/* Command line parsing.
Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
- 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation,
+ 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation,
Inc.
This file is part of GNU Wget.
#include "http.h" /* for save_cookies */
#include "ptimer.h"
#include "warc.h"
-
#include <getopt.h>
#include <getpass.h>
#include <quote.h>
static struct cmdline_option option_data[] =
{
{ "accept", 'A', OPT_VALUE, "accept", -1 },
+ { "accept-regex", 0, OPT_VALUE, "acceptregex", -1 },
{ "adjust-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 },
{ "append-output", 'a', OPT__APPEND_OUTPUT, NULL, required_argument },
{ "ask-password", 0, OPT_BOOLEAN, "askpassword", -1 },
{ "backups", 0, OPT_BOOLEAN, "backups", -1 },
{ "base", 'B', OPT_VALUE, "base", -1 },
{ "bind-address", 0, OPT_VALUE, "bindaddress", -1 },
+ { "body-data", 0, OPT_VALUE, "bodydata", -1 },
+ { "body-file", 0, OPT_VALUE, "bodyfile", -1 },
{ IF_SSL ("ca-certificate"), 0, OPT_VALUE, "cacertificate", -1 },
{ IF_SSL ("ca-directory"), 0, OPT_VALUE, "cadirectory", -1 },
{ "cache", 0, OPT_BOOLEAN, "cache", -1 },
{ "load-cookies", 0, OPT_VALUE, "loadcookies", -1 },
{ "local-encoding", 0, OPT_VALUE, "localencoding", -1 },
{ "max-redirect", 0, OPT_VALUE, "maxredirect", -1 },
+ { "method", 0, OPT_VALUE, "method", -1 },
{ "mirror", 'm', OPT_BOOLEAN, "mirror", -1 },
{ "no", 'n', OPT__NO, NULL, required_argument },
{ "no-clobber", 0, OPT_BOOLEAN, "noclobber", -1 },
{ "read-timeout", 0, OPT_VALUE, "readtimeout", -1 },
{ "recursive", 'r', OPT_BOOLEAN, "recursive", -1 },
{ "referer", 0, OPT_VALUE, "referer", -1 },
+ { "regex-type", 0, OPT_VALUE, "regextype", -1 },
{ "reject", 'R', OPT_VALUE, "reject", -1 },
+ { "reject-regex", 0, OPT_VALUE, "rejectregex", -1 },
{ "relative", 'L', OPT_BOOLEAN, "relativeonly", -1 },
{ "remote-encoding", 0, OPT_VALUE, "remoteencoding", -1 },
{ "remove-listing", 0, OPT_BOOLEAN, "removelisting", -1 },
+ { "report-speed", 0, OPT_BOOLEAN, "reportspeed", -1 },
{ "restrict-file-names", 0, OPT_BOOLEAN, "restrictfilenames", -1 },
{ "retr-symlinks", 0, OPT_BOOLEAN, "retrsymlinks", -1 },
{ "retry-connrefused", 0, OPT_BOOLEAN, "retryconnrefused", -1 },
{ "wait", 'w', OPT_VALUE, "wait", -1 },
{ "waitretry", 0, OPT_VALUE, "waitretry", -1 },
{ "warc-cdx", 0, OPT_BOOLEAN, "warccdx", -1 },
+#ifdef HAVE_LIBZ
{ "warc-compression", 0, OPT_BOOLEAN, "warccompression", -1 },
+#endif
{ "warc-dedup", 0, OPT_VALUE, "warccdxdedup", -1 },
{ "warc-digests", 0, OPT_BOOLEAN, "warcdigests", -1 },
{ "warc-file", 0, OPT_VALUE, "warcfile", -1 },
-v, --verbose be verbose (this is the default).\n"),
N_("\
-nv, --no-verbose turn off verboseness, without being quiet.\n"),
+ N_("\
