1
0
mirror of https://github.com/moparisthebest/wget synced 2024-07-03 16:38:41 -04:00
wget/src/init.c

1493 lines
39 KiB
C
Raw Normal View History

1999-12-02 02:42:23 -05:00
/* Reading/parsing the initialization file.
Copyright (C) 1996-2005 Free Software Foundation, Inc.
1999-12-02 02:42:23 -05:00
2001-05-27 15:35:15 -04:00
This file is part of GNU Wget.
1999-12-02 02:42:23 -05:00
2001-05-27 15:35:15 -04:00
GNU Wget is free software; you can redistribute it and/or modify
1999-12-02 02:42:23 -05:00
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
2001-05-27 15:35:15 -04:00
GNU Wget is distributed in the hope that it will be useful,
1999-12-02 02:42:23 -05:00
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, write to the Free Software Foundation, Inc.,
51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
In addition, as a special exception, the Free Software Foundation
gives permission to link the code of its release of Wget with the
OpenSSL project's "OpenSSL" library (or with modified versions of it
that use the same license as the "OpenSSL" library), and distribute
the linked executables. You must obey the GNU General Public License
in all respects for all of the code used other than "OpenSSL". If you
modify this file, you may extend this exception to your version of the
file, but you are not obligated to do so. If you do not wish to do
so, delete this exception statement from your version. */
1999-12-02 02:42:23 -05:00
#include <config.h>
#include <stdio.h>
#include <stdlib.h>
#ifdef HAVE_UNISTD_H
# include <unistd.h>
#endif
2005-06-19 18:34:58 -04:00
#include <string.h>
1999-12-02 02:42:23 -05:00
#include <errno.h>
#ifdef HAVE_PWD_H
# include <pwd.h>
1999-12-02 02:42:23 -05:00
#endif
#include <assert.h>
1999-12-02 02:42:23 -05:00
#include "wget.h"
#include "utils.h"
#include "init.h"
#include "host.h"
#include "netrc.h"
#include "progress.h"
#include "recur.h" /* for INFINITE_RECURSION */
#include "convert.h" /* for convert_cleanup */
#include "res.h" /* for res_cleanup */
#include "http.h" /* for http_cleanup */
#include "retr.h" /* for output_stream */
1999-12-02 02:42:23 -05:00
/* We want tilde expansion enabled only when reading `.wgetrc' lines;
otherwise, it will be performed by the shell. This variable will
be set by the wgetrc-reading function. */
static bool enable_tilde_expansion;
1999-12-02 02:42:23 -05:00
#define CMD_DECLARE(func) static bool func (const char *, const char *, void *)
1999-12-02 02:42:23 -05:00
CMD_DECLARE (cmd_boolean);
CMD_DECLARE (cmd_bytes);
CMD_DECLARE (cmd_bytes_sum);
#ifdef HAVE_SSL
CMD_DECLARE (cmd_cert_type);
#endif
CMD_DECLARE (cmd_directory_vector);
1999-12-02 02:42:23 -05:00
CMD_DECLARE (cmd_number);
CMD_DECLARE (cmd_number_inf);
CMD_DECLARE (cmd_string);
CMD_DECLARE (cmd_file);
CMD_DECLARE (cmd_directory);
1999-12-02 02:42:23 -05:00
CMD_DECLARE (cmd_time);
CMD_DECLARE (cmd_vector);
1999-12-02 02:42:23 -05:00
CMD_DECLARE (cmd_spec_dirstruct);
CMD_DECLARE (cmd_spec_header);
CMD_DECLARE (cmd_spec_htmlify);
CMD_DECLARE (cmd_spec_mirror);
CMD_DECLARE (cmd_spec_prefer_family);
CMD_DECLARE (cmd_spec_progress);
1999-12-02 02:42:23 -05:00
CMD_DECLARE (cmd_spec_recursive);
CMD_DECLARE (cmd_spec_restrict_file_names);
#ifdef HAVE_SSL
CMD_DECLARE (cmd_spec_secure_protocol);
#endif
CMD_DECLARE (cmd_spec_timeout);
1999-12-02 02:42:23 -05:00
CMD_DECLARE (cmd_spec_useragent);
CMD_DECLARE (cmd_spec_verbose);
1999-12-02 02:42:23 -05:00
2005-04-26 13:30:45 -04:00
/* List of recognized commands, each consisting of name, place and
2003-11-04 18:19:39 -05:00
function. When adding a new command, simply add it to the list,
but be sure to keep the list sorted alphabetically, as
command_by_name depends on it. Also, be sure to add any entries
2005-05-30 08:52:44 -04:00
that allocate memory (e.g. cmd_string and cmd_vector) to the
[svn] * TODO: Removed done item: we now have an option (-G) that makes it easy to download a single HTML document and all its constituents. * po/*.{gmo,po,pot}: Regenerated after adding new options. * po/hr.po: Hrvoje forgot '\n's on his translations of my altered messages, causing msgfmt to balk and `make install' to fail. * wget.texi (Recursive Retrieval Options): In -K description, added a link to the discussion of interaction with -N. (Recursive Accept/Reject Options): Did some alphabetizing and added descriptions of new --follow-tags and -G / --ignore-tags options. (Following Links): Changed "the loads of" to "loads of". (Wgetrc Commands): Added descriptions of new follow_tags and ignore_tags commands. * html.c (idmatch): Implemented checking of my new --follow-tags and --ignore-tags options. * init.c (commands): Added comment reminding people adding new entries doing allocation to add corresponding freeing in cleanup(). (commands): Added new followtags and ignoretags commands. (cleanup): Free storage for new followtags and ignoretags. * main.c: Use of "comma-separated list" was random -- normalized it. Did some alphabetization. Added comments pointing out "Options without arguments" and "Options accepting an argument" sections of long_options[]. Added new options --follow-tags and -G / --ignore-tags. Added comment that Damir's --referer is currently undocumented. Added comment that Heiko's --waitretry is partially undocumented (mentioned in --help but not in wget.texi). Moved improperly sorted 24, 129, and 'G' cases. * options.h (struct options): Added new fields follow_tags and ignore_tags. * wget.h: Added "#define EQ 0" so we can say "strcmp(a, b) == EQ".
