/* Command line parsing.
- 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>
-#ifdef HAVE_UNISTD_H
-# include <unistd.h>
-#endif /* HAVE_UNISTD_H */
+#include <unistd.h>
#include <string.h>
#include <signal.h>
#ifdef ENABLE_NLS
#include <errno.h>
#include <time.h>
+#include "exits.h"
#include "utils.h"
#include "init.h"
#include "retr.h"
#include "convert.h"
#include "spider.h"
#include "http.h" /* for save_cookies */
-
+#include "ptimer.h"
+#include "warc.h"
+#include "closeout.h"
#include <getopt.h>
#include <getpass.h>
#include <quote.h>
+#ifdef WINDOWS
+# include <io.h>
+# include <fcntl.h>
+#endif
+
+#ifdef __VMS
+# include "vms.h"
+#endif /* __VMS */
+
#ifndef PATH_SEPARATOR
# define PATH_SEPARATOR '/'
#endif
+#ifndef ENABLE_IRI
+struct iri dummy_iri;
+#endif
+
struct options opt;
/* defined in version.c */
extern char *system_getrc;
extern char *link_string;
/* defined in build_info.c */
-extern char *compiled_features[];
+extern const char *compiled_features[];
/* Used for --version output in print_version */
-static const int max_chars_per_line = 72;
+#define MAX_CHARS_PER_LINE 72
+#define TABULATION 4
#if defined(SIGHUP) || defined(SIGUSR1)
static void redirect_output_signal (int);
#endif
const char *exec_name;
+
+/* Number of successfully downloaded URLs */
+int numurls = 0;
\f
#ifndef TESTING
/* Initialize I18N/L10N. That amounts to invoking setlocale, and
static struct cmdline_option option_data[] =
{
{ "accept", 'A', OPT_VALUE, "accept", -1 },
+ { "adjust-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 },
{ "append-output", 'a', OPT__APPEND_OUTPUT, NULL, required_argument },
{ "ask-password", 0, OPT_BOOLEAN, "askpassword", -1 },
{ "auth-no-challenge", 0, OPT_BOOLEAN, "authnochallenge", -1 },
{ "backups", 0, OPT_BOOLEAN, "backups", -1 },
{ "base", 'B', OPT_VALUE, "base", -1 },
{ "bind-address", 0, OPT_VALUE, "bindaddress", -1 },
+ { "bits", 0, OPT_BOOLEAN, "bits", -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 },
{ IF_SSL ("certificate-type"), 0, OPT_VALUE, "certificatetype", -1 },
{ IF_SSL ("check-certificate"), 0, OPT_BOOLEAN, "checkcertificate", -1 },
{ "clobber", 0, OPT__CLOBBER, NULL, optional_argument },
+ { "config", 0, OPT_VALUE, "chooseconfig", -1 },
{ "connect-timeout", 0, OPT_VALUE, "connecttimeout", -1 },
{ "continue", 'c', OPT_BOOLEAN, "continue", -1 },
{ "convert-links", 'k', OPT_BOOLEAN, "convertlinks", -1 },
{ "content-disposition", 0, OPT_BOOLEAN, "contentdisposition", -1 },
+ { "content-on-error", 0, OPT_BOOLEAN, "contentonerror", -1 },
{ "cookies", 0, OPT_BOOLEAN, "cookies", -1 },
{ "cut-dirs", 0, OPT_VALUE, "cutdirs", -1 },
{ WHEN_DEBUG ("debug"), 'd', OPT_BOOLEAN, "debug", -1 },
{ "dns-timeout", 0, OPT_VALUE, "dnstimeout", -1 },
{ "domains", 'D', OPT_VALUE, "domains", -1 },
{ "dont-remove-listing", 0, OPT__DONT_REMOVE_LISTING, NULL, no_argument },
- { "dot-style", 0, OPT_VALUE, "dotstyle", -1 },
+ { "dot-style", 0, OPT_VALUE, "dotstyle", -1 }, /* deprecated */
{ "egd-file", 0, OPT_VALUE, "egdfile", -1 },
{ "exclude-directories", 'X', OPT_VALUE, "excludedirectories", -1 },
{ "exclude-domains", 0, OPT_VALUE, "excludedomains", -1 },
{ "force-directories", 'x', OPT_BOOLEAN, "dirstruct", -1 },
{ "force-html", 'F', OPT_BOOLEAN, "forcehtml", -1 },
{ "ftp-password", 0, OPT_VALUE, "ftppassword", -1 },
+#ifdef __VMS
+ { "ftp-stmlf", 0, OPT_BOOLEAN, "ftpstmlf", -1 },
+#endif /* def __VMS */
{ "ftp-user", 0, OPT_VALUE, "ftpuser", -1 },
{ "glob", 0, OPT_BOOLEAN, "glob", -1 },
{ "header", 0, OPT_VALUE, "header", -1 },
{ "help", 'h', OPT_FUNCALL, (void *)print_help, no_argument },
{ "host-directories", 0, OPT_BOOLEAN, "addhostdir", -1 },
- { "html-extension", 'E', OPT_BOOLEAN, "htmlextension", -1 },
+ { "html-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 }, /* deprecated */
{ "htmlify", 0, OPT_BOOLEAN, "htmlify", -1 },
{ "http-keep-alive", 0, OPT_BOOLEAN, "httpkeepalive", -1 },
{ "http-passwd", 0, OPT_VALUE, "httppassword", -1 }, /* deprecated */
