convert.c

一个从网络上自动下载文件的自由工具

/* Conversion of links to local files.
   Copyright (C) 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 "convert.h"
#include "url.h"
#include "recur.h"
#include "utils.h"
#include "hash.h"
#include "ptimer.h"
#include "res.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
   conversion after Wget is done.  */
struct hash_table *downloaded_html_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)
{
  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 (cnt == 0)
    goto cleanup;
  file_array = alloca_array (char *, cnt);
  string_set_to_array (downloaded_html_set, file_array);

  for (i = 0; i < cnt; i++)
    {
      struct urlpos *urls, *cur_url;
      char *url;
      char *file = file_array[i];

      /* Determine the URL of the HTML file.  get_urls_html will need
         it.  */
      url = hash_table_get (dl_file_url_map, file);
      if (!url)
        {
          DEBUGP (("Apparently %s has been removed.\n", file));
          continue;
        }

      DEBUGP (("Scanning %s (from %s)\n", file, url));

      /* Parse the HTML file...  */
      urls = get_urls_html (file, url, NULL);

      /* We don't respect meta_disallow_follow here because, even if
         the file is not followed, we might still want to convert the
         links that have been followed from other files.  */

      for (cur_url = urls; cur_url; cur_url = cur_url->next)
        {
          char *local_name;
          struct url *u = cur_url->url;

          if (cur_url->link_base_p)
            {
              /* Base references have been resolved by our parser, so
                 we turn the base URL into an empty string.  (Perhaps
                 we should remove the tag entirely?)  */
              cur_url->convert = CO_NULLIFY_BASE;
              continue;
            }

          /* We decide the direction of conversion according to whether
             a URL was downloaded.  Downloaded URLs will be converted
             ABS2REL, whereas non-downloaded will be converted REL2ABS.  */
          local_name = hash_table_get (dl_url_file_map, u->url);

          /* Decide on the conversion type.  */
          if (local_name)
            {
              /* We've downloaded this URL.  Convert it to relative
                 form.  We do this even if the URL already is in
                 relative form, because our directory structure may
                 not be identical to that on the server (think `-nd',
                 `--cut-dirs', etc.)  */
              cur_url->convert = CO_CONVERT_TO_RELATIVE;
              cur_url->local_name = xstrdup (local_name);
              DEBUGP (("will convert url %s to local %s\n", u->url, local_name));
            }
          else
            {
              /* We haven't downloaded this URL.  If it's not already
                 complete (including a full host name), convert it to
                 that form, so it can be reached while browsing this
                 HTML locally.  */
              if (!cur_url->link_complete_p)
                cur_url->convert = CO_CONVERT_TO_COMPLETE;
              cur_url->local_name = NULL;
              DEBUGP (("will convert url %s to complete\n", u->url));
            }
        }

      /* Convert the links in the file.  */
      convert_links (file, urls);
      ++file_count;

      /* Free the data.  */
      free_urlpos (urls);
    }

  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_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
   document, along with their positions and the desired direction of
   the conversion.  */
static void
convert_links (const char *file, struct urlpos *links)
{
  struct file_memory *fm;
  FILE *fp;
  const char *p;
  downloaded_file_t downloaded_file_return;

  struct urlpos *link;
  int to_url_count = 0, to_file_count = 0;

  logprintf (LOG_VERBOSE, _("Converting %s... "), file);

  {
    /* First we do a "dry run": go through the list L and see whether
       any URL needs to be converted in the first place.  If not, just
       leave the file alone.  */
    int dry_count = 0;
    struct urlpos *dry;
    for (dry = links; dry; dry = dry->next)
      if (dry->convert != CO_NOCONVERT)
        ++dry_count;
    if (!dry_count)
      {
        logputs (LOG_VERBOSE, _("nothing to do.\n"));
        return;
      }
  }

  fm = read_file (file);
  if (!fm)
    {
      logprintf (LOG_NOTQUIET, _("Cannot convert links in %s: %s\n"),
                 file, strerror (errno));
      return;
    }

  downloaded_file_return = downloaded_file (CHECK_FOR_FILE, file);
  if (opt.backup_converted && downloaded_file_return)
    write_backup_file (file, downloaded_file_return);

  /* Before opening the file for writing, unlink the file.  This is
     important if the data in FM is mmaped.  In such case, nulling the
     file, which is what fopen() below does, would make us read all
     zeroes from the mmaped region.  */
  if (unlink (file) < 0 && errno != ENOENT)
    {
      logprintf (LOG_NOTQUIET, _("Unable to delete `%s': %s\n"),
                 file, strerror (errno));
      read_file_free (fm);
      return;
    }
  /* Now open the file for writing.  */
  fp = fopen (file, "wb");
  if (!fp)
    {
      logprintf (LOG_NOTQUIET, _("Cannot convert links in %s: %s\n"),
                 file, strerror (errno));
      read_file_free (fm);
      return;
    }

  /* Here we loop through all the URLs in file, replacing those of
     them that are downloaded with relative references.  */
  p = fm->content;
  for (link = links; link; link = link->next)
    {
      char *url_start = fm->content + link->pos;

      if (link->pos >= fm->length)
        {
          DEBUGP (("Something strange is going on.  Please investigate."));
          break;
        }
      /* If the URL is not to be converted, skip it.  */
      if (link->convert == CO_NOCONVERT)
        {
          DEBUGP (("Skipping %s at position %d.\n", link->url->url, link->pos));
          continue;
        }

      /* Echo the file contents, up to the offending URL's opening
         quote, to the outfile.  */
      fwrite (p, 1, url_start - p, fp);
      p = url_start;

      switch (link->convert)
        {
        case CO_CONVERT_TO_RELATIVE:
          /* Convert absolute URL to relative. */
          {
            char *newname = construct_relative (file, link->local_name);
            char *quoted_newname = local_quote_string (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);
            xfree (quoted_newname);
            ++to_file_count;
            break;
          }
        case CO_CONVERT_TO_COMPLETE:
          /* Convert the link to absolute URL. */
          {
            char *newlink = link->url->url;
            char *quoted_newlink = html_quote_string (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);
            ++to_url_count;
            break;
          }
        case CO_NULLIFY_BASE:
          /* Change the base href to "". */
          p = replace_attr (p, link->size, fp, "");
          break;
        case CO_NOCONVERT:
          abort ();
          break;
        }
    }

  /* Output the rest of the file. */
  if (p - fm->content < fm->length)
    fwrite (p, 1, fm->length - (p - fm->content), fp);
  fclose (fp);
  read_file_free (fm);

  logprintf (LOG_VERBOSE, "%d-%d\n", to_file_count, to_url_count);
}

/* Construct and return a link that points from BASEFILE to LINKFILE.
   Both files should be local file names, BASEFILE of the referrering
   file, and LINKFILE of the referred file.

   Examples:

   cr("foo", "bar")         -> "bar"
   cr("A/foo", "A/bar")     -> "bar"
   cr("A/foo", "A/B/bar")   -> "B/bar"
   cr("A/X/foo", "A/Y/bar") -> "../Y/bar"
   cr("X/", "Y/bar")        -> "../Y/bar" (trailing slash does matter in BASE)

   Both files should be absolute or relative, otherwise strange
   results might ensue.  The function makes no special efforts to