2000-03-11 01:48:06 -05:00
cleanup() function below. */
2003-11-04 18:19:39 -05:00
1999-12-02 02:42:23 -05:00
static struct {
const char *name;
2005-04-26 13:30:45 -04:00
void *place;
bool (*action) (const char *, const char *, void *);
1999-12-02 02:42:23 -05:00
} commands[] = {
{ "accept", &opt.accepts, cmd_vector },
{ "addhostdir", &opt.add_hostdir, cmd_boolean },
{ "alwaysrest", &opt.always_rest, cmd_boolean }, /* deprecated */
{ "background", &opt.background, cmd_boolean },
{ "backupconverted", &opt.backup_converted, cmd_boolean },
1999-12-02 02:42:23 -05:00
{ "backups", &opt.backups, cmd_number },
{ "base", &opt.base_href, cmd_string },
{ "bindaddress", &opt.bind_address, cmd_string },
#ifdef HAVE_SSL
{ "cacertificate", &opt.ca_cert, cmd_file },
#endif
{ "cache", &opt.allow_cache, cmd_boolean },
#ifdef HAVE_SSL
{ "cadirectory", &opt.ca_directory, cmd_directory },
{ "certificate", &opt.cert_file, cmd_file },
{ "certificatetype", &opt.cert_type, cmd_cert_type },
{ "checkcertificate", &opt.check_cert, cmd_boolean },
#endif
{ "connecttimeout", &opt.connect_timeout, cmd_time },
1999-12-02 02:42:23 -05:00
{ "continue", &opt.always_rest, cmd_boolean },
{ "convertlinks", &opt.convert_links, cmd_boolean },
2001-04-08 18:25:24 -04:00
{ "cookies", &opt.cookies, cmd_boolean },
1999-12-02 02:42:23 -05:00
{ "cutdirs", &opt.cut_dirs, cmd_number },
2003-10-07 20:05:51 -04:00
#ifdef ENABLE_DEBUG
1999-12-02 02:42:23 -05:00
{ "debug", &opt.debug, cmd_boolean },
#endif
{ "deleteafter", &opt.delete_after, cmd_boolean },
{ "dirprefix", &opt.dir_prefix, cmd_directory },
1999-12-02 02:42:23 -05:00
{ "dirstruct", NULL, cmd_spec_dirstruct },
2003-09-10 15:41:54 -04:00
{ "dnscache", &opt.dns_cache, cmd_boolean },
{ "dnstimeout", &opt.dns_timeout, cmd_time },
1999-12-02 02:42:23 -05:00
{ "domains", &opt.domains, cmd_vector },
{ "dotbytes", &opt.dot_bytes, cmd_bytes },
{ "dotsinline", &opt.dots_in_line, cmd_number },
{ "dotspacing", &opt.dot_spacing, cmd_number },
2001-11-27 21:44:18 -05:00
{ "dotstyle", &opt.dot_style, cmd_string },
#ifdef HAVE_SSL
2005-04-26 13:30:45 -04:00
{ "egdfile", &opt.egd_file, cmd_file },
#endif
1999-12-02 02:42:23 -05:00
{ "excludedirectories", &opt.excludes, cmd_directory_vector },
{ "excludedomains", &opt.exclude_domains, cmd_vector },
{ "followftp", &opt.follow_ftp, cmd_boolean },
[svn] * TODO: Removed done item: we now have an option (-G) that makes it easy to download a single HTML document and all its constituents. * po/*.{gmo,po,pot}: Regenerated after adding new options. * po/hr.po: Hrvoje forgot '\n's on his translations of my altered messages, causing msgfmt to balk and `make install' to fail. * wget.texi (Recursive Retrieval Options): In -K description, added a link to the discussion of interaction with -N. (Recursive Accept/Reject Options): Did some alphabetizing and added descriptions of new --follow-tags and -G / --ignore-tags options. (Following Links): Changed "the loads of" to "loads of". (Wgetrc Commands): Added descriptions of new follow_tags and ignore_tags commands. * html.c (idmatch): Implemented checking of my new --follow-tags and --ignore-tags options. * init.c (commands): Added comment reminding people adding new entries doing allocation to add corresponding freeing in cleanup(). (commands): Added new followtags and ignoretags commands. (cleanup): Free storage for new followtags and ignoretags. * main.c: Use of "comma-separated list" was random -- normalized it. Did some alphabetization. Added comments pointing out "Options without arguments" and "Options accepting an argument" sections of long_options[]. Added new options --follow-tags and -G / --ignore-tags. Added comment that Damir's --referer is currently undocumented. Added comment that Heiko's --waitretry is partially undocumented (mentioned in --help but not in wget.texi). Moved improperly sorted 24, 129, and 'G' cases. * options.h (struct options): Added new fields follow_tags and ignore_tags. * wget.h: Added "#define EQ 0" so we can say "strcmp(a, b) == EQ".
2000-03-11 01:48:06 -05:00
{ "followtags", &opt.follow_tags, cmd_vector },
1999-12-02 02:42:23 -05:00
{ "forcehtml", &opt.force_html, cmd_boolean },
{ "ftppasswd", &opt.ftp_passwd, cmd_string }, /* deprecated */
{ "ftppassword", &opt.ftp_passwd, cmd_string },
1999-12-02 02:42:23 -05:00
{ "ftpproxy", &opt.ftp_proxy, cmd_string },
{ "ftpuser", &opt.ftp_user, cmd_string },
1999-12-02 02:42:23 -05:00
{ "glob", &opt.ftp_glob, cmd_boolean },
2005-05-30 09:31:24 -04:00
{ "header", NULL, cmd_spec_header },
{ "htmlextension", &opt.html_extension, cmd_boolean },
1999-12-02 02:42:23 -05:00
{ "htmlify", NULL, cmd_spec_htmlify },
{ "httpkeepalive", &opt.http_keep_alive, cmd_boolean },
{ "httppasswd", &opt.http_passwd, cmd_string }, /* deprecated */
{ "httppassword", &opt.http_passwd, cmd_string },
1999-12-02 02:42:23 -05:00
{ "httpproxy", &opt.http_proxy, cmd_string },
{ "httpsproxy", &opt.https_proxy, cmd_string },
{ "httpuser", &opt.http_user, cmd_string },
{ "ignorecase", &opt.ignore_case, cmd_boolean },
1999-12-02 02:42:23 -05:00
{ "ignorelength", &opt.ignore_length, cmd_boolean },
[svn] * TODO: Removed done item: we now have an option (-G) that makes it easy to download a single HTML document and all its constituents. * po/*.{gmo,po,pot}: Regenerated after adding new options. * po/hr.po: Hrvoje forgot '\n's on his translations of my altered messages, causing msgfmt to balk and `make install' to fail. * wget.texi (Recursive Retrieval Options): In -K description, added a link to the discussion of interaction with -N. (Recursive Accept/Reject Options): Did some alphabetizing and added descriptions of new --follow-tags and -G / --ignore-tags options. (Following Links): Changed "the loads of" to "loads of". (Wgetrc Commands): Added descriptions of new follow_tags and ignore_tags commands. * html.c (idmatch): Implemented checking of my new --follow-tags and --ignore-tags options. * init.c (commands): Added comment reminding people adding new entries doing allocation to add corresponding freeing in cleanup(). (commands): Added new followtags and ignoretags commands. (cleanup): Free storage for new followtags and ignoretags. * main.c: Use of "comma-separated list" was random -- normalized it. Did some alphabetization. Added comments pointing out "Options without arguments" and "Options accepting an argument" sections of long_options[]. Added new options --follow-tags and -G / --ignore-tags. Added comment that Damir's --referer is currently undocumented. Added comment that Heiko's --waitretry is partially undocumented (mentioned in --help but not in wget.texi). Moved improperly sorted 24, 129, and 'G' cases. * options.h (struct options): Added new fields follow_tags and ignore_tags. * wget.h: Added "#define EQ 0" so we can say "strcmp(a, b) == EQ".
2000-03-11 01:48:06 -05:00
{ "ignoretags", &opt.ignore_tags, cmd_vector },
1999-12-02 02:42:23 -05:00
{ "includedirectories", &opt.includes, cmd_directory_vector },
#ifdef ENABLE_IPV6
{ "inet4only", &opt.ipv4_only, cmd_boolean },
{ "inet6only", &opt.ipv6_only, cmd_boolean },
#endif
2003-11-14 20:49:52 -05:00
{ "input", &opt.input_filename, cmd_file },
{ "keepsessioncookies", &opt.keep_session_cookies, cmd_boolean },
1999-12-02 02:42:23 -05:00
{ "killlonger", &opt.kill_longer, cmd_boolean },
{ "limitrate", &opt.limit_rate, cmd_bytes },
{ "loadcookies", &opt.cookies_input, cmd_file },
{ "logfile", &opt.lfilename, cmd_file },
{ "login", &opt.ftp_user, cmd_string },/* deprecated*/
1999-12-02 02:42:23 -05:00
{ "mirror", NULL, cmd_spec_mirror },
{ "netrc", &opt.netrc, cmd_boolean },
{ "noclobber", &opt.noclobber, cmd_boolean },
{ "noparent", &opt.no_parent, cmd_boolean },
{ "noproxy", &opt.no_proxy, cmd_vector },
[svn] * TODO: Removed done item: we now have an option (-G) that makes it easy to download a single HTML document and all its constituents. * po/*.{gmo,po,pot}: Regenerated after adding new options. * po/hr.po: Hrvoje forgot '\n's on his translations of my altered messages, causing msgfmt to balk and `make install' to fail. * wget.texi (Recursive Retrieval Options): In -K description, added a link to the discussion of interaction with -N. (Recursive Accept/Reject Options): Did some alphabetizing and added descriptions of new --follow-tags and -G / --ignore-tags options. (Following Links): Changed "the loads of" to "loads of". (Wgetrc Commands): Added descriptions of new follow_tags and ignore_tags commands. * html.c (idmatch): Implemented checking of my new --follow-tags and --ignore-tags options. * init.c (commands): Added comment reminding people adding new entries doing allocation to add corresponding freeing in cleanup(). (commands): Added new followtags and ignoretags commands. (cleanup): Free storage for new followtags and ignoretags. * main.c: Use of "comma-separated list" was random -- normalized it. Did some alphabetization. Added comments pointing out "Options without arguments" and "Options accepting an argument" sections of long_options[]. Added new options --follow-tags and -G / --ignore-tags. Added comment that Damir's --referer is currently undocumented. Added comment that Heiko's --waitretry is partially undocumented (mentioned in --help but not in wget.texi). Moved improperly sorted 24, 129, and 'G' cases. * options.h (struct options): Added new fields follow_tags and ignore_tags. * wget.h: Added "#define EQ 0" so we can say "strcmp(a, b) == EQ".