{ "inet6-only", '6', OPT_BOOLEAN, "inet6only", -1 },
#endif
{ "input-file", 'i', OPT_VALUE, "input", -1 },
+ { "iri", 0, OPT_BOOLEAN, "iri", -1 },
{ "keep-session-cookies", 0, OPT_BOOLEAN, "keepsessioncookies", -1 },
{ "level", 'l', OPT_VALUE, "reclevel", -1 },
{ "limit-rate", 0, OPT_VALUE, "limitrate", -1 },
{ "load-cookies", 0, OPT_VALUE, "loadcookies", -1 },
+ { "local-encoding", 0, OPT_VALUE, "localencoding", -1 },
{ "max-redirect", 0, OPT_VALUE, "maxredirect", -1 },
{ "mirror", 'm', OPT_BOOLEAN, "mirror", -1 },
{ "no", 'n', OPT__NO, NULL, required_argument },
{ "referer", 0, OPT_VALUE, "referer", -1 },
{ "reject", 'R', OPT_VALUE, "reject", -1 },
{ "relative", 'L', OPT_BOOLEAN, "relativeonly", -1 },
+ { "remote-encoding", 0, OPT_VALUE, "remoteencoding", -1 },
{ "remove-listing", 0, OPT_BOOLEAN, "removelisting", -1 },
{ "restrict-file-names", 0, OPT_BOOLEAN, "restrictfilenames", -1 },
{ "retr-symlinks", 0, OPT_BOOLEAN, "retrsymlinks", -1 },
{ "timeout", 'T', OPT_VALUE, "timeout", -1 },
{ "timestamping", 'N', OPT_BOOLEAN, "timestamping", -1 },
{ "tries", 't', OPT_VALUE, "tries", -1 },
+ { "unlink", 0, OPT_BOOLEAN, "unlink", -1 },
+ { "trust-server-names", 0, OPT_BOOLEAN, "trustservernames", -1 },
+ { "use-server-timestamps", 0, OPT_BOOLEAN, "useservertimestamps", -1 },
{ "user", 0, OPT_VALUE, "user", -1 },
{ "user-agent", 'U', OPT_VALUE, "useragent", -1 },
{ "verbose", 'v', OPT_BOOLEAN, "verbose", -1 },
{ "version", 'V', OPT_FUNCALL, (void *) print_version, no_argument },
{ "wait", 'w', OPT_VALUE, "wait", -1 },
{ "waitretry", 0, OPT_VALUE, "waitretry", -1 },
-#ifdef MSDOS
+ { "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 },
+ { "warc-header", 0, OPT_VALUE, "warcheader", -1 },
+ { "warc-keep-log", 0, OPT_BOOLEAN, "warckeeplog", -1 },
+ { "warc-max-size", 0, OPT_VALUE, "warcmaxsize", -1 },
+ { "warc-tempdir", 0, OPT_VALUE, "warctempdir", -1 },
+#ifdef USE_WATT32
{ "wdebug", 0, OPT_BOOLEAN, "wdebug", -1 },
#endif
};
}
/* Print the usage message. */
-static void
-print_usage (void)
+static int
+print_usage (int error)
{
- printf (_("Usage: %s [OPTION]... [URL]...\n"), exec_name);
+ return fprintf (error ? stderr : stdout,
+ _("Usage: %s [OPTION]... [URL]...\n"), exec_name);
}
/* Print the help message, describing all the available options. If
N_("\
-d, --debug print lots of debugging information.\n"),
#endif
-#ifdef MSDOS
+#ifdef USE_WATT32
N_("\
--wdebug print Watt-32 debug output.\n"),
#endif
N_("\
-F, --force-html treat input file as HTML.\n"),
N_("\
- -B, --base=URL prepends URL to relative links in -F -i file.\n"),
+ -B, --base=URL resolves HTML input-file links (-i -F)\n\
+ relative to URL.\n"),
+ N_("\
+ --config=FILE Specify config file to use.\n"),
"\n",
N_("\
-O, --output-document=FILE write documents to FILE.\n"),
N_("\
-nc, --no-clobber skip downloads that would download to\n\
- existing files.\n"),
+ existing files (overwriting them).\n"),
N_("\
-c, --continue resume getting a partially-downloaded file.\n"),
N_("\
-N, --timestamping don't re-retrieve files unless newer than\n\
local.\n"),
N_("\
+ --no-use-server-timestamps don't set the local file's timestamp by\n\
+ the one on the server.\n"),
+ N_("\
-S, --server-response print server response.\n"),
N_("\
--spider don't download anything.\n"),
N_("\
--waitretry=SECONDS wait 1..SECONDS between retries of a retrieval.\n"),
N_("\
- --random-wait wait from 0...2*WAIT secs between retrievals.\n"),
+ --random-wait wait from 0.5*WAIT...1.5*WAIT secs between retrievals.\n"),
N_("\
--no-proxy explicitly turn off proxy.\n"),
N_("\
--password=PASS set both ftp and http password to PASS.\n"),
N_("\
--ask-password prompt for passwords.\n"),
+ N_("\
+ --no-iri turn off IRI support.\n"),
+ N_("\
+ --local-encoding=ENC use ENC as the local encoding for IRIs.\n"),
+ N_("\
+ --remote-encoding=ENC use ENC as the default remote encoding.\n"),
+ N_("\
+ --unlink remove file before clobber.\n"),
"\n",
N_("\
