X-Git-Url: http://sjero.net/git/?p=wget;a=blobdiff_plain;f=src%2Fhttp.c;h=d24db5fc38f3393ef6937c32e1ba03e3e6eba24b;hp=20f0619ad570835dc531349a9cc1a8d5f482cc0d;hb=9dadbf6fe9577a6a6b7e7bab4e4b782fc1a6f86c;hpb=9fa15c310f6a676812474c02e00178f4efd7d194 diff --git a/src/http.c b/src/http.c index 20f0619a..d24db5fc 100644 --- a/src/http.c +++ b/src/http.c @@ -1,6 +1,6 @@ /* HTTP support. Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, - 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc. + 2004, 2005, 2006, 2007, 2008, 2009 Free Software Foundation, Inc. This file is part of GNU Wget. @@ -66,14 +66,20 @@ as that of the covered work. */ #include "test.h" #endif +#ifdef __VMS +# include "vms.h" +#endif /* def __VMS */ + extern char *version_string; /* Forward decls. */ +struct http_stat; static char *create_authorization_line (const char *, const char *, const char *, const char *, const char *, bool *); static char *basic_authentication_encode (const char *, const char *); static bool known_authentication_scheme_p (const char *, const char *); +static void ensure_extension (struct http_stat *, const char *, int *); static void load_cookies (void); #ifndef MIN @@ -86,6 +92,7 @@ static struct cookie_jar *wget_cookie_jar; #define TEXTHTML_S "text/html" #define TEXTXHTML_S "application/xhtml+xml" +#define TEXTCSS_S "text/css" /* Some status code validation macros: */ #define H_20X(x) (((x) >= 200) && ((x) < 300)) @@ -139,6 +146,8 @@ struct request { int hcount, hcapacity; }; +extern int numurls; + /* Create a new, empty request. At least request_set_method must be called before the request can be used. */ @@ -403,13 +412,13 @@ maybe_send_basic_creds (const char *hostname, const char *user, else if (basic_authed_hosts && hash_table_contains(basic_authed_hosts, hostname)) { - DEBUGP(("Found `%s' in basic_authed_hosts.\n", hostname)); + DEBUGP(("Found %s in basic_authed_hosts.\n", quote (hostname))); do_challenge = true; } else { - DEBUGP(("Host `%s' has not issued a general basic challenge.\n", - hostname)); + DEBUGP(("Host %s has not issued a general basic challenge.\n", + quote (hostname))); } if (do_challenge) { @@ -430,7 +439,7 @@ register_basic_auth_host (const char *hostname) if (!hash_table_contains(basic_authed_hosts, hostname)) { hash_table_put (basic_authed_hosts, xstrdup(hostname), NULL); - DEBUGP(("Inserted `%s' into basic_authed_hosts\n", hostname)); + DEBUGP(("Inserted %s into basic_authed_hosts\n", quote (hostname))); } } @@ -810,7 +819,8 @@ print_response_line(const char *prefix, const char *b, const char *e) { char *copy; BOUNDED_TO_ALLOCA(b, e, copy); - logprintf (LOG_ALWAYS, "%s%s\n", prefix, escnonprint(copy)); + logprintf (LOG_ALWAYS, "%s%s\n", prefix, + quotearg_style (escape_quoting_style, copy)); } /* Print the server response, line by line, omitting the trailing CRLF @@ -1295,6 +1305,7 @@ struct http_stat char *remote_time; /* remote time-stamp string */ char *error; /* textual HTTP error */ int statcode; /* status code */ + char *message; /* status message */ wgint rd_size; /* amount of data read from socket */ double dltime; /* time it took to download the data */ const char *referer; /* value of the referer header. */ @@ -1321,6 +1332,7 @@ free_hstat (struct http_stat *hs) xfree_null (hs->rderrmsg); xfree_null (hs->local_file); xfree_null (hs->orig_file_name); + xfree_null (hs->message); /* Guard against being called twice. */ hs->newloc = NULL; @@ -1333,13 +1345,27 @@ free_hstat (struct http_stat *hs) && (c_isspace (line[sizeof (string_constant) - 1]) \ || !line[sizeof (string_constant) - 1])) +#ifdef __VMS #define SET_USER_AGENT(req) do { \ if (!opt.useragent) \ request_set_header (req, "User-Agent", \ - aprintf ("Wget/%s", version_string), rel_value); \ + aprintf ("Wget/%s (VMS %s %s)", \ + version_string, vms_arch(), vms_vers()), \ + rel_value); \ else if (*opt.useragent) \ request_set_header (req, "User-Agent", opt.useragent, rel_none); \ } while (0) +#else /* def __VMS */ +#define SET_USER_AGENT(req) do { \ + if (!opt.useragent) \ + request_set_header (req, "User-Agent", \ + aprintf ("Wget/%s (%s)", \ + version_string, OS_TYPE), \ + rel_value); \ + else if (*opt.useragent) \ + request_set_header (req, "User-Agent", opt.useragent, rel_none); \ +} while (0) +#endif /* def __VMS [else] */ /* The flags that allow clobbering the file (opening with "wb"). Defined here to avoid repetition later. #### This will require @@ -1358,7 +1384,8 @@ free_hstat (struct http_stat *hs) If PROXY is non-NULL, the connection will be made to the proxy server, and u->url will be requested. */ static uerr_t -gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) +gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy, + struct iri *iri) { struct request *req; @@ -1440,6 +1467,7 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) hs->newloc = NULL; hs->remote_time = NULL; hs->error = NULL; + hs->message = NULL; conn = u; @@ -1488,9 +1516,10 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) user = user ? user : (opt.http_user ? opt.http_user : opt.user); passwd = passwd ? passwd : (opt.http_passwd ? opt.http_passwd : opt.passwd); - if (user && passwd - && !u->user) /* We only do "site-wide" authentication with "global" - user/password values; URL user/password info overrides. */ + /* We only do "site-wide" authentication with "global" user/password + * values unless --auth-no-challange has been requested; URL user/password + * info overrides. */ + if (user && passwd && (!u->user || opt.auth_without_challenge)) { /* If this is a host for which we've already received a Basic * challenge, we'll go ahead and send Basic authentication creds. */ @@ -1544,8 +1573,8 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) post_data_size = file_size (opt.post_file_name); if (post_data_size == -1) { - logprintf (LOG_NOTQUIET, _("POST data file `%s' missing: %s\n"), - opt.post_file_name, strerror (errno)); + logprintf (LOG_NOTQUIET, _("POST data file %s missing: %s\n"), + quote (opt.post_file_name), strerror (errno)); post_data_size = 0; } } @@ -1629,7 +1658,8 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) sock = pconn.socket; using_ssl = pconn.ssl; logprintf (LOG_VERBOSE, _("Reusing existing connection to %s:%d.\n"), - escnonprint (pconn.host), pconn.port); + quotearg_style (escape_quoting_style, pconn.host), + pconn.port); DEBUGP (("Reusing fd %d.\n", sock)); if (pconn.authorized) /* If the connection is already authorized, the "Basic" @@ -1641,8 +1671,8 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) { request_free (req); logprintf(LOG_NOTQUIET, - _("%s: unable to resolve host address `%s'\n"), - exec_name, relevant->host); + _("%s: unable to resolve host address %s\n"), + exec_name, quote (relevant->host)); return HOSTERR; } } @@ -1711,13 +1741,14 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) resp = resp_new (head); statcode = resp_status (resp, &message); + hs->message = xstrdup (message); resp_free (resp); xfree (head); if (statcode != 200) { failed_tunnel: logprintf (LOG_NOTQUIET, _("Proxy tunneling failed: %s"), - message ? escnonprint (message) : "?"); + message ? quotearg_style (escape_quoting_style, message) : "?"); xfree_null (message); return CONSSLERR; } @@ -1731,7 +1762,7 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) if (conn->scheme == SCHEME_HTTPS) { - if (!ssl_connect (sock) || !ssl_check_certificate (sock, u->host)) + if (!ssl_connect_wget (sock) || !ssl_check_certificate (sock, u->host)) { fd_close (sock); return CONSSLERR; @@ -1793,15 +1824,111 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) /* Check for status line. */ message = NULL; statcode = resp_status (resp, &message); + hs->message = xstrdup (message); if (!opt.server_response) logprintf (LOG_VERBOSE, "%2d %s\n", statcode, - message ? escnonprint (message) : ""); + message ? quotearg_style (escape_quoting_style, message) : ""); else { logprintf (LOG_VERBOSE, "\n"); print_server_response (resp, " "); } + /* Check for keep-alive related responses. */ + if (!inhibit_keep_alive && contlen != -1) + { + if (resp_header_copy (resp, "Keep-Alive", NULL, 0)) + keep_alive = true; + else if (resp_header_copy (resp, "Connection", hdrval, sizeof (hdrval))) + { + if (0 == strcasecmp (hdrval, "Keep-Alive")) + keep_alive = true; + } + } + + if (keep_alive) + /* The server has promised that it will not close the