+ --report-speed=TYPE Output bandwidth as TYPE. TYPE can be bits.\n"),
N_("\
-i, --input-file=FILE download URLs found in local or external FILE.\n"),
N_("\
--post-data=STRING use the POST method; send STRING as the data.\n"),
N_("\
--post-file=FILE use the POST method; send contents of FILE.\n"),
+ N_("\
+ --method=HTTPMethod use method \"HTTPMethod\" in the header.\n"),
+ N_("\
+ --body-data=STRING Send STRING as data. --method MUST be set.\n"),
+ N_("\
+ --body-file=FILE Send contents of FILE. --method MUST be set.\n"),
N_("\
--content-disposition honor the Content-Disposition header when\n\
choosing local file names (EXPERIMENTAL).\n"),
--warc-cdx write CDX index files.\n"),
N_("\
--warc-dedup=FILENAME do not store records listed in this CDX file.\n"),
+#ifdef HAVE_LIBZ
N_("\
--no-warc-compression do not compress WARC files with GZIP.\n"),
+#endif
N_("\
--no-warc-digests do not calculate SHA1 digests.\n"),
N_("\
-A, --accept=LIST comma-separated list of accepted extensions.\n"),
N_("\
-R, --reject=LIST comma-separated list of rejected extensions.\n"),
+ N_("\
+ --accept-regex=REGEX regex matching accepted URLs.\n"),
+ N_("\
+ --reject-regex=REGEX regex matching rejected URLs.\n"),
+#ifdef HAVE_LIBPCRE
+ N_("\
+ --regex-type=TYPE regex type (posix|pcre).\n"),
+#else
+ N_("\
+ --regex-type=TYPE regex type (posix).\n"),
+#endif
N_("\
-D, --domains=LIST comma-separated list of accepted domains.\n"),
N_("\
N_("\
-np, --no-parent don't ascend to the parent directory.\n"),
"\n",
-
N_("Mail bug reports and suggestions to <bug-wget@gnu.org>.\n")
};
for (i = 1; i < argc; i++)
argstring_length += strlen (argv[i]) + 2 + 1;
char *p = program_argstring = malloc (argstring_length * sizeof (char));
+ if (p == NULL)
+ {
+ fprintf (stderr, _("Memory allocation problem\n"));
+ exit (2);
+ }
for (i = 1; i < argc; i++)
- {
- *p++ = '"';
- int arglen = strlen (argv[i]);
- memcpy (p, argv[i], arglen);
- p += arglen;
- *p++ = '"';
- *p++ = ' ';
- }
+ {
+ *p++ = '"';
+ int arglen = strlen (argv[i]);
+ memcpy (p, argv[i], arglen);
+ p += arglen;
+ *p++ = '"';
+ *p++ = ' ';
+ }
*p = '\0';
/* Load the hard-coded defaults. */
short_options, long_options, &longindex)) != -1)
{
int confval;
- bool userrc_ret = true;
struct cmdline_option *config_opt;
/* There is no short option for "--config". */
config_opt = &option_data[confval & ~BOOLEAN_NEG_MARKER];
if (strcmp (config_opt->long_name, "config") == 0)
{
+ bool userrc_ret = true;
userrc_ret &= run_wgetrc (optarg);
use_userconfig = true;
+ if (userrc_ret)
+ break;
+ else
+ {
+ fprintf (stderr, _("Exiting due to error in %s\n"), optarg);
+ exit (2);
+ }
}
- if (!userrc_ret)
- {
- printf ("Exiting due to error in %s\n", optarg);
- exit (2);
- }
- else
- break;
}
}
{
if (ret == '?')
