Loading docs/libcurl/curl_easy_setopt.3 +85 −0 Original line number Diff line number Diff line Loading @@ -84,6 +84,54 @@ If this option is set and libcurl has been built with the standard name resolver, timeouts will not occur while the name resolve takes place. Consider building libcurl with c-ares support to enable asynchronous DNS lookups, which enables nice timeouts for name resolves without signals. .IP CURLOPT_WILDCARDMATCH Set this option to 1 if you want to transfer multiple files according to a file name pattern. The pattern can be specified as part of the \fICURLOPT_URL\fP option, using an fnmatch-like pattern (Shell Pattern Matching) in the last part of URL (file name). By default, libcurl uses its internal implementation of fnmatch(). You can provide your own matching function by the \fICURLOPT_FNMATCH_FUNCTION\fR option. This feature is only supported by the FTP download for now. A brief introduction of its syntax follows: .RS .IP "\fB*\fR - ASTERISK" \&ftp://example.com/some/path/\fB*.txt\fR (for all txt's from the root directory) .RE .RS .IP "\fB?\fR - QUESTION MARK" Question mark matches any (exactly one) character. \&ftp://example.com/some/path/\fBphoto?.jpeg\fR .RE .RS .IP "\fB[\fR - BRACKET EXPRESSION" The left bracket opens a bracket expression. The question mark and asterisk have no special meaning in a bracket expression. Each bracket expression ends by the right bracket and matches exactly one character. Some examples follow: \fB[a-zA-Z0\-9]\fR or \fB[f\-gF\-G]\fR \- character interval \fB[abc]\fR - character enumeration \fB[^abc]\fR or \fB[!abc]\fR - negation \fB[[:\fR\fIname\fR\fB:]]\fR class expression. Supported classes are \fBalnum\fR,\fBlower\fR, \fBspace\fR, \fBalpha\fR, \fBdigit\fR, \fBprint\fR, \fBupper\fR, \fBblank\fR, \fBgraph\fR, \fBxdigit\fR. \fB[][-!^]\fR - special case \- matches only '\-', ']', '[', '!' or '^'. These characters have no special purpose. \fB[\\[\\]\\\\]\fR - escape syntax. Matches '[', ']' or '\\'. Using the rules above, a file name pattern can be constructed: \&ftp://example.com/some/path/\fB[a-z[:upper:]\\\\].jpeg\fR .RE .PP .SH CALLBACK OPTIONS .IP CURLOPT_WRITEFUNCTION Loading Loading @@ -424,6 +472,43 @@ in 7.20.0) .IP CURLOPT_INTERLEAVEDATA This is the stream that will be passed to \fICURLOPT_INTERLEAVEFUNCTION\fP when interleaved RTP data is received. (Added in 7.20.0) .IP CURLOPT_CHUNK_BGN_FUNCTION Function pointer that should match the following prototype: \fBlong function (const void *transfer_info, void *ptr, int remains)\fR. This function gets called by libcurl before a part of the stream is going to be transferred (if the transfer supports chunks). This callback makes sense only when using the \fICURLOPT_WILDCARDMATCH\fR option for now. The target of transfer_info parameter is a "feature depended" structure. For the FTP wildcard download, the target is curl_fileinfo structure (see \fIcurl/curl.h\fR). The parameter ptr is a pointer given by \fICURLOPT_CHUNK_DATA\fR. The parameter remains contains number of chunks remaining per the transfer. If the feature is not available, the parameter has zero value. Return \fICURL_CHUNK_BGN_FUNC_OK\fR if everything is fine, \fICURL_CHUNK_BGN_FUNC_SKIP\fR if you want to skip the concrete chunk or \fICURL_CHUNK_BGN_FUNC_FAIL\fR to tell libcurl to stop if some error occurred. .IP CURLOPT_CHUNK_END_FUNCTION Function pointer that should match the following prototype: \fBlong function(void *ptr)\fR. This function gets called by libcurl as soon as a part of the stream