
      if (dl_url_file_map && hash_table_contains (dl_url_file_map, url))
        {
          file = xstrdup (hash_table_get (dl_url_file_map, url));

          DEBUGP (("Already downloaded \"%s\", reusing it from \"%s\".\n",
                   url, file));

          if (html_allowed
              && downloaded_html_set
              && string_set_contains (downloaded_html_set, file))
            descend = true;
        }
      else
        {
          int dt = 0;
          char *redirected = NULL;

          status = retrieve_url (url, &file, &redirected, referer, &dt, false);

          if (html_allowed && file && status == RETROK
              && (dt & RETROKF) && (dt & TEXTHTML))
            descend = true;

          if (redirected)
            {
              /* We have been redirected, possibly to another host, or
                 different path, or wherever.  Check whether we really
                 want to follow it.  */
              if (descend)
                {
                  if (!descend_redirect_p (redirected, url, depth,
                                           start_url_parsed, blacklist))
                    descend = false;
                  else
                    /* Make sure that the old pre-redirect form gets
                       blacklisted. */
                    string_set_add (blacklist, url);
                }

              xfree (url);
              url = redirected;
            }
        }

      if (opt.spider)
        {
          visited_url (url, referer);
        }

      if (descend
          && depth >= opt.reclevel && opt.reclevel != INFINITE_RECURSION)
        {
          if (opt.page_requisites
              && (depth == opt.reclevel || depth == opt.reclevel + 1))
            {
              /* When -p is specified, we are allowed to exceed the
                 maximum depth, but only for the "inline" links,
                 i.e. those that are needed to display the page.
                 Originally this could exceed the depth at most by
                 one, but we allow one more level so that the leaf
                 pages that contain frames can be loaded
                 correctly.  */
              dash_p_leaf_HTML = true;
            }
          else
            {
              /* Either -p wasn't specified or it was and we've
                 already spent the two extra (pseudo-)levels that it
                 affords us, so we need to bail out. */
              DEBUGP (("Not descending further; at depth %d, max. %d.\n",
                       depth, opt.reclevel));
              descend = false;
            }
        }

      /* If the downloaded document was HTML, parse it and enqueue the
         links it contains. */

      if (descend)
        {
          bool meta_disallow_follow = false;
          struct urlpos *children
            = get_urls_html (file, url, &meta_disallow_follow);

          if (opt.use_robots && meta_disallow_follow)
            {
              free_urlpos (children);
              children = NULL;