X-Git-Url: http://sjero.net/git/?a=blobdiff_plain;f=src%2Furl.c;h=ed4a0f4689c4275f6ef40735aea700a9ed918a55;hb=7415b33b6c35134f72aa6b2a76ef99b9598234af;hp=c7a3a721c29ea53e3559c9834d7d6bfdaaf061bf;hpb=ccd62071dcbdfc0269813746b9f51ff9c23261db;p=wget diff --git a/src/url.c b/src/url.c index c7a3a721..ed4a0f46 100644 --- a/src/url.c +++ b/src/url.c @@ -1,6 +1,6 @@ /* URL handling. - 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 Free Software Foundation, Inc. This file is part of GNU Wget. @@ -42,7 +42,10 @@ as that of the covered work. */ #include "utils.h" #include "url.h" #include "host.h" /* for is_valid_ipv6_address */ -#include "iri.h" + +#ifdef __VMS +#include "vms.h" +#endif /* def __VMS */ #ifdef TESTING #include "test.h" @@ -253,6 +256,15 @@ url_escape (const char *s) return url_escape_1 (s, urlchr_unsafe, false); } +/* URL-escape the unsafe and reserved characters (see urlchr_table) in + a given string, returning a freshly allocated string. */ + +char * +url_escape_unsafe_and_reserved (const char *s) +{ + return url_escape_1 (s, urlchr_unsafe|urlchr_reserved, false); +} + /* URL-escape the unsafe characters (see urlchr_table) in a given string. If no characters are unsafe, S is returned. */ @@ -620,18 +632,20 @@ static const char *parse_errors[] = { #define PE_NO_ERROR 0 N_("No error"), #define PE_UNSUPPORTED_SCHEME 1 - N_("Unsupported scheme"), -#define PE_INVALID_HOST_NAME 2 + N_("Unsupported scheme %s"), /* support for format token only here */ +#define PE_MISSING_SCHEME 2 + N_("Scheme missing"), +#define PE_INVALID_HOST_NAME 3 N_("Invalid host name"), -#define PE_BAD_PORT_NUMBER 3 +#define PE_BAD_PORT_NUMBER 4 N_("Bad port number"), -#define PE_INVALID_USER_NAME 4 +#define PE_INVALID_USER_NAME 5 N_("Invalid user name"), -#define PE_UNTERMINATED_IPV6_ADDRESS 5 +#define PE_UNTERMINATED_IPV6_ADDRESS 6 N_("Unterminated IPv6 numeric address"), -#define PE_IPV6_NOT_SUPPORTED 6 +#define PE_IPV6_NOT_SUPPORTED 7 N_("IPv6 addresses not supported"), -#define PE_INVALID_IPV6_ADDRESS 7 +#define PE_INVALID_IPV6_ADDRESS 8 N_("Invalid IPv6 numeric address") }; @@ -641,7 +655,7 @@ static const char *parse_errors[] = { error, and if ERROR is not NULL, also set *ERROR to the appropriate error code. */ struct url * -url_parse (const char *url, int *error, struct iri *iri) +url_parse (const char *url, int *error, struct iri *iri, bool percent_encode) { struct url *u; const char *p; @@ -660,26 +674,36 @@ url_parse (const char *url, int *error, struct iri *iri) int port; char *user = NULL, *passwd = NULL; - char *url_encoded = NULL, *new_url = NULL; + const char *url_encoded = NULL; + char *new_url = NULL; int error_code; scheme = url_scheme (url); if (scheme == SCHEME_INVALID) { - error_code = PE_UNSUPPORTED_SCHEME; + if (url_has_scheme (url)) + error_code = PE_UNSUPPORTED_SCHEME; + else + error_code = PE_MISSING_SCHEME; goto error; } if (iri && iri->utf8_encode) { - url_unescape ((char *) url); - iri->utf8_encode = remote_to_utf8 (iri, url, (const char **) &new_url); + iri->utf8_encode = remote_to_utf8 (iri, iri->orig_url ? iri->orig_url : url, (const char **) &new_url); if (!iri->utf8_encode) new_url = NULL; + else + iri->orig_url = xstrdup (url); } - url_encoded = reencode_escapes (new_url ? new_url : url); + /* XXX XXX Could that change introduce (security) bugs ??? XXX XXX*/ + if (percent_encode) + url_encoded = reencode_escapes (new_url ? new_url : url); + else + url_encoded = new_url ? new_url : url; + p = url_encoded; if (new_url && url_encoded != new_url) @@ -890,7 +914,7 @@ url_parse (const char *url, int *error, struct iri *iri) if (url_encoded == url) u->url = xstrdup (url); else - u->url = url_encoded; + u->url = (char *) url_encoded; } return u; @@ -898,7 +922,7 @@ url_parse (const char *url, int *error, struct iri *iri) error: /* Cleanup in case of error: */ if (url_encoded && url_encoded != url) - xfree (url_encoded); + xfree ((char *) url_encoded); /* Transmit the error code to the caller, if the caller wants to know. */ @@ -910,11 +934,29 @@ url_parse (const char *url, int *error, struct iri *iri) /* Return the error message string from ERROR_CODE, which should have been retrieved from url_parse. The error message is translated. */ -const char * -url_error (int error_code) +char * +url_error (const char *url, int error_code) { assert (error_code >= 0 && ((size_t) error_code) < countof (parse_errors)); - return _(parse_errors[error_code]); + + if (error_code == PE_UNSUPPORTED_SCHEME) + { + char *error, *p; + char *scheme = xstrdup (url); + assert (url_has_scheme (url)); + + if ((p = strchr (scheme, ':'))) + *p = '\0'; + if (!strcasecmp (scheme, "https")) + error = aprintf (_("HTTPS support not compiled in")); + else + error = aprintf (_(parse_errors[error_code]), quote (scheme)); + xfree (scheme); + + return error; + } + else + return xstrdup (_(parse_errors[error_code])); } /* Split PATH into DIR and FILE. PATH comes from the URL and is @@ -1249,7 +1291,9 @@ enum { filechr_control = 4 /* a control character, e.g. 0-31 */ }; -#define FILE_CHAR_TEST(c, mask) (filechr_table[(unsigned char)(c)] & (mask)) +#define FILE_CHAR_TEST(c, mask) \ + ((opt.restrict_files_nonascii && !c_isascii ((unsigned char)(c))) || \ + (filechr_table[(unsigned char)(c)] & (mask))) /* Shorthands for the table: */ #define U filechr_not_unix @@ -1454,11 +1498,17 @@ url_file_name (const struct url *u) const char *u_file, *u_query; char *fname, *unique; + char *index_filename = "index.html"; /* The default index file is index.html */ fnres.base = NULL; fnres.size = 0; fnres.tail = 0; + /* If an alternative index file was defined, change index_filename */ + if (opt.default_page) + index_filename = opt.default_page; + + /* Start with the directory prefix, if specified. */ if (opt.dir_prefix) append_string (opt.dir_prefix, &fnres); @@ -1500,7 +1550,7 @@ url_file_name (const struct url *u) /* Add the file name. */ if (fnres.tail) append_char ('/', &fnres); - u_file = *u->file ? u->file : "index.html"; + u_file = *u->file ? u->file : index_filename; append_uri_pathel (u_file, u_file + strlen (u_file), false, &fnres); /* Append "?query" to the file name. */ @@ -1527,11 +1577,30 @@ url_file_name (const struct url *u) if ((opt.noclobber || opt.always_rest || opt.timestamping || opt.dirstruct) && !(file_exists_p (fname) && !file_non_directory_p (fname))) - return fname; + { + unique = fname; + } + else + { + unique = unique_name (fname, true); + if (unique != fname) + xfree (fname); + } + +/* On VMS, alter the name as required. */ +#ifdef __VMS + { + char *unique2; + + unique2 = ods_conform( unique); + if (unique2 != unique) + { + xfree (unique); + unique = unique2; + } + } +#endif /* def __VMS */ - unique = unique_name (fname, true); - if (unique != fname) - xfree (fname); return unique; } @@ -1969,12 +2038,12 @@ schemes_are_similar_p (enum url_scheme a, enum url_scheme b) static int getchar_from_escaped_string (const char *str, char *c) -{ +{ const char *p = str; assert (str && *str); assert (c); - + if (p[0] == '%') { if (!c_isxdigit(p[1]) || !c_isxdigit(p[2])) @@ -2024,7 +2093,7 @@ are_urls_equal (const char *u1, const char *u2) p += pp; q += qq; } - + return (*p == 0 && *q == 0 ? true : false); } @@ -2133,7 +2202,7 @@ test_append_uri_pathel() } test_array[] = { { "http://www.yoyodyne.com/path/", "somepage.html", false, "http://www.yoyodyne.com/path/somepage.html" }, }; - + for (i = 0; i < sizeof(test_array)/sizeof(test_array[0]); ++i) { struct growable dest;