--default-page=NAME Change the default page name (normally\n\
this is `index.html'.).\n"),
N_("\
- -E, --html-extension save HTML documents with `.html' extension.\n"),
+ -E, --adjust-extension save HTML/CSS documents with proper extensions.\n"),
N_("\
--ignore-length ignore `Content-Length' header field.\n"),
N_("\
--content-disposition honor the Content-Disposition header when\n\
choosing local file names (EXPERIMENTAL).\n"),
N_("\
- --auth-no-challenge Send Basic HTTP authentication information\n\
+ --content-on-error output the received content on server errors.\n"),
+ N_("\
+ --auth-no-challenge send Basic HTTP authentication information\n\
without first waiting for the server's\n\
challenge.\n"),
"\n",
N_("\
FTP options:\n"),
+#ifdef __VMS
+ N_("\
+ --ftp-stmlf Use Stream_LF format for all binary FTP files.\n"),
+#endif /* def __VMS */
N_("\
--ftp-user=USER set ftp user to USER.\n"),
N_("\
--no-glob turn off FTP file name globbing.\n"),
N_("\
--no-passive-ftp disable the \"passive\" transfer mode.\n"),
+ N_("\
+ --preserve-permissions preserve remote file permissions.\n"),
N_("\
--retr-symlinks when recursing, get linked-to files (not dir).\n"),
+ "\n",
+
N_("\
- --preserve-permissions preserve remote file permissions.\n"),
+WARC options:\n"),
+ N_("\
+ --warc-file=FILENAME save request/response data to a .warc.gz file.\n"),
+ N_("\
+ --warc-header=STRING insert STRING into the warcinfo record.\n"),
+ N_("\
+ --warc-max-size=NUMBER set maximum size of WARC files to NUMBER.\n"),
+ 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_("\
+ --no-warc-keep-log do not store the log file in a WARC record.\n"),
+ N_("\
+ --warc-tempdir=DIRECTORY location for temporary files created by the\n\
+ WARC writer.\n"),
"\n",
N_("\
N_("\
-k, --convert-links make links in downloaded HTML or CSS point to\n\
local files.\n"),
+#ifdef __VMS
+ N_("\
+ -K, --backup-converted before converting file X, back up as X_orig.\n"),
+#else /* def __VMS */
N_("\
-K, --backup-converted before converting file X, back up as X.orig.\n"),
+#endif /* def __VMS [else] */
N_("\
-m, --mirror shortcut for -N -r -l inf --no-remove-listing.\n"),
N_("\
N_("\
-I, --include-directories=LIST list of allowed directories.\n"),
N_("\
+ --trust-server-names use the name specified by the redirection\n\
+ url last component.\n"),
+ N_("\
-X, --exclude-directories=LIST list of excluded directories.\n"),
N_("\
-np, --no-parent don't ascend to the parent directory.\n"),
"\n",
+ N_("\
+Output format:\n"),
+ N_("\
+ --bits Output bandwidth in bits.\n"),
+ "\n",
N_("Mail bug reports and suggestions to <bug-wget@gnu.org>.\n")
};
size_t i;
- printf (_("GNU Wget %s, a non-interactive network retriever.\n"),
- version_string);
- print_usage ();
+ if (printf (_("GNU Wget %s, a non-interactive network retriever.\n"),
+ version_string) < 0)
+ exit (3);
+ if (print_usage (0) < 0)
+ exit (3);
for (i = 0; i < countof (help); i++)
- fputs (_(help[i]), stdout);
+ if (fputs (_(help[i]), stdout) < 0)
+ exit (3);
exit (0);
}
prompt_for_password (void)
{
if (opt.user)
- printf (_("Password for user %s: "), quote (opt.user));
+ fprintf (stderr, _("Password for user %s: "), quote (opt.user));
else
- printf (_("Password: "));
+ fprintf (stderr, _("Password: "));
return getpass("");
}
to at most line_length. prefix is printed on the first line
and an appropriate number of spaces are added on subsequent
lines.*/
-static void
-format_and_print_line (char* prefix, char* line,
- int line_length)
+static int
+format_and_print_line (const char *prefix, const char *line,
+ int line_length)
{
- int leading_spaces;
int remaining_chars;
- char *token;
-
+ char *line_dup, *token;
+
assert (prefix != NULL);
assert (line != NULL);
+ line_dup = xstrdup (line);
+
if (line_length <= 0)
- line_length = max_chars_per_line;
+ line_length = MAX_CHARS_PER_LINE - TABULATION;
- leading_spaces = strlen (prefix);
- printf ("%s", prefix);
- remaining_chars = line_length - leading_spaces;