2000-03-11 01:48:06 -05:00
{ "numtries", &opt.ntry, cmd_number_inf },/* deprecated*/
{ "outputdocument", &opt.output_document, cmd_file },
{ "pagerequisites", &opt.page_requisites, cmd_boolean },
{ "passiveftp", &opt.ftp_pasv, cmd_boolean },
{ "passwd", &opt.ftp_passwd, cmd_string },/* deprecated*/
{ "password", &opt.passwd, cmd_string },
{ "postdata", &opt.post_data, cmd_string },
{ "postfile", &opt.post_file_name, cmd_file },
{ "preferfamily", NULL, cmd_spec_prefer_family },
{ "preservepermissions", &opt.preserve_perm, cmd_boolean },
#ifdef HAVE_SSL
{ "privatekey", &opt.private_key, cmd_file },
{ "privatekeytype", &opt.private_key_type, cmd_cert_type },
#endif
{ "progress", &opt.progress_type, cmd_spec_progress },
{ "protocoldirectories", &opt.protocol_directories, cmd_boolean },
{ "proxypasswd", &opt.proxy_passwd, cmd_string }, /* deprecated */
{ "proxypassword", &opt.proxy_passwd, cmd_string },
1999-12-02 02:42:23 -05:00
{ "proxyuser", &opt.proxy_user, cmd_string },
{ "quiet", &opt.quiet, cmd_boolean },
{ "quota", &opt.quota, cmd_bytes_sum },
#ifdef HAVE_SSL
{ "randomfile", &opt.random_file, cmd_file },
#endif
{ "randomwait", &opt.random_wait, cmd_boolean },
{ "readtimeout", &opt.read_timeout, cmd_time },
1999-12-02 02:42:23 -05:00
{ "reclevel", &opt.reclevel, cmd_number_inf },
{ "recursive", NULL, cmd_spec_recursive },
2000-03-02 09:23:22 -05:00
{ "referer", &opt.referer, cmd_string },
1999-12-02 02:42:23 -05:00
{ "reject", &opt.rejects, cmd_vector },
{ "relativeonly", &opt.relative_only, cmd_boolean },
{ "removelisting", &opt.remove_listing, cmd_boolean },
{ "restrictfilenames", NULL, cmd_spec_restrict_file_names },
1999-12-02 02:42:23 -05:00
{ "retrsymlinks", &opt.retr_symlinks, cmd_boolean },
{ "retryconnrefused", &opt.retry_connrefused, cmd_boolean },
1999-12-02 02:42:23 -05:00
{ "robots", &opt.use_robots, cmd_boolean },
{ "savecookies", &opt.cookies_output, cmd_file },
1999-12-02 02:42:23 -05:00
{ "saveheaders", &opt.save_headers, cmd_boolean },
#ifdef HAVE_SSL
{ "secureprotocol", &opt.secure_protocol, cmd_spec_secure_protocol },
#endif
1999-12-02 02:42:23 -05:00
{ "serverresponse", &opt.server_response, cmd_boolean },
{ "spanhosts", &opt.spanhost, cmd_boolean },
{ "spider", &opt.spider, cmd_boolean },
{ "strictcomments", &opt.strict_comments, cmd_boolean },
{ "timeout", NULL, cmd_spec_timeout },
1999-12-02 02:42:23 -05:00
{ "timestamping", &opt.timestamping, cmd_boolean },
{ "tries", &opt.ntry, cmd_number_inf },
{ "useproxy", &opt.use_proxy, cmd_boolean },
{ "user", &opt.user, cmd_string },
1999-12-02 02:42:23 -05:00
{ "useragent", NULL, cmd_spec_useragent },
{ "verbose", NULL, cmd_spec_verbose },
{ "wait", &opt.wait, cmd_time },
{ "waitretry", &opt.waitretry, cmd_time }
1999-12-02 02:42:23 -05:00
};
2003-11-04 18:19:39 -05:00
/* Look up CMDNAME in the commands[] and return its position in the
array. If CMDNAME is not found, return -1. */
1999-12-02 02:42:23 -05:00
static int
2003-11-04 18:19:39 -05:00
command_by_name (const char *cmdname)
1999-12-02 02:42:23 -05:00
{
2003-11-04 18:19:39 -05:00
/* Use binary search for speed. Wget has ~100 commands, which
guarantees a worst case performance of 7 string comparisons. */
int lo = 0, hi = countof (commands) - 1;
1999-12-02 02:42:23 -05:00
while (lo <= hi)
1999-12-02 02:42:23 -05:00
{
int mid = (lo + hi) >> 1;
2003-11-04 18:19:39 -05:00
int cmp = strcasecmp (cmdname, commands[mid].name);
if (cmp < 0)
hi = mid - 1;
else if (cmp > 0)
lo = mid + 1;
1999-12-02 02:42:23 -05:00
else
return mid;
1999-12-02 02:42:23 -05:00
}
return -1;
}
/* Reset the variables to default values. */
static void
defaults (void)
{
char *tmp;
/* Most of the default values are 0 (and 0.0, NULL, and false).
Just reset everything, and fill in the non-zero values. Note
that initializing pointers to NULL this way is technically
illegal, but porting Wget to a machine where NULL is not all-zero
bit pattern will be the least of the implementors' worries. */
xzero (opt);
1999-12-02 02:42:23 -05:00
opt.cookies = true;
1999-12-02 02:42:23 -05:00
opt.verbose = -1;
opt.ntry = 20;
opt.reclevel = 5;
opt.add_hostdir = true;
opt.netrc = true;
opt.ftp_glob = true;
opt.htmlify = true;
opt.http_keep_alive = true;
opt.use_proxy = true;
1999-12-02 02:42:23 -05:00
tmp = getenv ("no_proxy");
if (tmp)
opt.no_proxy = sepstring (tmp);
opt.allow_cache = true;
1999-12-02 02:42:23 -05:00
opt.read_timeout = 900;
opt.use_robots = true;
1999-12-02 02:42:23 -05:00
opt.remove_listing = true;
1999-12-02 02:42:23 -05:00
opt.dot_bytes = 1024;
opt.dot_spacing = 10;
opt.dots_in_line = 50;
2003-09-10 15:41:54 -04:00
opt.dns_cache = true;
opt.ftp_pasv = true;
#ifdef HAVE_SSL
opt.check_cert = true;
#endif
/* The default for file name restriction defaults to the OS type. */
#if !defined(WINDOWS) && !defined(__CYGWIN__)
opt.restrict_files_os = restrict_unix;
#else
opt.restrict_files_os = restrict_windows;
#endif
opt.restrict_files_ctrl = true;
1999-12-02 02:42:23 -05:00
}
/* Return the user's home directory (strdup-ed), or NULL if none is
found. */
char *
home_dir (void)
{
char *home = getenv ("HOME");
if (!home)
{
#ifndef WINDOWS
/* If HOME is not defined, try getting it from the password
file. */
struct passwd *pwd = getpwuid (getuid ());
if (!pwd || !pwd->pw_dir)
return NULL;
home = pwd->pw_dir;
#else /* WINDOWS */
/* Under Windows, if $HOME isn't defined, use the directory where
`wget.exe' resides. */
home = ws_mypath ();
1999-12-02 02:42:23 -05:00
#endif /* WINDOWS */
}
return home ? xstrdup (home) : NULL;
}
/* Return the path to the user's .wgetrc. This is either the value of
`WGETRC' environment variable, or `$HOME/.wgetrc'.