{
- print_usage (0);
- printf ("\n");
- printf (_("Try `%s --help' for more options.\n"), exec_name);
+ print_usage (1);
+ fprintf (stderr, "\n");
+ fprintf (stderr, _("Try `%s --help' for more options.\n"),
+ exec_name);
exit (2);
}
/* Find the short option character in the mapping. */
{
fprintf (stderr,
_("Both --no-clobber and --convert-links were specified,"
- "only --convert-links will be used.\n"));
+ " only --convert-links will be used.\n"));
opt.noclobber = false;
}
if (opt.verbose == -1)
opt.verbose = !opt.quiet;
+
/* Sanity checks. */
if (opt.verbose && opt.quiet)
{
}
if (opt.warc_keep_log)
{
- opt.progress_type = "dot";
+ opt.progress_type = xstrdup ("dot");
}
}
/* No URL specified. */
fprintf (stderr, _("%s: missing URL\n"), exec_name);
print_usage (1);
- printf ("\n");
+ fprintf (stderr, "\n");
/* #### Something nicer should be printed here -- similar to the
pre-1.5 `--help' page. */
fprintf (stderr, _("Try `%s --help' for more options.\n"), exec_name);
exit (1);
}
+ /* Compile the regular expressions. */
+ switch (opt.regex_type)
+ {
+#ifdef HAVE_LIBPCRE
+ case regex_type_pcre:
+ opt.regex_compile_fun = compile_pcre_regex;
+ opt.regex_match_fun = match_pcre_regex;
+ break;
+#endif
+
+ case regex_type_posix:
+ default:
+ opt.regex_compile_fun = compile_posix_regex;
+ opt.regex_match_fun = match_posix_regex;
+ break;
+ }
+ if (opt.acceptregex_s)
+ {
+ opt.acceptregex = opt.regex_compile_fun (opt.acceptregex_s);
+ if (!opt.acceptregex)
+ exit (1);
+ }
+ if (opt.rejectregex_s)
+ {
+ opt.rejectregex = opt.regex_compile_fun (opt.rejectregex_s);
+ if (!opt.rejectregex)
+ exit (1);
+ }
+ if (opt.post_data || opt.post_file_name)
+ {
+ if (opt.post_data && opt.post_file_name)
+ {
+ fprintf (stderr, _("You cannot specify both --post-data and --post-file.\n"));
+ exit (1);
+ }
+ else if (opt.method)
+ {
+ fprintf (stderr, _("You cannot use --post-data or --post-file along with --method. "
+ "--method expects data through --body-data and --body-file options"));
+ exit (1);
+ }
+ }
+ if (opt.body_data || opt.body_file)
+ {
+ if (!opt.method)
+ {
+ fprintf (stderr, _("You must specify a method through --method=HTTPMethod "
+ "to use with --body-data or --body-file.\n"));
+ exit (1);
+ }
+ else if (opt.body_data && opt.body_file)
+ {
+ fprintf (stderr, _("You cannot specify both --body-data and --body-file.\n"));
+ exit (1);
+ }
+ }
+
+ /* Convert post_data to body-data and post_file_name to body-file options.
+ This is required so as to remove redundant code later on in gethttp().
+ The --post-data and --post-file options may also be removed in
+ the future hence it makes sense to convert them to aliases for
+ the more generic --method options.
+ This MUST occur only after the sanity checks so as to prevent the
+ user from setting both post and body options simultaneously.
+ */
+ if (opt.post_data || opt.post_file_name)
+ {
+ setoptval ("method", "POST", "method");
+ if (opt.post_data)
+ {
+ setoptval ("bodydata", opt.post_data, "body-data");
+ opt.post_data = NULL;
+ }
+ else
+ {
+ setoptval ("bodyfile", opt.post_file_name, "body-file");
+ opt.post_file_name = NULL;
+ }
+ }
+
#ifdef ENABLE_IRI
if (opt.enable_iri)
{
/* Fill in the arguments. */
url = alloca_array (char *, nurl + 1);
+ if (url == NULL)
+ {
+ fprintf (stderr, _("Memory allocation problem\n"));
+ exit (2);
+ }
for (i = 0; i < nurl; i++, optind++)
{
char *rewritten = rewrite_shorthand_url (argv[optind]);
&dt, opt.recursive, iri, true);
}
- if (opt.delete_after && file_exists_p(filename))
+ if (opt.delete_after && filename != NULL && file_exists_p (filename))
{
DEBUGP (("Removing file due to --delete-after in main():\n"));
logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
if (opt.convert_links && !opt.delete_after)
convert_all_links ();
- /* Close WARC file. */
- if (opt.warc_filename != 0)
- warc_close ();
-
- log_close ();
-
- for (i = 0; i < nurl; i++)
- xfree (url[i]);
cleanup ();
exit (get_exit_status ());