+ if (printf ("%s", prefix) < 0)
+ return -1;
+ remaining_chars = line_length;
/* We break on spaces. */
- token = strtok (line, " ");
- while (token != NULL)
+ token = strtok (line_dup, " ");
+ while (token != NULL)
{
/* If however a token is much larger than the maximum
line length, all bets are off and we simply print the
token on the next line. */
- if (remaining_chars <= strlen (token))
+ if (remaining_chars <= strlen (token))
{
- int j;
- printf ("\n");
- j = 0;
- for (j = 0; j < leading_spaces; j++)
- {
- printf (" ");
- }
- remaining_chars = line_length - leading_spaces;
+ if (printf ("\n%*c", TABULATION, ' ') < 0)
+ return -1;
+ remaining_chars = line_length - TABULATION;
}
- printf ("%s ", token);
- remaining_chars -= strlen (token) + 1; // account for " "
+ if (printf ("%s ", token) < 0)
+ return -1;
+ remaining_chars -= strlen (token) + 1; /* account for " " */
token = strtok (NULL, " ");
}
- printf ("\n");
+ if (printf ("\n") < 0)
+ return -1;
- /* FIXME: Responsibility for deallocation should be handled by
- whatever allocated it, wherever possible. These two lines result
- in unnecessary strdup calls in the print_version function. */
- xfree (prefix);
- xfree (line);
+ xfree (line_dup);
+ return 0;
}
static void
print_version (void)
{
- const char *options_title = "Options : ";
- const char *wgetrc_title = "Wgetrc : ";
- const char *locale_title = "Locale : ";
- const char *compile_title = "Compile : ";
- const char *link_title = "Link : ";
- const char *prefix_spaces = " ";
- const int prefix_space_length = strlen (prefix_spaces);
+ const char *wgetrc_title = _("Wgetrc: ");
+ const char *locale_title = _("Locale: ");
+ const char *compile_title = _("Compile: ");
+ const char *link_title = _("Link: ");
char *env_wgetrc, *user_wgetrc;
int i;
- printf ("GNU Wget %s\n", version_string);
- printf (options_title);
- /* compiled_features is a char*[]. We limit the characters per
- line to max_chars_per_line and prefix each line with a constant
- number of spaces for proper alignment. */
- for (i = 0; compiled_features[i] != NULL; )
+ if (printf (_("GNU Wget %s built on %s.\n\n"), version_string, OS_TYPE) < 0)
+ exit (3);
+
+ for (i = 0; compiled_features[i] != NULL; )
{
- int line_length = max_chars_per_line - prefix_space_length;
- while ((line_length > 0) && (compiled_features[i] != NULL))
+ int line_length = MAX_CHARS_PER_LINE;
+ while ((line_length > 0) && (compiled_features[i] != NULL))
{
- printf ("%s ", compiled_features[i]);
+ if (printf ("%s ", compiled_features[i]) < 0)
+ exit (3);
line_length -= strlen (compiled_features[i]) + 2;
i++;
}
- printf ("\n");
- if (compiled_features[i] != NULL)
- {
- printf (prefix_spaces);
- }
+ if (printf ("\n") < 0)
+ exit (3);
}
- /* Handle the case when $WGETRC is unset and $HOME/.wgetrc is
+ if (printf ("\n") < 0)
+ exit (3);
+
+ /* Handle the case when $WGETRC is unset and $HOME/.wgetrc is
absent. */
- printf (wgetrc_title);
+ if (printf ("%s\n", wgetrc_title) < 0)
+ exit (3);
+
env_wgetrc = wgetrc_env_file_name ();
- if (env_wgetrc && *env_wgetrc)
+ if (env_wgetrc && *env_wgetrc)
{
- printf ("%s (env)\n%s", env_wgetrc, prefix_spaces);
+ if (printf (_(" %s (env)\n"), env_wgetrc) < 0)
+ exit (3);
xfree (env_wgetrc);
}
user_wgetrc = wgetrc_user_file_name ();
- if (user_wgetrc)
+ if (user_wgetrc)
{
- printf ("%s (user)\n%s", user_wgetrc, prefix_spaces);
+ if (printf (_(" %s (user)\n"), user_wgetrc) < 0)
+ exit (3);
xfree (user_wgetrc);
}
#ifdef SYSTEM_WGETRC
- printf ("%s (system)\n", SYSTEM_WGETRC);
-#else
- putchar ('\n');
+ if (printf (_(" %s (system)\n"), SYSTEM_WGETRC) < 0)
+ exit (3);
#endif
- format_and_print_line (strdup (locale_title),
- strdup (LOCALEDIR),
- max_chars_per_line);
-
- format_and_print_line (strdup (compile_title),
- strdup (compilation_string),
- max_chars_per_line);
-
- format_and_print_line (strdup (link_title),
- strdup (link_string),
- max_chars_per_line);
- printf ("\n");
+#ifdef ENABLE_NLS