connection + when we're done. This means that we can register it. */ + register_persistent (conn->host, conn->port, sock, using_ssl); + + if (statcode == HTTP_STATUS_UNAUTHORIZED) + { + /* Authorization is required. */ + if (keep_alive && !head_only && skip_short_body (sock, contlen)) + CLOSE_FINISH (sock); + else + CLOSE_INVALIDATE (sock); + pconn.authorized = false; + if (!auth_finished && (user && passwd)) + { + /* IIS sends multiple copies of WWW-Authenticate, one with + the value "negotiate", and other(s) with data. Loop over + all the occurrences and pick the one we recognize. */ + int wapos; + const char *wabeg, *waend; + char *www_authenticate = NULL; + for (wapos = 0; + (wapos = resp_header_locate (resp, "WWW-Authenticate", wapos, + &wabeg, &waend)) != -1; + ++wapos) + if (known_authentication_scheme_p (wabeg, waend)) + { + BOUNDED_TO_ALLOCA (wabeg, waend, www_authenticate); + break; + } + + if (!www_authenticate) + { + /* If the authentication header is missing or + unrecognized, there's no sense in retrying. */ + logputs (LOG_NOTQUIET, _("Unknown authentication scheme.\n")); + } + else if (!basic_auth_finished + || !BEGINS_WITH (www_authenticate, "Basic")) + { + char *pth; + pth = url_full_path (u); + request_set_header (req, "Authorization", + create_authorization_line (www_authenticate, + user, passwd, + request_method (req), + pth, + &auth_finished), + rel_value); + if (BEGINS_WITH (www_authenticate, "NTLM")) + ntlm_seen = true; + else if (!u->user && BEGINS_WITH (www_authenticate, "Basic")) + { + /* Need to register this host as using basic auth, + * so we automatically send creds next time. */ + register_basic_auth_host (u->host); + } + xfree (pth); + xfree_null (message); + resp_free (resp); + xfree (head); + goto retry_with_auth; + } + else + { + /* We already did Basic auth, and it failed. Gotta + * give up. */ + } + } + logputs (LOG_NOTQUIET, _("Authorization failed.\n")); + request_free (req); + xfree_null (message); + resp_free (resp); + xfree (head); + return AUTHFAILED; + } + else /* statcode != HTTP_STATUS_UNAUTHORIZED */ + { + /* Kludge: if NTLM is used, mark the TCP connection as authorized. */ + if (ntlm_seen) + pconn.authorized = true; + } + /* Determine the local filename if needed. Notice that if -O is used * hstat.local_file is set by http_loop to the argument of -O. */ if (!hs->local_file) @@ -1817,7 +1944,7 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) hs->local_file = url_file_name (u); } } - + /* TODO: perform this check only once. */ if (!hs->existence_checked && file_exists_p (hs->local_file)) { @@ -1827,7 +1954,7 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy) retrieve the file. But if the output_document was given, then this test was already done and the file didn't exist. Hence the !opt.output_document */ logprintf (LOG_VERBOSE, _("\ -File `%s' already there; not retrieving.\n\n"), hs->local_file); +File %s already there; not retrieving.\n\n"), quote (hs->local_file)); /* If the file is there, we suppose it's retrieved OK. */ *dt |= RETROKF; @@ -1836,6 +1963,8 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); if (has_html_suffix_p (hs->local_file)) *dt |= TEXTHTML; + xfree (head); + xfree_null (message); return RETRUNNEEDED; } else if (!ALLOW_CLOBBER) @@ -1853,7 +1982,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); if (opt.timestamping && !hs->timestamp_checked) { size_t filename_len = strlen (hs->local_file); - char *filename_plus_orig_suffix = alloca (filename_len + sizeof (".orig")); + char *filename_plus_orig_suffix = alloca (filename_len + sizeof (ORIG_SFX)); bool local_dot_orig_file_exists = false; char *local_filename = NULL; struct_stat st; @@ -1878,7 +2007,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); --hniksic */ memcpy (filename_plus_orig_suffix, hs->local_file, filename_len); memcpy (filename_plus_orig_suffix + filename_len, - ".orig", sizeof (".orig")); + ORIG_SFX, sizeof (ORIG_SFX)); /* Try to stat() the .orig file. */ if (stat (filename_plus_orig_suffix, &st) == 0) @@ -1886,7 +2015,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); local_dot_orig_file_exists = true; local_filename = filename_plus_orig_suffix; } - } + } if (!local_dot_orig_file_exists) /* Couldn't stat() .orig, so try to stat() . */ @@ -1934,93 +2063,6 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); contlen = parsed; } - /* Check for keep-alive related responses. */ - if (!inhibit_keep_alive && contlen != -1) - { - if (resp_header_copy (resp, "Keep-Alive", NULL, 0)) - keep_alive = true; - else if (resp_header_copy (resp, "Connection", hdrval, sizeof (hdrval))) - { - if (0 == strcasecmp (hdrval, "Keep-Alive")) - keep_alive = true; - } - } - if (keep_alive) - /* The server has promised that it will not close the connection - when we're done. This means that we can register it. */ - register_persistent (conn->host, conn->port, sock, using_ssl); - - if (statcode == HTTP_STATUS_UNAUTHORIZED) - { - /* Authorization is required. */ - if (keep_alive && !head_only && skip_short_body (sock, contlen)) - CLOSE_FINISH (sock); - else - CLOSE_INVALIDATE (sock); - pconn.authorized = false; - if (!auth_finished && (user && passwd)) - { - /* IIS sends multiple copies of WWW-Authenticate, one with - the value "negotiate", and other(s) with data. Loop over - all the occurrences and pick the one we recognize. */ - int wapos; - const char *wabeg, *waend; - char *www_authenticate = NULL; - for (wapos = 0; - (wapos = resp_header_locate (resp, "WWW-Authenticate", wapos, - &wabeg, &waend)) != -1; - ++wapos) - if (known_authentication_scheme_p (wabeg, waend)) - { - BOUNDED_TO_ALLOCA (wabeg, waend, www_authenticate); - break; - } - - if (!www_authenticate) - { - /* If the authentication header is missing or - unrecognized, there's no sense in retrying. */ - logputs (LOG_NOTQUIET, _("Unknown authentication scheme.\n")); - } - else if (!basic_auth_finished - || !BEGINS_WITH (www_authenticate, "Basic")) - { - char *pth; - pth = url_full_path (u); - request_set_header (req, "Authorization", - create_authorization_line (www_authenticate, - user, passwd, - request_method (req), - pth, - &auth_finished), - rel_value); - if (BEGINS_WITH (www_authenticate, "NTLM")) - ntlm_seen = true; - else if (!u->user && BEGINS_WITH (www_authenticate, "Basic")) - { - /* Need to register this host as using basic auth, - * so we automatically send creds next time. */ - register_basic_auth_host (u->host); - } - xfree (pth); - goto retry_with_auth; - } - else - { - /* We already did Basic auth, and it failed. Gotta - * give up. */ - } - } - logputs (LOG_NOTQUIET, _("Authorization failed.\n")); - request_free (req); - return AUTHFAILED; - } - else /* statcode != HTTP_STATUS_UNAUTHORIZED */ - { - /* Kludge: if NTLM is used, mark the TCP connection as authorized. */ - if (ntlm_seen) - pconn.authorized = true; - } request_free (req); hs->statcode = statcode; @@ -2038,9 +2080,20 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); char *tmp = strchr (type, ';'); if (tmp) { + /* sXXXav: only needed if IRI support is enabled */ + char *tmp2 = tmp + 1; + while (tmp > type && c_isspace (tmp[-1])) --tmp; *tmp = '\0'; + + /* Try to get remote encoding if needed */ + if (opt.enable_iri && !opt.encoding_remote) + { + tmp = parse_charset (tmp2); + if (tmp) + set_content_encoding (iri, tmp); + } } } hs->newloc = resp_header_strdup (resp, "Location"); @@ -2101,6 +2154,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); else CLOSE_INVALIDATE (sock); xfree_null (type); + xfree (head); return NEWLOCATION; } } @@ -2115,42 +2169,37 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); else *dt &= ~TEXTHTML; - if (opt.html_extension && (*dt & TEXTHTML)) - /* -E / --html-extension / html_extension = on was specified, and this is a - text/html file. If some case-insensitive variation on ".htm[l]" isn't - already the file's suffix, tack on ".html". */ - { - char *last_period_in_local_filename = strrchr (hs->local_file, '.'); + if (type && + 0 == strncasecmp (type, TEXTCSS_S, strlen (TEXTCSS_S))) + *dt |= TEXTCSS; + else + *dt &= ~TEXTCSS; - if (last_period_in_local_filename == NULL - || !