X-Git-Url: http://sjero.net/git/?p=wget;a=blobdiff_plain;f=src%2Fres.c;h=4b0ff82ba5b5a15ca4cae87e607ea2ac37f016e6;hp=69abd12dd5b60dc8d03193b3ddcd55f1003a7c63;hb=d763f8bf6d6e13ce006ffab616cc8a77e747a633;hpb=ccd62071dcbdfc0269813746b9f51ff9c23261db diff --git a/src/res.c b/src/res.c index 69abd12d..4b0ff82b 100644 --- a/src/res.c +++ b/src/res.c @@ -538,18 +538,34 @@ res_retrieve_file (const char *url, char **file, struct iri *iri) uerr_t err; char *robots_url = uri_merge (url, RES_SPECS_LOCATION); int saved_ts_val = opt.timestamping; - int saved_sp_val = opt.spider; + int saved_sp_val = opt.spider, url_err; + struct url * url_parsed; /* Copy server URI encoding for a possible IDNA transformation, no need to encode the full URI in UTF-8 because "robots.txt" is plain ASCII */ - set_uri_encoding (i, iri->uri_encoding); + set_uri_encoding (i, iri->uri_encoding, false); i->utf8_encode = false; logputs (LOG_VERBOSE, _("Loading robots.txt; please ignore errors.\n")); *file = NULL; opt.timestamping = false; opt.spider = false; - err = retrieve_url (robots_url, file, NULL, NULL, NULL, false, i); + + url_parsed = url_parse (robots_url, &url_err, iri, true); + if (!url_parsed) + { + char *error = url_error (robots_url, url_err); + logprintf (LOG_NOTQUIET, "%s: %s.\n", robots_url, error); + xfree (error); + err = URLERROR; + } + else + { + err = retrieve_url (url_parsed, robots_url, file, NULL, NULL, NULL, + false, i); + url_free(url_parsed); + } + opt.timestamping = saved_ts_val; opt.spider = saved_sp_val; xfree (robots_url);