has been transferred (or skipped). Return \fICURL_CHUNK_END_FUNC_OK\fR if everything is fine or \fBCURL_CHUNK_END_FUNC_FAIL\fR to tell the lib to stop if some error occurred. .IP CURLOPT_CHUNK_DATA Pass a pointer that will be untouched by libcurl and passed as the ptr argument to the \fICURL_CHUNK_BGN_FUNTION\fR and \fICURL_CHUNK_END_FUNTION\fR. .IP CURLOPT_FNMATCH_FUNCTION Function pointer that should match \fBint function(const char *pattern, const char *string)\fR prototype (see \fIcurl/curl.h\fR). It is used internally for the wildcard matching feature. Return \fICURL_FNMATCHFUNC_MATCH\fR if pattern matches the string, \fICURL_FNMATCHFUNC_NOMATCH\fR if not or \fICURL_FNMATCHFUNC_FAIL\fR if an error occurred. .SH ERROR OPTIONS .IP CURLOPT_ERRORBUFFER Pass a char * to a buffer that the libcurl may store human readable error Loading docs/libcurl/libcurl-errors.3 +10 −0 Original line number Diff line number Diff line Loading @@ -218,6 +218,16 @@ return code is only returned from \fIcurl_easy_recv(3)\fP and Failed to load CRL file (Added in 7.19.0) .IP "CURLE_SSL_ISSUER_ERROR (83)" Issuer check failed (Added in 7.19.0) .IP "CURLE_FTP_PRET_FAILED (84)" PRET command failed .IP "CURLE_RTSP_CSEQ_ERROR (85)" Mismatch of RTSP CSeq numbers. .IP "CURLE_RTSP_SESSION_ERROR (86)" Mismatch of RTSP Session Identifiers. .IP "CURLE_FTP_BAD_FILE_LIST (87)" Unable to parse FTP file list (during FTP wildcard downloading). .IP "CURLE_CHUNK_FAILED (88)" Chunk callback reported error. .IP "CURLE_OBSOLETE*" These error codes will never be returned. They were used in an old libcurl version and are currently unused. Loading include/curl/curl.h +109 −0 Original line number Diff line number Diff line Loading @@ -198,6 +198,96 @@ typedef size_t (*curl_write_callback)(char *buffer, size_t nitems, void *outstream); /* enumeration of file types */ typedef enum { CURLFILETYPE_FILE = 0, CURLFILETYPE_DIRECTORY, CURLFILETYPE_SYMLINK, CURLFILETYPE_DEVICE_BLOCK, CURLFILETYPE_DEVICE_CHAR, CURLFILETYPE_NAMEDPIPE, CURLFILETYPE_SOCKET, CURLFILETYPE_DOOR, /* is possible only on Sun Solaris now */ CURLFILETYPE_UNKNOWN /* should never occur */ } curlfiletype; #define CURLFINFOFLAG_KNOWN_FILENAME (1<<0) #define CURLFINFOFLAG_KNOWN_FILETYPE (1<<1) #define CURLFINFOFLAG_KNOWN_TIME (1<<2) #define CURLFINFOFLAG_KNOWN_PERM (1<<3) #define CURLFINFOFLAG_KNOWN_UID (1<<4) #define CURLFINFOFLAG_KNOWN_GID (1<<5) #define CURLFINFOFLAG_KNOWN_SIZE (1<<6) #define CURLFINFOFLAG_KNOWN_HLINKCOUNT (1<<7) /* Content of this structure depends on information which is known and is achievable (e.g. by FTP LIST parsing). Please see the url_easy_setopt(3) man page for callbacks returning this structure -- some fields are mandatory, some others are optional. The FLAG field has special meaning. */ struct curl_fileinfo { char *filename; curlfiletype filetype; time_t time; int32_t perm; int uid; int gid; curl_off_t size; long int hardlinks; struct { /* If some of these fields is not NULL, it is a pointer to b_data. */ char *time; char *perm; char *user; char *group; char *target; /* pointer to the target filename of a symlink */ } strings; int32_t flags; /* used internally */ char * b_data; size_t b_size; size_t b_used; }; /* return codes for CURLOPT_CHUNK_BGN_FUNCTION */ #define CURL_CHUNK_BGN_FUNC_OK 0 #define CURL_CHUNK_BGN_FUNC_FAIL 1 /* tell the lib to end the task */ #define CURL_CHUNK_BGN_FUNC_SKIP 2 /* skip this chunk over */ /* if splitting of data transfer is enabled, this callback is called before download of an individual