+ if (format_and_print_line (locale_title,
+ LOCALEDIR,
+ MAX_CHARS_PER_LINE) < 0)
+ exit (3);
+#endif /* def ENABLE_NLS */
+
+ if (compilation_string != NULL)
+ if (format_and_print_line (compile_title,
+ compilation_string,
+ MAX_CHARS_PER_LINE) < 0)
+ exit (3);
+
+ if (link_string != NULL)
+ if (format_and_print_line (link_title,
+ link_string,
+ MAX_CHARS_PER_LINE) < 0)
+ exit (3);
+
+ if (printf ("\n") < 0)
+ exit (3);
+
/* TRANSLATORS: When available, an actual copyright character
- (cirle-c) should be used in preference to "(C)". */
- fputs (_("\
-Copyright (C) 2008 Free Software Foundation, Inc.\n"), stdout);
- fputs (_("\
+ (circle-c) should be used in preference to "(C)". */
+ if (fputs (_("\
+Copyright (C) 2011 Free Software Foundation, Inc.\n"), stdout) < 0)
+ exit (3);
+ if (fputs (_("\
License GPLv3+: GNU GPL version 3 or later\n\
<http://www.gnu.org/licenses/gpl.html>.\n\
This is free software: you are free to change and redistribute it.\n\
-There is NO WARRANTY, to the extent permitted by law.\n"), stdout);
+There is NO WARRANTY, to the extent permitted by law.\n"), stdout) < 0)
+ exit (3);
/* TRANSLATORS: When available, please use the proper diacritics for
names such as this one. See en_US.po for reference. */
- fputs (_("\nOriginally written by Hrvoje Niksic <hniksic@xemacs.org>.\n"),
- stdout);
- fputs (_("Currently maintained by Micah Cowan <micah@cowan.name>.\n"),
- stdout);
+ if (fputs (_("\nOriginally written by Hrvoje Niksic <hniksic@xemacs.org>.\n"),
+ stdout) < 0)
+ exit (3);
+ if (fputs (_("Please send bug reports and questions to <bug-wget@gnu.org>.\n"),
+ stdout) < 0)
+ exit (3);
+
exit (0);
}
char *program_name; /* Needed by lib/error.c. */
+char *program_argstring; /* Needed by wget_warc.c. */
int
main (int argc, char **argv)
{
char **url, **t;
int i, ret, longindex;
- int nurl, status;
+ int nurl;
bool append_to_log = false;
+ total_downloaded_bytes = 0;
+
program_name = argv[0];
+ struct ptimer *timer = ptimer_new ();
+ double start_time = ptimer_measure (timer);
+
i18n_initialize ();
+ atexit (close_stdout);
+
/* Construct the name of the executable, without the directory part. */
+#ifdef __VMS
+ /* On VMS, lose the "dev:[dir]" prefix and the ".EXE;nnn" suffix. */
+ exec_name = vms_basename (argv[0]);
+#else /* def __VMS */
exec_name = strrchr (argv[0], PATH_SEPARATOR);
if (!exec_name)
exec_name = argv[0];
else
++exec_name;
+#endif /* def __VMS [else] */
#ifdef WINDOWS
/* Drop extension (typically .EXE) from executable filename. */
windows_main ((char **) &exec_name);
#endif
- /* Set option defaults; read the system wgetrc and ~/.wgetrc. */
- initialize ();
+ /* Construct the arguments string. */
+ int argstring_length = 1;
+ 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 = '\0';
+
+ /* Load the hard-coded defaults. */
+ defaults ();
init_switches ();
+
+ /* This separate getopt_long is needed to find the user config file
+ option ("--config") and parse it before the other user options. */
+ longindex = -1;
+ int retconf;
+ bool use_userconfig = false;
+
+ while ((retconf = getopt_long (argc, argv,
+ short_options, long_options, &longindex)) != -1)
+ {
+ int confval;
+ bool userrc_ret = true;
+ struct cmdline_option *config_opt;
+
+ /* There is no short option for "--config". */
+ if (longindex >= 0)
+ {
+ confval = long_options[longindex].val;
+ config_opt = &option_data[confval & ~BOOLEAN_NEG_MARKER];
+ if (strcmp (config_opt->long_name, "config") == 0)
+ {
+ userrc_ret &= run_wgetrc (optarg);
+ use_userconfig = true;
+ }
+ if (!userrc_ret)
+ {
+ fprintf (stderr, "Exiting due to error in %s\n", optarg);
+ exit (2);
+ }
+ else
+ break;
+ }
+ }
+
+ /* If the user did not specify a config, read the system wgetrc and ~/.wgetrc. */
+ if (use_userconfig == false)
+ initialize ();
+
+ opterr = 0;
+ optind = 0;
+
longindex = -1;
while ((ret = getopt_long (argc, argv,
short_options, long_options, &longindex)) != -1)
{
if (ret == '?')