If the `WGETRC' variable exists but the file does not exist, the
function will exit(). */
static char *
wgetrc_file_name (void)
{
char *env, *home;
char *file = NULL;
/* Try the environment. */
env = getenv ("WGETRC");
if (env && *env)
{
if (!file_exists_p (env))
{
2003-09-21 15:17:45 -04:00
fprintf (stderr, _("%s: WGETRC points to %s, which doesn't exist.\n"),
exec_name, env);
1999-12-02 02:42:23 -05:00
exit (1);
}
return xstrdup (env);
}
/* If that failed, try $HOME/.wgetrc. */
home = home_dir ();
if (home)
file = aprintf ("%s/.wgetrc", home);
xfree_null (home);
1999-12-02 02:42:23 -05:00
#ifdef WINDOWS
/* Under Windows, if we still haven't found .wgetrc, look for the file
`wget.ini' in the directory where `wget.exe' resides; we do this for
backward compatibility with previous versions of Wget.
SYSTEM_WGETRC should not be defined under WINDOWS. */
if (!file || !file_exists_p (file))
1999-12-02 02:42:23 -05:00
{
xfree_null (file);
file = NULL;
home = ws_mypath ();
if (home)
file = aprintf ("%s/wget.ini", home);
1999-12-02 02:42:23 -05:00
}
#endif /* WINDOWS */
if (!file)
return NULL;
if (!file_exists_p (file))
{
xfree (file);
1999-12-02 02:42:23 -05:00
return NULL;
}
return file;
}
/* Return values of parse_line. */
enum parse_line {
line_ok,
line_empty,
line_syntax_error,
line_unknown_command
};
2005-06-19 18:34:58 -04:00
static enum parse_line parse_line (const char *, char **, char **, int *);
static bool setval_internal (int, const char *, const char *);
/* Initialize variables from a wgetrc file. Returns zero (failure) if
there were errors in the file. */
static bool
1999-12-02 02:42:23 -05:00
run_wgetrc (const char *file)
{
FILE *fp;
char *line;
int ln;
int errcnt = 0;
1999-12-02 02:42:23 -05:00
fp = fopen (file, "rb");
if (!fp)
{
fprintf (stderr, _("%s: Cannot read %s (%s).\n"), exec_name,
file, strerror (errno));
return true; /* not a fatal error */
1999-12-02 02:42:23 -05:00
}
enable_tilde_expansion = true;
1999-12-02 02:42:23 -05:00
ln = 1;
2005-03-05 21:11:10 -05:00
while ((line = read_whole_line (fp)) != NULL)
1999-12-02 02:42:23 -05:00
{
char *com = NULL, *val = NULL;
int comind;
1999-12-02 02:42:23 -05:00
/* Parse the line. */
switch (parse_line (line, &com, &val, &comind))
1999-12-02 02:42:23 -05:00
{
case line_ok:
/* If everything is OK, set the value. */
if (!setval_internal (comind, com, val))
{
fprintf (stderr, _("%s: Error in %s at line %d.\n"),
exec_name, file, ln);
++errcnt;
}
break;
case line_syntax_error:
fprintf (stderr, _("%s: Syntax error in %s at line %d.\n"),
exec_name, file, ln);
++errcnt;
break;
case line_unknown_command:
fprintf (stderr, _("%s: Unknown command `%s' in %s at line %d.\n"),
exec_name, com, file, ln);
++errcnt;
break;
case line_empty:
break;
default:
abort ();
1999-12-02 02:42:23 -05:00
}
xfree_null (com);
xfree_null (val);
xfree (line);
1999-12-02 02:42:23 -05:00
++ln;
}
enable_tilde_expansion = false;
1999-12-02 02:42:23 -05:00
fclose (fp);
return errcnt == 0;
1999-12-02 02:42:23 -05:00
}
/* Initialize the defaults and run the system wgetrc and user's own
wgetrc. */
void
initialize (void)
{
char *file;
int ok = true;
1999-12-02 02:42:23 -05:00
/* Load the hard-coded defaults. */
defaults ();
/* If SYSTEM_WGETRC is defined, use it. */
#ifdef SYSTEM_WGETRC
if (file_exists_p (SYSTEM_WGETRC))
ok &= run_wgetrc (SYSTEM_WGETRC);
1999-12-02 02:42:23 -05:00
#endif
/* Override it with your own, if one exists. */
file = wgetrc_file_name ();
if (!file)
return;
/* #### We should canonicalize `file' and SYSTEM_WGETRC with
something like realpath() before comparing them with `strcmp' */
1999-12-02 02:42:23 -05:00
#ifdef SYSTEM_WGETRC
if (!strcmp (file, SYSTEM_WGETRC))
{
fprintf (stderr, _("\
%s: Warning: Both system and user wgetrc point to `%s'.\n"),
exec_name, file);
}
else
#endif
ok &= run_wgetrc (file);
/* If there were errors processing either `.wgetrc', abort. */
if (!ok)
exit (2);
xfree (file);
1999-12-02 02:42:23 -05:00
return;
}
/* Remove dashes and underscores from S, modifying S in the
process. */
static void
dehyphen (char *s)
{
char *t = s; /* t - tortoise */
char *h = s; /* h - hare */
while (*h)
if (*h == '_' || *h == '-')
++h;
else
*t++ = *h++;
*t = '\0';
}
1999-12-02 02:42:23 -05:00
/* Parse the line pointed by line, with the syntax:
2003-09-21 15:17:45 -04:00
<sp>* command <sp>* = <sp>* value <sp>*
1999-12-02 02:42:23 -05:00
Uses malloc to allocate space for command and value.
Returns one of line_ok, line_empty, line_syntax_error, or
line_unknown_command.