(0 == strcasecmp (last_period_in_local_filename, ".htm") - || 0 == strcasecmp (last_period_in_local_filename, ".html"))) + if (opt.adjust_extension) + { + if (*dt & TEXTHTML) + /* -E / --adjust-extension / adjust_extension = on was specified, + and this is a text/html file. If some case-insensitive + variation on ".htm[l]" isn't already the file's suffix, + tack on ".html". */ { - int local_filename_len = strlen (hs->local_file); - /* Resize the local file, allowing for ".html" preceded by - optional ".NUMBER". */ - hs->local_file = xrealloc (hs->local_file, - local_filename_len + 24 + sizeof (".html")); - strcpy(hs->local_file + local_filename_len, ".html"); - /* If clobbering is not allowed and the file, as named, - exists, tack on ".NUMBER.html" instead. */ - if (!ALLOW_CLOBBER && file_exists_p (hs->local_file)) - { - int ext_num = 1; - do - sprintf (hs->local_file + local_filename_len, - ".%d.html", ext_num++); - while (file_exists_p (hs->local_file)); - } - *dt |= ADDED_HTML_EXTENSION; + ensure_extension (hs, ".html", dt); + } + else if (*dt & TEXTCSS) + { + ensure_extension (hs, ".css", dt); } } - if (statcode == HTTP_STATUS_RANGE_NOT_SATISFIABLE) + if (statcode == HTTP_STATUS_RANGE_NOT_SATISFIABLE + || (hs->restval > 0 && statcode == HTTP_STATUS_OK + && contrange == 0 && hs->restval >= contlen) + ) { /* If `-c' is in use and the file has been fully downloaded (or the remote file has shrunk), Wget effectively requests bytes - after the end of file and the server response with 416. */ + after the end of file and the server response with 416 + (or 200 with a <= Content-Length. */ logputs (LOG_VERBOSE, _("\ \n The file is already fully retrieved; nothing to do.\n\n")); /* In case the caller inspects. */ @@ -2161,6 +2210,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); xfree_null (type); CLOSE_INVALIDATE (sock); /* would be CLOSE_FINISH, but there might be more bytes in the body. */ + xfree (head); return RETRUNNEEDED; } if ((contrange != 0 && contrange != hs->restval) @@ -2170,6 +2220,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); Bail out. */ xfree_null (type); CLOSE_INVALIDATE (sock); + xfree (head); return RANGEERR; } if (contlen == -1) @@ -2206,7 +2257,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); logputs (LOG_VERBOSE, opt.ignore_length ? _("ignored") : _("unspecified")); if (type) - logprintf (LOG_VERBOSE, " [%s]\n", escnonprint (type)); + logprintf (LOG_VERBOSE, " [%s]\n", quotearg_style (escape_quoting_style, type)); else logputs (LOG_VERBOSE, "\n"); } @@ -2233,9 +2284,20 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); CLOSE_FINISH (sock); else CLOSE_INVALIDATE (sock); + xfree (head); return RETRFINISHED; } +/* 2005-06-17 SMS. + For VMS, define common fopen() optional arguments. +*/ +#ifdef __VMS +# define FOPEN_OPT_ARGS "fop=sqo", "acc", acc_cb, &open_id +# define FOPEN_BIN_FLAG 3 +#else /* def __VMS */ +# define FOPEN_BIN_FLAG 1 +#endif /* def __VMS [else] */ + /* Open the local file. */ if (!output_stream) { @@ -2243,9 +2305,27 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); if (opt.backups) rotate_backups (hs->local_file); if (hs->restval) - fp = fopen (hs->local_file, "ab"); + { +#ifdef __VMS + int open_id; + + open_id = 21; + fp = fopen (hs->local_file, "ab", FOPEN_OPT_ARGS); +#else /* def __VMS */ + fp = fopen (hs->local_file, "ab"); +#endif /* def __VMS [else] */ + } else if (ALLOW_CLOBBER) - fp = fopen (hs->local_file, "wb"); + { +#ifdef __VMS + int open_id; + + open_id = 22; + fp = fopen (hs->local_file, "wb", FOPEN_OPT_ARGS); +#else /* def __VMS */ + fp = fopen (hs->local_file, "wb"); +#endif /* def __VMS [else] */ + } else { fp = fopen_excl (hs->local_file, true); @@ -2259,6 +2339,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); _("%s has sprung into existence.