{
- print_usage ();
- 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. */
short options for convenience and backward
compatibility. */
char *p;
- for (p = optarg; *p; p++)
+ for (p = optarg; p && *p; p++)
switch (*p)
{
case 'v':
setoptval ("noparent", "1", opt->long_name);
break;
default:
- printf (_("%s: illegal option -- `-n%c'\n"), exec_name, *p);
- print_usage ();
- printf ("\n");
- printf (_("Try `%s --help' for more options.\n"), exec_name);
+ fprintf (stderr, _("%s: illegal option -- `-n%c'\n"),
+ exec_name, *p);
+ print_usage (1);
+ fprintf (stderr, "\n");
+ fprintf (stderr, _("Try `%s --help' for more options.\n"),
+ exec_name);
exit (1);
}
break;
/* All user options have now been processed, so it's now safe to do
interoption dependency checks. */
+ if (opt.noclobber && opt.convert_links)
+ {
+ fprintf (stderr,
+ _("Both --no-clobber and --convert-links were specified,"
+ "only --convert-links will be used.\n"));
+ opt.noclobber = false;
+ }
+
if (opt.reclevel == 0)
opt.reclevel = INFINITE_RECURSION; /* see recur.h for commentary */
/* Sanity checks. */
if (opt.verbose && opt.quiet)
{
- printf (_("Can't be verbose and quiet at the same time.\n"));
- print_usage ();
+ fprintf (stderr, _("Can't be verbose and quiet at the same time.\n"));
+ print_usage (1);
exit (1);
}
if (opt.timestamping && opt.noclobber)
{
- printf (_("\
+ fprintf (stderr, _("\
Can't timestamp and not clobber old files at the same time.\n"));
- print_usage ();
+ print_usage (1);
exit (1);
}
#ifdef ENABLE_IPV6
if (opt.ipv4_only && opt.ipv6_only)
{
- printf (_("Cannot specify both --inet4-only and --inet6-only.\n"));
- print_usage ();
+ fprintf (stderr,
+ _("Cannot specify both --inet4-only and --inet6-only.\n"));
+ print_usage (1);
exit (1);
}
#endif
if (opt.output_document)
{
- if (opt.convert_links
+ if (opt.convert_links
&& (nurl > 1 || opt.page_requisites || opt.recursive))
{
fputs (_("\
Cannot specify both -k and -O if multiple URLs are given, or in combination\n\
-with -p or -r. See the manual for details.\n\n"), stdout);
- print_usage ();
+with -p or -r. See the manual for details.\n\n"), stderr);
+ print_usage (1);
exit (1);
}
if (opt.page_requisites
for details.\n\n"));
opt.timestamping = false;
}
- if (opt.noclobber && file_exists_p(opt.output_document))
- {
+ if (opt.noclobber && file_exists_p(opt.output_document))
+ {
/* Check if output file exists; if it does, exit. */
- logprintf (LOG_VERBOSE, _("File `%s' already there; not retrieving.\n"), opt.output_document);
+ logprintf (LOG_VERBOSE,
+ _("File `%s' already there; not retrieving.\n"),
+ opt.output_document);
exit(1);
- }
+ }
+ }
+
+ if (opt.warc_filename != 0)
+ {
+ if (opt.noclobber)
+ {
+ fprintf (stderr,
+ _("WARC output does not work with --no-clobber, "
+ "--no-clobber will be disabled.\n"));
+ opt.noclobber = false;
+ }
+ if (opt.timestamping)
+ {
+ fprintf (stderr,
+ _("WARC output does not work with timestamping, "
+ "timestamping will be disabled.\n"));
+ opt.timestamping = false;
+ }
+ if (opt.spider)
+ {
+ fprintf (stderr,
+ _("WARC output does not work with --spider.\n"));
+ exit (1);
+ }
+ if (opt.always_rest)
+ {
+ fprintf (stderr,
+ _("WARC output does not work with --continue, "
+ "--continue will be disabled.\n"));
+ opt.always_rest = false;
+ }
+ if (opt.warc_cdx_dedup_filename != 0 && !opt.warc_digests_enabled)
+ {
+ fprintf (stderr,
+ _("Digests are disabled; WARC deduplication will "
+ "not find duplicate records.\n"));
+ }
+ if (opt.warc_keep_log)
+ {
+ opt.progress_type = xstrdup ("dot");