chunk started. Note that parameter "remains" works only for FTP wildcard downloading (for now), otherwise is not used */ typedef long (*curl_chunk_bgn_callback)(const void *transfer_info, void *ptr, int remains); /* return codes for CURLOPT_CHUNK_END_FUNCTION */ #define CURL_CHUNK_END_FUNC_OK 0 #define CURL_CHUNK_END_FUNC_FAIL 1 /* tell the lib to end the task */ /* If splitting of data transfer is enabled this callback is called after download of an individual chunk finished. Note! After this callback was set then it have to be called FOR ALL chunks. Even if downloading of this chunk was skipped in CHUNK_BGN_FUNC. This is the reason why we don't need "transfer_info" parameter in this callback and we are not interested in "remains" parameter too. */ typedef long (*curl_chunk_end_callback)(void *ptr); /* return codes for FNMATCHFUNCTION */ #define CURL_FNMATCHFUNC_MATCH 0 /* string corresponds to the pattern */ #define CURL_FNMATCHFUNC_NOMATCH 1 /* pattern doesn't match the string */ #define CURL_FNMATCHFUNC_FAIL 2 /* an error occurred */ /* callback type for wildcard downloading pattern matching. If the string matches the pattern, return CURL_FNMATCHFUNC_MATCH value, etc. */ typedef int (*curl_fnmatch_callback)(const char *pattern, const char *string); /* These are the return codes for the seek callbacks */ #define CURL_SEEKFUNC_OK 0 #define CURL_SEEKFUNC_FAIL 1 /* fail the entire transfer */ Loading Loading @@ -409,6 +499,8 @@ typedef enum { CURLE_FTP_PRET_FAILED, /* 84 - a PRET command failed */ CURLE_RTSP_CSEQ_ERROR, /* 85 - mismatch of RTSP CSeq numbers */ CURLE_RTSP_SESSION_ERROR, /* 86 - mismatch of RTSP Session Identifiers */ CURLE_FTP_BAD_FILE_LIST, /* 87 - unable to parse FTP file list */ CURLE_CHUNK_FAILED, /* 88 - chunk callback reported error */ CURL_LAST /* never use! */ } CURLcode; Loading Loading @@ -1322,6 +1414,23 @@ typedef enum { /* Let the application define a custom write method for RTP data */ CINIT(INTERLEAVEFUNCTION, FUNCTIONPOINT, 196), /* Turn on wildcard matching */ CINIT(WILDCARDMATCH, LONG, 197), /* Directory matching callback called before downloading of an individual file (chunk) started */ CINIT(CHUNK_BGN_FUNCTION, FUNCTIONPOINT, 198), /* Directory matching callback called after the file (chunk) was downloaded, or skipped */ CINIT(CHUNK_END_FUNCTION, FUNCTIONPOINT, 199), /* Change match (fnmatch-like) callback for wildcard matching */ CINIT(FNMATCH_FUNCTION, FUNCTIONPOINT, 200), /* Let the application define custom chunk data pointer */ CINIT(CHUNK_DATA, OBJECTPOINT, 201), CURLOPT_LASTENTRY /* the last unused */ } CURLoption; Loading lib/Makefile.inc +2 −0 Original line number Diff line number Diff line Loading @@ -4,6 +4,7 @@ CSOURCES = file.c timeval.c base64.c hostip.c progress.c formdata.c \ cookie.c http.c sendf.c ftp.c url.c dict.c if2ip.c speedcheck.c \ ldap.c ssluse.c version.c getenv.c escape.c mprintf.c telnet.c \ netrc.c getinfo.c transfer.c strequal.c easy.c security.c krb4.c \ curl_fnmatch.c fileinfo.c ftplistparser.c wildcard.c \ krb5.c memdebug.c http_chunks.c strtok.c connect.c llist.c hash.c \ multi.c content_encoding.c share.c http_digest.c md5.c curl_rand.c \ http_negotiate.c http_ntlm.c inet_pton.c strtoofft.c strerror.c \ Loading @@ -18,6 +19,7 @@ HHEADERS = arpa_telnet.h netrc.h file.h timeval.h qssl.h hostip.h \ progress.h formdata.h cookie.h http.h sendf.h ftp.h url.h dict.h \ if2ip.h speedcheck.h urldata.h curl_ldap.h ssluse.h escape.h telnet.h \ getinfo.h strequal.h krb4.h memdebug.h http_chunks.h