2003-09-21 15:17:45 -04:00
In case of line_ok, *COM and *VAL point to freshly allocated
2003-09-21 15:17:45 -04:00
strings, and *COMIND points to com's index. In case of error or
empty line, their values are unmodified. */
static enum parse_line
parse_line (const char *line, char **com, char **val, int *comind)
1999-12-02 02:42:23 -05:00
{
const char *p;
const char *end = line + strlen (line);
const char *cmdstart, *cmdend;
const char *valstart, *valend;
1999-12-02 02:42:23 -05:00
char *cmdcopy;
int ind;
/* Skip leading and trailing whitespace. */
while (*line && ISSPACE (*line))
++line;
while (end > line && ISSPACE (end[-1]))
--end;
1999-12-02 02:42:23 -05:00
/* Skip empty lines and comments. */
if (!*line || *line == '#')
return line_empty;
1999-12-02 02:42:23 -05:00
p = line;
cmdstart = p;
while (p < end && (ISALNUM (*p) || *p == '_' || *p == '-'))
++p;
cmdend = p;
/* Skip '=', as well as any space before or after it. */
while (p < end && ISSPACE (*p))
++p;
if (p == end || *p != '=')
return line_syntax_error;
++p;
while (p < end && ISSPACE (*p))
++p;
1999-12-02 02:42:23 -05:00
valstart = p;
valend = end;
/* The syntax is valid (even though the command might not be). Fill
in the command name and value. */
*com = strdupdelim (cmdstart, cmdend);
*val = strdupdelim (valstart, valend);
/* The line now known to be syntactically correct. Check whether
the command is valid. */
BOUNDED_TO_ALLOCA (cmdstart, cmdend, cmdcopy);
dehyphen (cmdcopy);
2003-11-04 18:19:39 -05:00
ind = command_by_name (cmdcopy);
if (ind == -1)
return line_unknown_command;
/* Report success to the caller. */
*comind = ind;
return line_ok;
1999-12-02 02:42:23 -05:00
}
/* Run commands[comind].action. */
static bool
setval_internal (int comind, const char *com, const char *val)
1999-12-02 02:42:23 -05:00
{
assert (0 <= comind && comind < countof (commands));
DEBUGP (("Setting %s (%s) to %s\n", com, commands[comind].name, val));
2005-06-19 18:34:58 -04:00
return commands[comind].action (com, val, commands[comind].place);
}
1999-12-02 02:42:23 -05:00
/* Run command COM with value VAL. If running the command produces an
error, report the error and exit.
2003-10-30 06:09:18 -05:00
This is intended to be called from main() to modify Wget's behavior
through command-line switches. Since COM is hard-coded in main(),
it is not canonicalized, and this aborts when COM is not found.
If COMIND's are exported to init.h, this function will be changed
to accept COMIND directly. */
void
setoptval (const char *com, const char *val, const char *optname)
{
/* Prepend "--" to OPTNAME. */
char *dd_optname = (char *) alloca (2 + strlen (optname) + 1);
dd_optname[0] = '-';
dd_optname[1] = '-';
strcpy (dd_optname + 2, optname);
assert (val != NULL);
if (!setval_internal (command_by_name (com), dd_optname, val))
exit (2);
}
2003-09-21 15:17:45 -04:00
/* Parse OPT into command and value and run it. For example,
run_command("foo=bar") is equivalent to setoptval("foo", "bar").
This is used by the `--execute' flag in main.c. */
void
run_command (const char *opt)
{
char *com, *val;
int comind;
switch (parse_line (opt, &com, &val, &comind))
1999-12-02 02:42:23 -05:00
{
case line_ok:
if (!setval_internal (comind, com, val))
exit (2);
xfree (com);
xfree (val);
break;
default:
2003-09-21 15:17:45 -04:00
fprintf (stderr, _("%s: Invalid --execute command `%s'\n"),
exec_name, opt);
exit (2);
1999-12-02 02:42:23 -05:00
}
}
/* Generic helper functions, for use with `commands'. */
/* Forward declarations: */
struct decode_item {
const char *name;
int code;
};
static bool decode_string (const char *, const struct decode_item *, int, int *);
static bool simple_atoi (const char *, const char *, int *);
static bool simple_atof (const char *, const char *, double *);
2003-09-21 15:17:45 -04:00
#define CMP1(p, c0) (TOLOWER((p)[0]) == (c0) && (p)[1] == '\0')
#define CMP2(p, c0, c1) (TOLOWER((p)[0]) == (c0) \
&& TOLOWER((p)[1]) == (c1) \
&& (p)[2] == '\0')
#define CMP3(p, c0, c1, c2) (TOLOWER((p)[0]) == (c0) \
&& TOLOWER((p)[1]) == (c1) \
&& TOLOWER((p)[2]) == (c2) \
&& (p)[3] == '\0')
1999-12-02 02:42:23 -05:00
2005-04-26 13:30:45 -04:00
/* Store the boolean value from VAL to PLACE. COM is ignored,
1999-12-02 02:42:23 -05:00
except for error messages. */
static bool
2005-04-26 13:30:45 -04:00
cmd_boolean (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
bool value;
2003-09-21 15:17:45 -04:00
if (CMP2 (val, 'o', 'n') || CMP3 (val, 'y', 'e', 's') || CMP1 (val, '1'))
/* "on", "yes" and "1" mean true. */
value = true;
2003-09-21 15:17:45 -04:00
else if (CMP3 (val, 'o', 'f', 'f') || CMP2 (val, 'n', 'o') || CMP1 (val, '0'))
/* "off", "no" and "0" mean false. */
value = false;
1999-12-02 02:42:23 -05:00
else
{
2003-09-21 15:17:45 -04:00
fprintf (stderr,
_("%s: %s: Invalid boolean `%s'; use `on' or `off'.\n"),
2003-09-21 15:17:45 -04:00
exec_name, com, val);
return false;
1999-12-02 02:42:23 -05:00
}
*(bool *) place = value;
return true;
1999-12-02 02:42:23 -05:00
}
2005-04-26 13:30:45 -04:00
/* Set the non-negative integer value from VAL to PLACE. With
1999-12-02 02:42:23 -05:00
incorrect specification, the number remains unchanged. */
static bool
2005-04-26 13:30:45 -04:00
cmd_number (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
2005-04-26 13:30:45 -04:00
if (!simple_atoi (val, val + strlen (val), place)
|| *(int *) place < 0)
1999-12-02 02:42:23 -05:00
{
2003-09-21 15:17:45 -04:00
fprintf (stderr, _("%s: %s: Invalid number `%s'.\n"),
1999-12-02 02:42:23 -05:00
exec_name, com, val);
return false;
1999-12-02 02:42:23 -05:00
}
return true;
1999-12-02 02:42:23 -05:00
}
/* Similar to cmd_number(), only accepts `inf' as a synonym for 0. */
static bool
2005-04-26 13:30:45 -04:00
cmd_number_inf (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
if (!strcasecmp (val, "inf"))
{
*(int *) place = 0;
return true;
1999-12-02 02:42:23 -05:00
}
2005-04-26 13:30:45 -04:00
return cmd_number (com, val, place);
1999-12-02 02:42:23 -05:00
}
/* Copy (strdup) the string at COM to a new location and place a
2005-04-26 13:30:45 -04:00
pointer to *PLACE. */
static bool
2005-04-26 13:30:45 -04:00
cmd_string (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
2005-04-26 13:30:45 -04:00
char **pstring = (char **)place;
1999-12-02 02:42:23 -05:00
xfree_null (*pstring);
1999-12-02 02:42:23 -05:00
*pstring = xstrdup (val);
return true;
1999-12-02 02:42:23 -05:00
}
#ifndef WINDOWS
# define ISSEP(c) ((c) == '/')
#else
# define ISSEP(c) ((c) == '/' || (c) == '\\')
#endif
/* Like the above, but handles tilde-expansion when reading a user's