+ }
}
if (opt.ask_passwd && opt.passwd)
{
- printf (_("Cannot specify both --ask-password and --password.\n"));
- print_usage ();
+ fprintf (stderr,
+ _("Cannot specify both --ask-password and --password.\n"));
+ print_usage (1);
exit (1);
}
if (!nurl && !opt.input_filename)
{
/* No URL specified. */
- printf (_("%s: missing URL\n"), exec_name);
- print_usage ();
- printf ("\n");
+ fprintf (stderr, _("%s: missing URL\n"), exec_name);
+ print_usage (1);
+ fprintf (stderr, "\n");
/* #### Something nicer should be printed here -- similar to the
pre-1.5 `--help' page. */
- printf (_("Try `%s --help' for more options.\n"), exec_name);
+ fprintf (stderr, _("Try `%s --help' for more options.\n"), exec_name);
exit (1);
}
+#ifdef ENABLE_IRI
+ if (opt.enable_iri)
+ {
+ if (opt.locale && !check_encoding_name (opt.locale))
+ opt.locale = NULL;
+
+ if (!opt.locale)
+ opt.locale = find_locale ();
+
+ if (opt.encoding_remote && !check_encoding_name (opt.encoding_remote))
+ opt.encoding_remote = NULL;
+ }
+#else
+ memset (&dummy_iri, 0, sizeof (dummy_iri));
+ if (opt.enable_iri || opt.locale || opt.encoding_remote)
+ {
+ /* sXXXav : be more specific... */
+ fprintf (stderr, _("This version does not have support for IRIs\n"));
+ exit(1);
+ }
+#endif
+
if (opt.ask_passwd)
{
opt.passwd = prompt_for_password ();
exit (1);
}
-#ifdef MSDOS
+#ifdef USE_WATT32
if (opt.wdebug)
dbug_init();
sock_init();
/* 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]);
/* Initialize logging. */
log_init (opt.lfilename, append_to_log);
- DEBUGP (("DEBUG output created by Wget %s on %s.\n\n", version_string,
- OS_TYPE));
+ /* Open WARC file. */
+ if (opt.warc_filename != 0)
+ warc_init ();
+
+ DEBUGP (("DEBUG output created by Wget %s on %s.\n\n",
+ version_string, OS_TYPE));
/* Open the output filename if necessary. */
+
+/* 2005-04-17 SMS.
+ Note that having the output_stream ("-O") file opened here for an FTP
+ URL rather than in getftp() (ftp.c) (and the http equivalent) rather
+ limits the ability in VMS to open the file differently for ASCII
+ versus binary FTP there. (Of course, doing it here allows a open
+ failure to be detected immediately, without first connecting to the
+ server.)
+*/
if (opt.output_document)
{
if (HYPHENP (opt.output_document))
{
#ifdef WINDOWS
- FILE *result;
- result = freopen (NULL, "wb", stdout);
- if (result == NULL)
- {
- logputs (LOG_NOTQUIET, _("\
-WARNING: Can't reopen standard output in binary mode;\n\
- downloaded file may contain inappropriate line endings.\n"));
- }
+ _setmode (_fileno (stdout), _O_BINARY);
#endif
output_stream = stdout;
}
else
{
struct_fstat st;
+
+#ifdef __VMS
+/* Common fopen() optional arguments:
+ sequential access only, access callback function.
+*/
+# define FOPEN_OPT_ARGS , "fop=sqo", "acc", acc_cb, &open_id
+ int open_id = 7;
+#else /* def __VMS */
+# define FOPEN_OPT_ARGS
+#endif /* def __VMS [else] */
+
output_stream = fopen (opt.output_document,
- opt.always_rest ? "ab" : "wb");
+ opt.always_rest ? "ab" : "wb"
+ FOPEN_OPT_ARGS);
if (output_stream == NULL)
{
perror (opt.output_document);
if (fstat (fileno (output_stream), &st) == 0 && S_ISREG (st.st_mode))
output_stream_regular = true;
}
+ if (!output_stream_regular && opt.convert_links)
+ {
+ fprintf (stderr, _("-k can be used together with -O only if \
+outputting to a regular file.\n"));
+ print_usage (1);
+ exit(1);
+ }
}
+#ifdef __VMS
+ /* Set global ODS5 flag according to the specified destination (if
+ any), otherwise according to the current default device.