\n"), hs->local_file); CLOSE_INVALIDATE (sock); + xfree (head); return FOPEN_EXCL_ERR; } } @@ -2266,6 +2347,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); { logprintf (LOG_NOTQUIET, "%s: %s\n", hs->local_file, strerror (errno)); CLOSE_INVALIDATE (sock); + xfree (head); return FOPENERR; } } @@ -2275,8 +2357,8 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); /* Print fetch message, if opt.verbose. */ if (opt.verbose) { - logprintf (LOG_NOTQUIET, _("Saving to: `%s'\n"), - HYPHENP (hs->local_file) ? "STDOUT" : hs->local_file); + logprintf (LOG_NOTQUIET, _("Saving to: %s\n"), + HYPHENP (hs->local_file) ? quote ("STDOUT") : quote (hs->local_file)); } /* This confuses the timestamping code that checks for file size. @@ -2324,7 +2406,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file); retried, and retried, and retried, and... */ uerr_t http_loop (struct url *u, char **newloc, char **local_file, const char *referer, - int *dt, struct url *proxy) + int *dt, struct url *proxy, struct iri *iri) { int count; bool got_head = false; /* used for time-stamping and filename detection */ @@ -2335,16 +2417,17 @@ http_loop (struct url *u, char **newloc, char **local_file, const char *referer, uerr_t err, ret = TRYLIMEXC; time_t tmr = -1; /* remote time-stamp */ struct http_stat hstat; /* HTTP status */ - struct_stat st; + struct_stat st; bool send_head_first = true; + char *file_name; /* Assert that no value for *LOCAL_FILE was passed. */ assert (local_file == NULL || *local_file == NULL); - + /* Set LOCAL_FILE parameter. */ if (local_file && opt.output_document) *local_file = HYPHENP (opt.output_document) ? NULL : xstrdup (opt.output_document); - + /* Reset NEWLOC parameter. */ *newloc = NULL; @@ -2381,8 +2464,8 @@ http_loop (struct url *u, char **newloc, char **local_file, const char *referer, retrieve the file. But if the output_document was given, then this test was already done and the file didn't exist. Hence the !opt.output_document */ logprintf (LOG_VERBOSE, _("\ -File `%s' already there; not retrieving.\n\n"), - hstat.local_file); +File %s already there; not retrieving.\n\n"), + quote (hstat.local_file)); /* If the file is there, we suppose it's retrieved OK. */ *dt |= RETROKF; @@ -2397,10 +2480,10 @@ File `%s' already there; not retrieving.\n\n"), /* Reset the counter. */ count = 0; - + /* Reset the document type. */ *dt = 0; - + /* Skip preliminary HEAD request if we're not in spider mode AND * if -O was given or HTTP Content-Disposition support is disabled. */ if (!opt.spider @@ -2409,10 +2492,12 @@ File `%s' already there; not retrieving.\n\n"), /* Send preliminary HEAD request if -N is given and we have an existing * destination file. */ - if (opt.timestamping + file_name = url_file_name (u); + if (opt.timestamping && !opt.content_disposition - && file_exists_p (url_file_name (u))) + && file_exists_p (file_name)) send_head_first = true; + xfree (file_name); /* THE loop */ do @@ -2420,10 +2505,10 @@ File `%s' already there; not retrieving.\n\n"), /* Increment the pass counter. */ ++count; sleep_between_retrievals (count); - + /* Get the current time string. */ tms = datetime_str (time (NULL)); - + if (opt.spider && !got_head) logprintf (LOG_VERBOSE, _("\ Spider mode enabled. Check if remote file exists.\n")); @@ -2432,20 +2517,20 @@ Spider mode enabled. Check if remote file exists.\n")); if (opt.verbose) { char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD); - - if (count > 1) + + if (count > 1) { char tmp[256]; sprintf (tmp, _("(try:%2d)"), count); logprintf (LOG_NOTQUIET, "--%s-- %s %s\n", tms, tmp, hurl); } - else + else { logprintf (LOG_NOTQUIET, "--%s-- %s\n", tms, hurl); } - + #ifdef WINDOWS ws_changetitle (hurl); #endif @@ -2455,7 +2540,7 @@ Spider mode enabled. Check if remote file exists.\n")); /* Default document type is empty. However, if spider mode is on or time-stamping is employed, HEAD_ONLY commands is encoded within *dt. */ - if (send_head_first && !got_head) + if (send_head_first && !got_head) *dt |= HEAD_ONLY; else *dt &= ~HEAD_ONLY; @@ -2488,11 +2573,11 @@ Spider mode enabled. Check if remote file exists.\n")); *dt &= ~SEND_NOCACHE; /* Try fetching the document, or at least its head. */ - err = gethttp (u, &hstat, dt, proxy); + err = gethttp (u, &hstat, dt, proxy, iri); /* Time? */ tms = datetime_str (time (NULL)); - + /* Get the new location (with or without the redirection). */ if (hstat.newloc) *newloc = xstrdup (hstat.newloc); @@ -2510,8 +2595,8 @@ Spider mode enabled. Check if remote file exists.\n")); case FWRITEERR: case FOPENERR: /* Another fatal error. */ logputs (LOG_VERBOSE, "\n"); - logprintf (LOG_NOTQUIET, _("Cannot write to `%s' (%s).\n"), - hstat.local_file, strerror (errno)); + logprintf (LOG_NOTQUIET, _("Cannot write to %s (%s).