`.wgetrc'. In that case, and if VAL begins with `~', the tilde
gets expanded to the user's home directory. */
static bool
2005-04-26 13:30:45 -04:00
cmd_file (const char *com, const char *val, void *place)
{
2005-04-26 13:30:45 -04:00
char **pstring = (char **)place;
xfree_null (*pstring);
2005-04-26 13:30:45 -04:00
/* #### If VAL is empty, perhaps should set *PLACE to NULL. */
if (!enable_tilde_expansion || !(*val == '~' && ISSEP (val[1])))
{
noexpand:
*pstring = xstrdup (val);
}
else
{
int homelen;
char *home = home_dir ();
if (!home)
goto noexpand;
homelen = strlen (home);
while (homelen && ISSEP (home[homelen - 1]))
home[--homelen] = '\0';
/* Skip the leading "~/". */
for (++val; ISSEP (*val); val++)
;
*pstring = concat_strings (home, "/", val, (char *) 0);
}
#ifdef WINDOWS
/* Convert "\" to "/". */
{
char *s;
for (s = *pstring; *s; s++)
if (*s == '\\')
*s = '/';
}
#endif
return true;
}
/* Like cmd_file, but strips trailing '/' characters. */
static bool
2005-04-26 13:30:45 -04:00
cmd_directory (const char *com, const char *val, void *place)
{
char *s, *t;
/* Call cmd_file() for tilde expansion and separator
canonicalization (backslash -> slash under Windows). These
things should perhaps be in a separate function. */
2005-04-26 13:30:45 -04:00
if (!cmd_file (com, val, place))
return false;
2005-04-26 13:30:45 -04:00
s = *(char **)place;
t = s + strlen (s);
while (t > s && *--t == '/')
*t = '\0';
return true;
}
2003-09-21 15:17:45 -04:00
/* Split VAL by space to a vector of values, and append those values
2005-04-26 13:30:45 -04:00
to vector pointed to by the PLACE argument. If VAL is empty, the
PLACE vector is cleared instead. */
2003-09-21 15:17:45 -04:00
static bool
2005-04-26 13:30:45 -04:00
cmd_vector (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
2005-04-26 13:30:45 -04:00
char ***pvec = (char ***)place;
1999-12-02 02:42:23 -05:00
if (*val)
*pvec = merge_vecs (*pvec, sepstring (val));
else
{
free_vec (*pvec);
*pvec = NULL;
}
return true;
1999-12-02 02:42:23 -05:00
}
static bool
2005-04-26 13:30:45 -04:00
cmd_directory_vector (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
2005-04-26 13:30:45 -04:00
char ***pvec = (char ***)place;
1999-12-02 02:42:23 -05:00
if (*val)
{
/* Strip the trailing slashes from directories. */
char **t, **seps;
seps = sepstring (val);
for (t = seps; t && *t; t++)
{
int len = strlen (*t);
/* Skip degenerate case of root directory. */
if (len > 1)
{
if ((*t)[len - 1] == '/')
(*t)[len - 1] = '\0';
}
}
*pvec = merge_vecs (*pvec, seps);
}
else
{
free_vec (*pvec);
*pvec = NULL;
}
return true;
1999-12-02 02:42:23 -05:00
}
/* Engine for cmd_bytes and cmd_bytes_sum: converts a string such as
"100k" or "2.5G" to a floating point number. */
1999-12-02 02:42:23 -05:00
static bool
parse_bytes_helper (const char *val, double *result)
1999-12-02 02:42:23 -05:00
{
double number, mult;
const char *end = val + strlen (val);
1999-12-02 02:42:23 -05:00
/* Check for "inf". */
if (0 == strcmp (val, "inf"))
1999-12-02 02:42:23 -05:00
{
*result = 0;
return true;
1999-12-02 02:42:23 -05:00
}
/* Strip trailing whitespace. */
while (val < end && ISSPACE (end[-1]))
--end;
if (val == end)
return false;
switch (TOLOWER (end[-1]))
1999-12-02 02:42:23 -05:00
{
case 'k':
--end, mult = 1024.0;
1999-12-02 02:42:23 -05:00
break;
case 'm':
--end, mult = 1048576.0;
1999-12-02 02:42:23 -05:00
break;
case 'g':
--end, mult = 1073741824.0;
break;
case 't':
--end, mult = 1099511627776.0;
1999-12-02 02:42:23 -05:00
break;
default:
/* Not a recognized suffix: assume it's a digit. (If not,
simple_atof will raise an error.) */
mult = 1;
1999-12-02 02:42:23 -05:00
}
/* Skip leading and trailing whitespace. */
while (val < end && ISSPACE (*val))
++val;
while (val < end && ISSPACE (end[-1]))
--end;
if (val == end)
return false;
if (!simple_atof (val, end, &number) || number < 0)
return false;
*result = number * mult;
return true;
}
2005-04-26 13:30:45 -04:00
/* Parse VAL as a number and set its value to PLACE (which should
point to a wgint).
By default, the value is assumed to be in bytes. If "K", "M", or
"G" are appended, the value is multiplied with 1<<10, 1<<20, or
1<<30, respectively. Floating point values are allowed and are
cast to integer before use. The idea is to be able to use things
like 1.5k instead of "1536".
The string "inf" is returned as 0.
In case of error, false is returned and memory pointed to by PLACE
remains unmodified. */
static bool
2005-04-26 13:30:45 -04:00
cmd_bytes (const char *com, const char *val, void *place)
{
double byte_value;
if (!parse_bytes_helper (val, &byte_value))
{
fprintf (stderr, _("%s: %s: Invalid byte value `%s'\n"),
exec_name, com, val);
return false;
}
2005-04-26 13:30:45 -04:00
*(wgint *)place = (wgint)byte_value;
return true;
}
2005-04-26 13:30:45 -04:00
/* Like cmd_bytes, but PLACE is interpreted as a pointer to
SIZE_SUM. It works by converting the string to double, therefore
working with values up to 2^53-1 without loss of precision. This
value (8192 TB) is large enough to serve for a while. */
static bool
cmd_bytes_sum (const char *com, const char *val, void *place)
{
double byte_value;
if (!parse_bytes_helper (val, &byte_value))
{
fprintf (stderr, _("%s: %s: Invalid byte value `%s'\n"),
exec_name, com, val);
return false;
}
*(SUM_SIZE_INT *) place = (SUM_SIZE_INT) byte_value;
return true;
1999-12-02 02:42:23 -05:00
}
/* Store the value of VAL to *OUT. The value is a time period, by
default expressed in seconds, but also accepting suffixes "m", "h",
"d", and "w" for minutes, hours, days, and weeks respectively. */
static bool
2005-04-26 13:30:45 -04:00
cmd_time (const char *com, const char *val, void *place)
1999-12-02 02:42:23 -05:00
{
double number, mult;
const char *end = val + strlen (val);
/* Strip trailing whitespace. */
while (val < end && ISSPACE (end[-1]))
--end;
if (val == end)
1999-12-02 02:42:23 -05:00
{
err:
2003-09-21 15:17:45 -04:00
fprintf (stderr, _("%s: %s: Invalid time period `%s'\n"),
exec_name, com, val);
return false;
1999-12-02 02:42:23 -05:00
}
switch (TOLOWER (end[-1]))
1999-12-02 02:42:23 -05:00
{
case 's':
--end, mult = 1; /* seconds */
1999-12-02 02:42:23 -05:00
break;
case 'm':
--end, mult = 60; /* minutes */
1999-12-02 02:42:23 -05:00
break;
case 'h':
--end, mult = 3600; /* hours */
1999-12-02 02:42:23 -05:00
break;
case 'd':
--end, mult = 86400.0; /* days */
1999-12-02 02:42:23 -05:00
break;
case 'w':
--end, mult = 604800.0; /* weeks */
1999-12-02 02:42:23 -05:00
break;
default:
/* Not a recognized suffix: assume it belongs to the number.