+ */
+ if (output_stream == NULL)
+ set_ods5_dest( "SYS$DISK");
+ else if (output_stream != stdout)
+ set_ods5_dest( opt.output_document);
+#endif /* def __VMS */
+
#ifdef WINDOWS
ws_startup ();
#endif
signal (SIGWINCH, progress_handle_sigwinch);
#endif
- status = RETROK; /* initialize it, just-in-case */
/* Retrieve the URLs from argument list. */
for (t = url; *t; t++)
{
char *filename = NULL, *redirected_URL = NULL;
- int dt;
-
- if ((opt.recursive || opt.page_requisites)
- && (url_scheme (*t) != SCHEME_FTP || url_uses_proxy (*t)))
- {
- int old_follow_ftp = opt.follow_ftp;
+ int dt, url_err;
+ /* Need to do a new struct iri every time, because
+ * retrieve_url may modify it in some circumstances,
+ * currently. */
+ struct iri *iri = iri_new ();
+ struct url *url_parsed;
- /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
- if (url_scheme (*t) == SCHEME_FTP)
- opt.follow_ftp = 1;
-
- status = retrieve_tree (*t);
+ set_uri_encoding (iri, opt.locale, true);
+ url_parsed = url_parse (*t, &url_err, iri, true);
- opt.follow_ftp = old_follow_ftp;
+ if (!url_parsed)
+ {
+ char *error = url_error (*t, url_err);
+ logprintf (LOG_NOTQUIET, "%s: %s.\n",*t, error);
+ xfree (error);
+ inform_exit_status (URLERROR);
}
else
- status = retrieve_url (*t, &filename, &redirected_URL, NULL, &dt, opt.recursive);
-
- if (opt.delete_after && file_exists_p(filename))
{
- DEBUGP (("Removing file due to --delete-after in main():\n"));
- logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
- if (unlink (filename))
- logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
- }
+ if ((opt.recursive || opt.page_requisites)
+ && (url_scheme (*t) != SCHEME_FTP || url_uses_proxy (url_parsed)))
+ {
+ int old_follow_ftp = opt.follow_ftp;
- xfree_null (redirected_URL);
- xfree_null (filename);
+ /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
+ if (url_scheme (*t) == SCHEME_FTP)
+ opt.follow_ftp = 1;
+
+ retrieve_tree (url_parsed, NULL);
+
+ opt.follow_ftp = old_follow_ftp;
+ }
+ else
+ {
+ retrieve_url (url_parsed, *t, &filename, &redirected_URL, NULL,
+ &dt, opt.recursive, iri, true);
+ }
+
+ if (opt.delete_after && file_exists_p(filename))
+ {
+ DEBUGP (("Removing file due to --delete-after in main():\n"));
+ logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
+ if (unlink (filename))
+ logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
+ }
+ xfree_null (redirected_URL);
+ xfree_null (filename);
+ url_free (url_parsed);
+ }
+ iri_free (iri);
}
/* And then from the input file, if any. */
if (opt.input_filename)
{
int count;
+ int status;
status = retrieve_from_file (opt.input_filename, opt.force_html, &count);
+ inform_exit_status (status);
if (!count)
logprintf (LOG_NOTQUIET, _("No URLs found in %s.\n"),
opt.input_filename);
/* Print broken links. */
if (opt.recursive && opt.spider)
- {
- print_broken_links();
- }
-
+ print_broken_links ();
+
/* Print the downloaded sum. */
if ((opt.recursive || opt.page_requisites
|| nurl > 1
&&
total_downloaded_bytes != 0)
{
+ double end_time = ptimer_measure (timer);
+ ptimer_destroy (timer);
+
+ char *wall_time = xstrdup (secs_to_human_time (end_time - start_time));
+ char *download_time = xstrdup (secs_to_human_time (total_download_time));
logprintf (LOG_NOTQUIET,
- _("FINISHED --%s--\nDownloaded: %d files, %s in %s (%s)\n"),
- datetime_str (time (NULL)),
- opt.numurls,
- human_readable (total_downloaded_bytes),
- secs_to_human_time (total_download_time),
- retr_rate (total_downloaded_bytes, total_download_time));
+ _("FINISHED --%s--\nTotal wall clock time: %s\n"
+ "Downloaded: %d files, %s in %s (%s)\n"),
+ datetime_str (time (NULL)),
+ wall_time,
+ numurls,
+ human_readable (total_downloaded_bytes),
+ download_time,
+ retr_rate (total_downloaded_bytes, total_download_time));
+ xfree (wall_time);
+ xfree (download_time);
+
/* Print quota warning, if exceeded. */
if (opt.quota && total_downloaded_bytes > opt.quota)
logprintf (LOG_NOTQUIET,
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 ();
- if (status == RETROK)
- return 0;
- else
- return 1;
+ exit (get_exit_status ());
}
#endif /* TESTING */
\f