\n"), + quote (hstat.local_file), strerror (errno)); case HOSTERR: case CONIMPOSSIBLE: case PROXERR: case AUTHFAILED: case SSLINITFAILED: case CONTNOTSUPPORTED: /* Fatal errors just return from the function. */ @@ -2531,7 +2616,7 @@ Spider mode enabled. Check if remote file exists.\n")); hstat.statcode); ret = WRONGCODE; } - else + else { ret = NEWLOCATION; } @@ -2547,7 +2632,7 @@ Spider mode enabled. Check if remote file exists.\n")); /* All possibilities should have been exhausted. */ abort (); } - + if (!(*dt & RETROKF)) { char *hurl = NULL; @@ -2566,11 +2651,13 @@ Spider mode enabled. Check if remote file exists.\n")); continue; } /* Maybe we should always keep track of broken links, not just in - * spider mode. */ - else if (opt.spider) + * spider mode. + * Don't log error if it was UTF-8 encoded because we will try + * once unencoded. */ + else if (opt.spider && !iri->utf8_encode) { /* #### Again: ugly ugly ugly! */ - if (!hurl) + if (!hurl) hurl = url_string (u, URL_AUTH_HIDE_PASSWD); nonexisting_url (hurl); logprintf (LOG_NOTQUIET, _("\ @@ -2579,7 +2666,8 @@ Remote file does not exist -- broken link!!!\n")); else { logprintf (LOG_NOTQUIET, _("%s ERROR %d: %s.\n"), - tms, hstat.statcode, escnonprint (hstat.error)); + tms, hstat.statcode, + quotearg_style (escape_quoting_style, hstat.error)); } logputs (LOG_VERBOSE, "\n"); ret = WRONGCODE; @@ -2633,8 +2721,8 @@ Last-modified header invalid -- time-stamp ignored.\n")); || hstat.orig_file_size == hstat.contlen) { logprintf (LOG_VERBOSE, _("\ -Server file no newer than local file `%s' -- not retrieving.\n\n"), - hstat.orig_file_name); +Server file no newer than local file %s -- not retrieving.\n\n"), + quote (hstat.orig_file_name)); ret = RETROK; goto exit; } @@ -2659,19 +2747,20 @@ The sizes do not match (local %s) -- retrieving.\n"), if (opt.spider) { + bool finished = true; if (opt.recursive) { if (*dt & TEXTHTML) { logputs (LOG_VERBOSE, _("\ Remote file exists and could contain links to other resources -- retrieving.\n\n")); + finished = false; } else { logprintf (LOG_VERBOSE, _("\ Remote file exists but does not contain any link -- not retrieving.\n\n")); ret = RETROK; /* RETRUNNEEDED is not for caller. */ - goto exit; } } else @@ -2688,6 +2777,14 @@ but recursion is disabled -- not retrieving.\n\n")); Remote file exists.\n\n")); } ret = RETROK; /* RETRUNNEEDED is not for caller. */ + } + + if (finished) + { + logprintf (LOG_NONVERBOSE, + _("%s URL: %s %2d %s\n"), + tms, u->url, hstat.statcode, + hstat.message ? quotearg_style (escape_quoting_style, hstat.message) : ""); goto exit; } } @@ -2703,16 +2800,8 @@ Remote file exists.\n\n")); && ((hstat.len == hstat.contlen) || ((hstat.res == 0) && (hstat.contlen == -1)))) { - /* #### This code repeats in http.c and ftp.c. Move it to a - function! */ const char *fl = NULL; - if (opt.output_document) - { - if (output_stream_regular) - fl = opt.output_document; - } - else - fl = hstat.local_file; + set_local_file (&fl, hstat.local_file); if (fl) { time_t newtmr = -1; @@ -2721,7 +2810,7 @@ Remote file exists.\n\n")); && hstat.remote_time && hstat.remote_time[0]) { newtmr = http_atotm (hstat.remote_time); - if (newtmr != -1) + if (newtmr != (time_t)-1) tmr = newtmr; } touch (fl, tmr); @@ -2736,9 +2825,14 @@ Remote file exists.\n\n")); { if (*dt & RETROKF) { + bool write_to_stdout = (opt.output_document && HYPHENP (opt.output_document)); + logprintf (LOG_VERBOSE, - _("%s (%s) - `%s' saved [%s/%s]\n\n"), - tms, tmrate, hstat.local_file, + write_to_stdout + ? _("%s (%s) - written to stdout %s[%s/%s]\n\n") + : _("%s (%s) - %s saved [%s/%s]\n\n"), + tms, tmrate, + write_to_stdout ? "" : quote (hstat.local_file), number_to_static_string (hstat.len), number_to_static_string (hstat.contlen)); logprintf (LOG_NONVERBOSE, @@ -2748,7 +2842,7 @@ Remote file exists.\n\n")); number_to_static_string (hstat.contlen), hstat.local_file, count); } - ++opt.numurls; + ++numurls; total_downloaded_bytes += hstat.len; /* Remember that we downloaded the file for later ".orig" code. */ @@ -2767,16 +2861,21 @@ Remote file exists.