2005-04-26 17:10:24 -04:00
(If not, simple_atof will raise an error.) */
mult = 1;
}
/* Skip leading and trailing whitespace. */
while (val < end && ISSPACE (*val))
++val;
while (val < end && ISSPACE (end[-1]))
--end;
if (val == end)
goto err;
if (!simple_atof (val, end, &number))
goto err;
2005-04-26 13:30:45 -04:00
*(double *)place = number * mult;
return true;
1999-12-02 02:42:23 -05:00
}
#ifdef HAVE_SSL
static bool
cmd_cert_type (const char *com, const char *val, void *place)
{
static const struct decode_item choices[] = {
{ "pem", keyfile_pem },
{ "der", keyfile_asn1 },
{ "asn1", keyfile_asn1 },
};
2005-04-26 13:30:45 -04:00
int ok = decode_string (val, choices, countof (choices), place);
if (!ok)
fprintf (stderr, _("%s: %s: Invalid value `%s'.\n"), exec_name, com, val);
return ok;
}
#endif
/* Specialized helper functions, used by `commands' to handle some
options specially. */
static bool check_user_specified_header (const char *);
1999-12-02 02:42:23 -05:00
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_dirstruct (const char *com, const char *val, void *place_ignored)
1999-12-02 02:42:23 -05:00
{
if (!cmd_boolean (com, val, &opt.dirstruct))
return false;
1999-12-02 02:42:23 -05:00
/* Since dirstruct behaviour is explicitly changed, no_dirstruct
must be affected inversely. */
if (opt.dirstruct)
opt.no_dirstruct = false;
1999-12-02 02:42:23 -05:00
else
opt.no_dirstruct = true;
return true;
1999-12-02 02:42:23 -05:00
}
static bool
2005-05-30 09:31:24 -04:00
cmd_spec_header (const char *com, const char *val, void *place_ignored)
1999-12-02 02:42:23 -05:00
{
2005-05-30 09:31:24 -04:00
/* Empty value means reset the list of headers. */
if (*val == '\0')
{
free_vec (opt.user_headers);
opt.user_headers = NULL;
return true;
2005-05-30 09:31:24 -04:00
}
if (!check_user_specified_header (val))
1999-12-02 02:42:23 -05:00
{
fprintf (stderr, _("%s: %s: Invalid header `%s'.\n"),
exec_name, com, val);
return false;
1999-12-02 02:42:23 -05:00
}
2005-05-30 09:31:24 -04:00
opt.user_headers = vec_append (opt.user_headers, val);
return true;
1999-12-02 02:42:23 -05:00
}
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_htmlify (const char *com, const char *val, void *place_ignored)
1999-12-02 02:42:23 -05:00
{
int flag = cmd_boolean (com, val, &opt.htmlify);
if (flag && !opt.htmlify)
opt.remove_listing = false;
1999-12-02 02:42:23 -05:00
return flag;
}
2003-09-21 15:17:45 -04:00
/* Set the "mirror" mode. It means: recursive download, timestamping,
no limit on max. recursion depth, and don't remove listings. */
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_mirror (const char *com, const char *val, void *place_ignored)
1999-12-02 02:42:23 -05:00
{
int mirror;
if (!cmd_boolean (com, val, &mirror))
return false;
1999-12-02 02:42:23 -05:00
if (mirror)
{
opt.recursive = true;
1999-12-02 02:42:23 -05:00
if (!opt.no_dirstruct)
opt.dirstruct = true;
opt.timestamping = true;
opt.reclevel = INFINITE_RECURSION;
opt.remove_listing = false;
1999-12-02 02:42:23 -05:00
}
return true;
1999-12-02 02:42:23 -05:00
}
/* Validate --prefer-family and set the choice. Allowed values are
"IPv4", "IPv6", and "none". */
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_prefer_family (const char *com, const char *val, void *place_ignored)
{
static const struct decode_item choices[] = {
{ "IPv4", prefer_ipv4 },
{ "IPv6", prefer_ipv6 },
{ "none", prefer_none },
};
int ok = decode_string (val, choices, countof (choices),
(int *) &opt.prefer_family);
if (!ok)
fprintf (stderr, _("%s: %s: Invalid value `%s'.\n"), exec_name, com, val);
return ok;
}
2003-09-21 15:17:45 -04:00
/* Set progress.type to VAL, but verify that it's a valid progress
implementation before that. */
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_progress (const char *com, const char *val, void *place_ignored)
{
if (!valid_progress_implementation_p (val))
{
fprintf (stderr, _("%s: %s: Invalid progress type `%s'.\n"),
exec_name, com, val);
return false;
}
xfree_null (opt.progress_type);
/* Don't call set_progress_implementation here. It will be called
in main() when it becomes clear what the log output is. */
opt.progress_type = xstrdup (val);
return true;
}
2003-09-21 15:17:45 -04:00
/* Set opt.recursive to VAL as with cmd_boolean. If opt.recursive is
set to true, also set opt.dirstruct to true, unless opt.no_dirstruct
2003-09-21 15:17:45 -04:00
is specified. */
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_recursive (const char *com, const char *val, void *place_ignored)
1999-12-02 02:42:23 -05:00
{
if (!cmd_boolean (com, val, &opt.recursive))
return false;
1999-12-02 02:42:23 -05:00
else
{
if (opt.recursive && !opt.no_dirstruct)
opt.dirstruct = true;
1999-12-02 02:42:23 -05:00
}
return true;
1999-12-02 02:42:23 -05:00
}
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_restrict_file_names (const char *com, const char *val, void *place_ignored)
{
int restrict_os = opt.restrict_files_os;
int restrict_ctrl = opt.restrict_files_ctrl;
const char *end = strchr (val, ',');
if (!end)
end = val + strlen (val);
#define VAL_IS(string_literal) BOUNDED_EQUAL (val, end, string_literal)
if (VAL_IS ("unix"))
restrict_os = restrict_unix;
else if (VAL_IS ("windows"))
restrict_os = restrict_windows;
else if (VAL_IS ("nocontrol"))
restrict_ctrl = 0;
else
{
err:
2003-09-21 15:17:45 -04:00
fprintf (stderr,
_("%s: %s: Invalid restriction `%s', use `unix' or `windows'.\n"),
exec_name, com, val);
return false;
}
#undef VAL_IS
if (*end)
{
if (!strcmp (end + 1, "nocontrol"))
restrict_ctrl = false;
else
goto err;
}
opt.restrict_files_os = restrict_os;
opt.restrict_files_ctrl = restrict_ctrl;
return true;
}
#ifdef HAVE_SSL
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_secure_protocol (const char *com, const char *val, void *place)
{
static const struct decode_item choices[] = {
{ "auto", secure_protocol_auto },
{ "sslv2", secure_protocol_sslv2 },
{ "sslv3", secure_protocol_sslv3 },
{ "tlsv1", secure_protocol_tlsv1 },
};
2005-04-26 13:30:45 -04:00
int ok = decode_string (val, choices, countof (choices), place);
if (!ok)
fprintf (stderr, _("%s: %s: Invalid value `%s'.\n"), exec_name, com, val);
return ok;
}
#endif
/* Set all three timeout values. */
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_timeout (const char *com, const char *val, void *place_ignored)
{
double value;
if (!cmd_time (com, val, &value))
return false;
opt.read_timeout = value;
opt.connect_timeout = value;
opt.dns_timeout = value;
return true;
}
static bool
2005-04-26 13:30:45 -04:00
cmd_spec_useragent (const char *com, const char *val, void *place_ignored)
1999-12-02 02:42:23 -05:00
{
/* Disallow embedded newlines. */
if (strchr (val, '\n'))
1999-12-02 02:42:23 -05:00
{
2003-09-21 15:17:45 -04:00
fprintf (stderr, _("%s: %s: Invalid value `%s'.\n"),
1999-12-02 02:42:23 -05:00
exec_name, com, val);
return false;
1999-12-02 02:42:23 -05:00
}
xfree_null (opt.useragent);
1999-12-02 02:42:23 -05:00
opt.useragent = xstrdup (val);
return true;
1999-12-02 02:42:23 -05:00
}
/* The "verbose" option cannot be cmd_boolean because the variable is
not bool -- it's of type int (-1 means uninitialized because of
some random hackery for disallowing -q -v). */
static bool