curl_rand.h \ curl_fnmatch.h wildcard.h fileinfo.h ftplistparser.h \ strtok.h connect.h llist.h hash.h content_encoding.h share.h \ curl_md5.h http_digest.h http_negotiate.h http_ntlm.h inet_pton.h \ strtoofft.h strerror.h inet_ntop.h curlx.h curl_memory.h setup.h \ Loading lib/curl_fnmatch.c 0 → 100644 +410 −0 Original line number Diff line number Diff line /*************************************************************************** * _ _ ____ _ * Project ___| | | | _ \| | * / __| | | | |_) | | * | (__| |_| | _ <| |___ * \___|\___/|_| \_\_____| * * Copyright (C) 1998 - 2009, Daniel Stenberg, <daniel@haxx.se>, et al. * * This software is licensed as described in the file COPYING, which * you should have received as part of this distribution. The terms * are also available at http://curl.haxx.se/docs/copyright.html. * * You may opt to use, copy, modify, merge, publish, distribute and/or sell * copies of the Software, and permit persons to whom the Software is * furnished to do so, under the terms of the COPYING file. * * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY * KIND, either express or implied. * ***************************************************************************/ #include "curl_fnmatch.h" #include "setup.h" #define CURLFNM_CHARSET_LEN (sizeof(char) * 256) #define CURLFNM_CHSET_SIZE (CURLFNM_CHARSET_LEN + 15) #define CURLFNM_NEGATE CURLFNM_CHARSET_LEN #define CURLFNM_ALNUM (CURLFNM_CHARSET_LEN + 1) #define CURLFNM_DIGIT (CURLFNM_CHARSET_LEN + 2) #define CURLFNM_XDIGIT (CURLFNM_CHARSET_LEN + 3) #define CURLFNM_ALPHA (CURLFNM_CHARSET_LEN + 4) #define CURLFNM_PRINT (CURLFNM_CHARSET_LEN + 5) #define CURLFNM_BLANK (CURLFNM_CHARSET_LEN + 6) #define CURLFNM_LOWER (CURLFNM_CHARSET_LEN + 7) #define CURLFNM_GRAPH (CURLFNM_CHARSET_LEN + 8) #define CURLFNM_SPACE (CURLFNM_CHARSET_LEN + 9) #define CURLFNM_UPPER (CURLFNM_CHARSET_LEN + 10) typedef enum { CURLFNM_LOOP_DEFAULT = 0, CURLFNM_LOOP_BACKSLASH } loop_state; typedef enum { CURLFNM_SCHS_DEFAULT = 0, CURLFNM_SCHS_MAYRANGE, CURLFNM_SCHS_MAYRANGE2, CURLFNM_SCHS_RIGHTBR, CURLFNM_SCHS_RIGHTBRLEFTBR } setcharset_state; typedef enum { CURLFNM_PKW_INIT = 0, CURLFNM_PKW_DDOT } parsekey_state; #define SETCHARSET_OK 1 #define SETCHARSET_FAIL 0 static int parsekeyword(unsigned char **pattern, unsigned char *charset) { parsekey_state state = CURLFNM_PKW_INIT; #define KEYLEN 10 char keyword[KEYLEN] = { 0 }; int found = FALSE; int i; register unsigned char *p = *pattern; for(i = 0; !found; i++) { char c = *p++; if(i >= KEYLEN) return SETCHARSET_FAIL; switch(state) { case CURLFNM_PKW_INIT: if(ISALPHA(c) && ISLOWER(c)) keyword[i] = c; else if(c == ':') state = CURLFNM_PKW_DDOT; else return 0; break; case CURLFNM_PKW_DDOT: if(c == ']') found = TRUE; else return SETCHARSET_FAIL; } } #undef KEYLEN *pattern = p; /* move caller's pattern pointer */ if(strcmp(keyword, "digit") == 0) charset[CURLFNM_DIGIT] = 1; else if(strcmp(keyword, "alnum") == 0) charset[CURLFNM_ALNUM] = 1; else if(strcmp(keyword, "alpha") == 0) charset[CURLFNM_ALPHA] = 1; else if(strcmp(keyword, "xdigit") == 0) charset[CURLFNM_XDIGIT] = 1; else if(strcmp(keyword, "print") == 0) charset[CURLFNM_PRINT] = 1; else if(strcmp(keyword, "graph") == 0) charset[CURLFNM_GRAPH] = 1; else if(strcmp(keyword, "space") == 0) charset[CURLFNM_SPACE] = 1; else if(strcmp(keyword, "blank") == 0) charset[CURLFNM_BLANK] = 1; else if(strcmp(keyword, "upper") == 0) charset[CURLFNM_UPPER] = 1; else if(strcmp(keyword, "lower") == 0) charset[CURLFNM_LOWER] = 1; else return