\n\n")); { if (*dt & RETROKF) { + bool write_to_stdout = (opt.output_document && HYPHENP (opt.output_document)); + logprintf (LOG_VERBOSE, - _("%s (%s) - `%s' saved [%s]\n\n"), - tms, tmrate, hstat.local_file, + write_to_stdout + ? _("%s (%s) - written to stdout %s[%s]\n\n") + : _("%s (%s) - %s saved [%s]\n\n"), + tms, tmrate, + write_to_stdout ? "" : quote (hstat.local_file), number_to_static_string (hstat.len)); logprintf (LOG_NONVERBOSE, "%s URL:%s [%s] -> \"%s\" [%d]\n", tms, u->url, number_to_static_string (hstat.len), hstat.local_file, count); } - ++opt.numurls; + ++numurls; total_downloaded_bytes += hstat.len; /* Remember that we downloaded the file for later ".orig" code. */ @@ -2797,10 +2896,18 @@ Remote file exists.\n\n")); printwhat (count, opt.ntry); continue; } - else + else if (hstat.len != hstat.restval) /* Getting here would mean reading more data than requested with content-length, which we never do. */ abort (); + else + { + /* Getting here probably means that the content-length was + * _less_ than the original, local size. We should probably + * truncate or re-read, or something. FIXME */ + ret = RETROK; + goto exit; + } } else /* from now on hstat.res can only be -1 */ { @@ -2905,13 +3012,24 @@ http_atotm (const char *time_string) Netscape cookie specification.) */ }; const char *oldlocale; - int i; + char savedlocale[256]; + size_t i; time_t ret = (time_t) -1; /* Solaris strptime fails to recognize English month names in non-English locales, which we work around by temporarily setting locale to C before invoking strptime. */ oldlocale = setlocale (LC_TIME, NULL); + if (oldlocale) + { + size_t l = strlen (oldlocale); + if (l >= sizeof savedlocale) + savedlocale[0] = '\0'; + else + memcpy (savedlocale, oldlocale, l); + } + else savedlocale[0] = '\0'; + setlocale (LC_TIME, "C"); for (i = 0; i < countof (time_formats); i++) @@ -2931,7 +3049,8 @@ http_atotm (const char *time_string) } /* Restore the previous locale. */ - setlocale (LC_TIME, oldlocale); + if (savedlocale[0]) + setlocale (LC_TIME, savedlocale); return ret; } @@ -3016,10 +3135,12 @@ digest_authentication_encode (const char *au, const char *user, au += 6; /* skip over `Digest' */ while (extract_param (&au, &name, &value, ',')) { - int i; + size_t i; + size_t namelen = name.e - name.b; for (i = 0; i < countof (options); i++) - if (name.e - name.b == strlen (options[i].name) - && 0 == strncmp (name.b, options[i].name, name.e - name.b)) + if (namelen == strlen (options[i].name) + && 0 == strncmp (name.b, options[i].name, + namelen)) { *options[i].variable = strdupdelim (value.b, value.e); break; @@ -3099,9 +3220,10 @@ username=\"%s\", realm=\"%s\", nonce=\"%s\", uri=\"%s\", response=\"%s\"", first argument and are followed by whitespace or terminating \0. The comparison is case-insensitive. */ #define STARTS(literal, b, e) \ - ((e) - (b) >= STRSIZE (literal) \ + ((e > b) \ + && ((size_t) ((e) - (b))) >= STRSIZE (literal) \ && 0 == strncasecmp (b, literal, STRSIZE (literal)) \ - && ((e) - (b) == STRSIZE (literal) \ + && ((size_t) ((e) - (b)) == STRSIZE (literal) \ || c_isspace (b[STRSIZE (literal)]))) static bool @@ -3184,6 +3306,42 @@ http_cleanup (void) cookie_jar_delete (wget_cookie_jar); } +void +ensure_extension (struct http_stat *hs, const char *ext, int *dt) +{ + char *last_period_in_local_filename = strrchr (hs->local_file, '.'); + char shortext[8]; + int len = strlen (ext); + if (len == 5) + { + strncpy (shortext, ext, len - 1); + shortext[len - 2] = '\0'; + } + + if (last_period_in_local_filename == NULL + || !(0 == strcasecmp (last_period_in_local_filename, shortext) + || 0 == strcasecmp (last_period_in_local_filename, ext))) + { + int local_filename_len = strlen (hs->local_file); + /* Resize the local file, allowing for ".html" preceded by + optional ".NUMBER". */ + hs->local_file = xrealloc (hs->local_file, + local_filename_len + 24 + len); + strcpy (hs->local_file + local_filename_len, ext); + /* If clobbering is not allowed and the file, as named, + exists, tack on ".NUMBER.html" instead. */ + if (!ALLOW_CLOBBER && file_exists_p (hs->local_file)) + { + int ext_num = 1; + do + sprintf (hs->local_file + local_filename_len, + ".%d%s", ext_num++, ext); + while (file_exists_p (hs->local_file)); + } + *dt |= ADDED_HTML_EXTENSION; + } +} + #ifdef TESTING