cmd_spec_verbose (const char *com, const char *val, void *place_ignored)
{
bool flag;
if (cmd_boolean (com, val, &flag))
{
opt.verbose = flag;
return true;
}
return false;
}
1999-12-02 02:42:23 -05:00
/* Miscellaneous useful routines. */
/* A very simple atoi clone, more useful than atoi because it works on
delimited strings, and has error reportage. Returns true on success,
false on failure. If successful, stores result to *DEST. */
2003-09-21 15:17:45 -04:00
static bool
2003-09-21 15:17:45 -04:00
simple_atoi (const char *beg, const char *end, int *dest)
1999-12-02 02:42:23 -05:00
{
2003-09-21 15:17:45 -04:00
int result = 0;
bool negative = false;
const char *p = beg;
1999-12-02 02:42:23 -05:00
while (p < end && ISSPACE (*p))
++p;
if (p < end && (*p == '-' || *p == '+'))
{
negative = (*p == '-');
++p;
}
if (p == end)
return false;
2003-09-21 15:17:45 -04:00
/* Read negative numbers in a separate loop because the most
negative integer cannot be represented as a positive number. */
if (!negative)
for (; p < end && ISDIGIT (*p); p++)
{
int next = (10 * result) + (*p - '0');
if (next < result)
return false; /* overflow */
result = next;
}
else
for (; p < end && ISDIGIT (*p); p++)
{
int next = (10 * result) - (*p - '0');
if (next > result)
return false; /* underflow */
result = next;
}
2003-09-21 15:17:45 -04:00
if (p != end)
return false;
2003-09-21 15:17:45 -04:00
*dest = result;
return true;
2003-09-21 15:17:45 -04:00
}
/* Trivial atof, with error reporting. Handles "<digits>[.<digits>]",
doesn't handle exponential notation. Returns true on success,
false on failure. In case of success, stores its result to
*DEST. */
2003-09-21 15:17:45 -04:00
static bool
2003-09-21 15:17:45 -04:00
simple_atof (const char *beg, const char *end, double *dest)
{
double result = 0;
bool negative = false;
bool seen_dot = false;
bool seen_digit = false;
2003-09-21 15:17:45 -04:00
double divider = 1;
const char *p = beg;
while (p < end && ISSPACE (*p))
++p;
if (p < end && (*p == '-' || *p == '+'))
{
negative = (*p == '-');
++p;
}
2003-09-21 15:17:45 -04:00
for (; p < end; p++)
2003-09-21 15:17:45 -04:00
{
char ch = *p;
if (ISDIGIT (ch))
{
if (!seen_dot)
result = (10 * result) + (ch - '0');
else
result += (ch - '0') / (divider *= 10);
seen_digit = true;
2003-09-21 15:17:45 -04:00
}
else if (ch == '.')
{
if (!seen_dot)
seen_dot = true;
2003-09-21 15:17:45 -04:00
else
return false;
2003-09-21 15:17:45 -04:00
}
else
return false;
2003-09-21 15:17:45 -04:00
}
if (!seen_digit)
return false;
if (negative)
result = -result;
2003-09-21 15:17:45 -04:00
*dest = result;
return true;
1999-12-02 02:42:23 -05:00
}
/* Verify that the user-specified header in S is valid. It must
contain a colon preceded by non-white-space characters and must not
contain newlines. */
static bool
1999-12-02 02:42:23 -05:00
check_user_specified_header (const char *s)
{
const char *p;
for (p = s; *p && *p != ':' && !ISSPACE (*p); p++)
;
1999-12-02 02:42:23 -05:00
/* The header MUST contain `:' preceded by at least one
non-whitespace character. */
if (*p != ':' || p == s)
return false;
1999-12-02 02:42:23 -05:00
/* The header MUST NOT contain newlines. */
if (strchr (s, '\n'))
return false;
return true;
1999-12-02 02:42:23 -05:00
}
/* Decode VAL into a number, according to ITEMS. */
static bool
decode_string (const char *val, const struct decode_item *items, int itemcount,
int *place)
{
int i;
for (i = 0; i < itemcount; i++)
if (0 == strcasecmp (val, items[i].name))
{
*place = items[i].code;
return true;
}
return false;
}
1999-12-02 02:42:23 -05:00
2005-06-19 18:34:58 -04:00
void cleanup_html_url (void);
1999-12-02 02:42:23 -05:00
/* Free the memory allocated by global variables. */
void
cleanup (void)
{
/* Free external resources, close files, etc. */
1999-12-02 02:42:23 -05:00
if (output_stream)
fclose (output_stream);
/* No need to check for error because Wget flushes its output (and
checks for errors) after any data arrives. */
/* We're exiting anyway so there's no real need to call free()
hundreds of times. Skipping the frees will make Wget exit
faster.
However, when detecting leaks, it's crucial to free() everything
because then you can find the real leaks, i.e. the allocated
memory which grows with the size of the program. */
#ifdef DEBUG_MALLOC
convert_cleanup ();
res_cleanup ();
http_cleanup ();
cleanup_html_url ();
host_cleanup ();
log_cleanup ();
{
extern acc_t *netrc_list;
free_netrc (netrc_list);
}
xfree_null (opt.lfilename);
xfree_null (opt.dir_prefix);
xfree_null (opt.input_filename);
xfree_null (opt.output_document);
1999-12-02 02:42:23 -05:00
free_vec (opt.accepts);
free_vec (opt.rejects);
free_vec (opt.excludes);
free_vec (opt.includes);
free_vec (opt.domains);
[svn] * TODO: Removed done item: we now have an option (-G) that makes it easy to download a single HTML document and all its constituents. * po/*.{gmo,po,pot}: Regenerated after adding new options. * po/hr.po: Hrvoje forgot '\n's on his translations of my altered messages, causing msgfmt to balk and `make install' to fail. * wget.texi (Recursive Retrieval Options): In -K description, added a link to the discussion of interaction with -N. (Recursive Accept/Reject Options): Did some alphabetizing and added descriptions of new --follow-tags and -G / --ignore-tags options. (Following Links): Changed "the loads of" to "loads of". (Wgetrc Commands): Added descriptions of new follow_tags and ignore_tags commands. * html.c (idmatch): Implemented checking of my new --follow-tags and --ignore-tags options. * init.c (commands): Added comment reminding people adding new entries doing allocation to add corresponding freeing in cleanup(). (commands): Added new followtags and ignoretags commands. (cleanup): Free storage for new followtags and ignoretags. * main.c: Use of "comma-separated list" was random -- normalized it. Did some alphabetization. Added comments pointing out "Options without arguments" and "Options accepting an argument" sections of long_options[]. Added new options --follow-tags and -G / --ignore-tags. Added comment that Damir's --referer is currently undocumented. Added comment that Heiko's --waitretry is partially undocumented (mentioned in --help but not in wget.texi). Moved improperly sorted 24, 129, and 'G' cases. * options.h (struct options): Added new fields follow_tags and ignore_tags. * wget.h: Added "#define EQ 0" so we can say "strcmp(a, b) == EQ".
2000-03-11 01:48:06 -05:00
free_vec (opt.follow_tags);
free_vec (opt.ignore_tags);
xfree_null (opt.progress_type);
xfree_null (opt.ftp_user);
xfree_null (opt.ftp_passwd);
xfree_null (opt.ftp_proxy);
xfree_null (opt.https_proxy);
xfree_null (opt.http_proxy);
1999-12-02 02:42:23 -05:00
free_vec (opt.no_proxy);
xfree_null (opt.useragent);
xfree_null (opt.referer);
xfree_null (opt.http_user);
xfree_null (opt.http_passwd);
free_vec (opt.user_headers);
# ifdef HAVE_SSL
xfree_null (opt.cert_file);
xfree_null (opt.private_key);
xfree_null (opt.ca_directory);
xfree_null (opt.ca_cert);
xfree_null (opt.random_file);
xfree_null (opt.egd_file);
# endif
xfree_null (opt.bind_address);
xfree_null (opt.cookies_input);
xfree_null (opt.cookies_output);
xfree_null (opt.user);
xfree_null (opt.passwd);
#endif /* DEBUG_MALLOC */
1999-12-02 02:42:23 -05:00
}