SETCHARSET_FAIL; return SETCHARSET_OK; } /* returns 1 (true) if pattern is OK, 0 if is bad ("p" is pattern pointer) */ static int setcharset(unsigned char **p, unsigned char *charset) { setcharset_state state = CURLFNM_SCHS_DEFAULT; unsigned char rangestart = 0; unsigned char lastchar = 0; bool something_found = FALSE; register unsigned char c; for(;;) { c = **p; switch(state){ case CURLFNM_SCHS_DEFAULT: if(ISALNUM(c)) { /* ASCII value */ rangestart = c; charset[c] = 1; (*p)++; state = CURLFNM_SCHS_MAYRANGE; something_found = TRUE; } else if(c == ']') { if(something_found) return SETCHARSET_OK; else something_found = TRUE; state = CURLFNM_SCHS_RIGHTBR; charset[c] = 1; (*p)++; } else if(c == '[') { char c2 = *((*p)+1); if(c2 == ':') { /* there has to be a keyword */ (*p) += 2; if(parsekeyword(p, charset)) { state = CURLFNM_SCHS_DEFAULT; } else return SETCHARSET_FAIL; } else { charset[c] = 1; (*p)++; } something_found = TRUE; } else if(c == '?' || c == '*') { something_found = TRUE; charset[c] = 1; (*p)++; } else if(c == '^' || c == '!') { if(!something_found) { if(charset[CURLFNM_NEGATE]) { charset[c] = 1; something_found = 1; } else charset[CURLFNM_NEGATE] = 1; /* negate charset */ } else charset[c] = 1; (*p)++; } else if(c == '\\') { c = *(++(*p)); if(ISPRINT((c))) { something_found = TRUE; state = CURLFNM_SCHS_MAYRANGE; charset[c] = 1; rangestart = c; (*p)++; } else return SETCHARSET_FAIL; } else if(c == '\0') { return SETCHARSET_FAIL; } else { charset[c] = 1; (*p)++; something_found = TRUE; } break; case CURLFNM_SCHS_MAYRANGE: if(c == '-'){ charset[c] = 1; (*p)++; lastchar = '-'; state = CURLFNM_SCHS_MAYRANGE2; } else if(c == '[') { state = CURLFNM_SCHS_DEFAULT; } else if(ISALNUM(c)) { charset[c] = 1; (*p)++; } else if(c == '\\') { c = *(++(*p)); if(isprint(c)) { charset[c] = 1; (*p)++; } else return SETCHARSET_FAIL; } else if(c == ']') { return SETCHARSET_OK; } else return SETCHARSET_FAIL; break; case CURLFNM_SCHS_MAYRANGE2: if(c == '\\') { c = *(++(*p)); if(!ISPRINT(c)) return SETCHARSET_FAIL; } if(c == ']') { return SETCHARSET_OK; } else if(c == '\\') { c = *(++(*p)); if(ISPRINT(c)) { charset[c] = 1; state = CURLFNM_SCHS_DEFAULT; (*p)++; } else return SETCHARSET_FAIL; } if(c >= rangestart) { if((ISLOWER(c) && ISLOWER(rangestart)) || (ISDIGIT(c) && ISDIGIT(rangestart)) || (ISUPPER(c) && ISUPPER(rangestart))) { charset[lastchar] = 0; rangestart++; while(rangestart++ <= c) charset[rangestart-1] = 1; (*p)++; state = CURLFNM_SCHS_DEFAULT; } else return SETCHARSET_FAIL; } break; case CURLFNM_SCHS_RIGHTBR: if(c == '[') { state = CURLFNM_SCHS_RIGHTBRLEFTBR; charset[c] = 1; (*p)++; } else if(c == ']') { return SETCHARSET_OK; } else if(c == '\0') { return SETCHARSET_FAIL; } else if(ISPRINT(c)){ charset[c] = 1; (*p)++; state = CURLFNM_SCHS_DEFAULT; } else return SETCHARSET_FAIL; break; case CURLFNM_SCHS_RIGHTBRLEFTBR: if(c == ']') { return SETCHARSET_OK; } else { state = CURLFNM_SCHS_DEFAULT; charset[c] = 1; (*p)++; } break; } } return SETCHARSET_FAIL; } static int loop(const unsigned char *pattern, const unsigned char *string) { loop_state state = CURLFNM_LOOP_DEFAULT; register unsigned char *p = (unsigned char *)pattern; register unsigned char *s = (unsigned char *)string; unsigned char charset[CURLFNM_CHSET_SIZE] = { 0 }; int rc = 0; for (;;) { switch(state) { case CURLFNM_LOOP_DEFAULT: if(*p == '*') { while(*(p+1) == '*') /* eliminate multiple stars */ p++; if(*s == '\0' && *(p+1) == '\0') return CURL_FNMATCH_MATCH; rc = loop(p + 1, s); /* *.txt matches .txt <=> .txt matches .txt */ if(rc == CURL_FNMATCH_MATCH) return CURL_FNMATCH_MATCH; if(*s) /* let the star eat up one character */ s++; else return CURL_FNMATCH_NOMATCH; } else if(*p == '?') { if(ISPRINT(*s)) { s++; p++; } else if(*s == '\0') return CURL_FNMATCH_NOMATCH; else return CURL_FNMATCH_FAIL; /* cannot deal with other character */ } else if(*p == '\0') { if(*s == '\0') return CURL_FNMATCH_MATCH; else return CURL_FNMATCH_NOMATCH; } else if(*p == '\\') { state = CURLFNM_LOOP_BACKSLASH; p++; } else if(*p == '[') { unsigned char *pp = p+1; /* cannot handle with pointer to register */ if(setcharset(&pp, charset)) { bool found = FALSE; if(charset[(unsigned int)*s]) found = TRUE; else if(charset[CURLFNM_ALNUM]) found = ISALNUM(*s); else if(charset[CURLFNM_ALPHA]) found = ISALPHA(*s); else if(charset[CURLFNM_DIGIT]) found = ISDIGIT(*s); else if(charset[CURLFNM_XDIGIT]) found = ISXDIGIT(*s); else if(charset[CURLFNM_PRINT]) found = ISPRINT(*s); else if(charset[CURLFNM_SPACE]) found = ISSPACE(*s); else if(charset[CURLFNM_UPPER]) found = ISUPPER(*s); else if(charset[CURLFNM_LOWER]) found = ISLOWER(*s); else if(charset[CURLFNM_BLANK]) found = ISBLANK(*s); else if(charset[CURLFNM_GRAPH]) found = ISGRAPH(*s); if(charset[CURLFNM_NEGATE]) found = !found; if(found) { p = pp+1; s++; memset(charset, 0, CURLFNM_CHSET_SIZE); } else return CURL_FNMATCH_NOMATCH; } else return CURL_FNMATCH_FAIL; } else { if(*p++ != *s++) return CURL_FNMATCH_NOMATCH; } break; case CURLFNM_LOOP_BACKSLASH: if(ISPRINT(*p)) { if(*p++ == *s++) state = CURLFNM_LOOP_DEFAULT; else return CURL_FNMATCH_NOMATCH; } else return CURL_FNMATCH_FAIL; break; } } } int Curl_fnmatch(const char *pattern, const char *string) { if(!pattern || !string) { return CURL_FNMATCH_FAIL; } return loop((unsigned char *)pattern, (unsigned char *)string); } Loading
docs/libcurl/curl_easy_setopt.3 +85 −0 Original line number Diff line number Diff line Loading @@ -84,6 +84,54 @@ If this option is set and libcurl has been built with the standard name resolver, timeouts will not occur while the name resolve takes place. Consider building libcurl with c-ares support to enable asynchronous DNS lookups, which enables nice timeouts for name resolves without signals. .IP CURLOPT_WILDCARDMATCH Set this option to 1 if you want to transfer multiple files according to a file name pattern. The pattern can be specified as part of the \fICURLOPT_URL\fP option, using an fnmatch-like pattern (Shell Pattern Matching) in the last part of URL (file name). By default, libcurl uses its internal implementation of fnmatch(). You can provide your own matching function by the \fICURLOPT_FNMATCH_FUNCTION\fR option. This feature is only supported by the FTP download for now. A brief introduction of its syntax follows: .RS .IP "\fB*\fR - ASTERISK" \&ftp://example.com/some/path/\fB*.txt\fR (for all txt's from the root directory) .RE .RS .IP "\fB?\fR - QUESTION MARK" Question mark matches any (exactly one) character. \&ftp://example.com/some/path/\fBphoto?.jpeg\fR .RE .RS .IP "\fB[\fR - BRACKET EXPRESSION" The left bracket opens a bracket expression. The question mark and asterisk have no special meaning in a bracket expression. Each bracket expression ends by the right bracket and matches exactly one character. Some examples follow: \fB[a-zA-Z0\-9]\fR or \fB[f\-gF\-G]\fR \- character interval \fB[abc]\fR - character enumeration \fB[^abc]\fR or \fB[!abc]\fR - negation \fB[[:\fR\fIname\fR\fB:]]\fR class expression. Supported classes are \fBalnum\fR,\fBlower\fR, \fBspace\fR, \fBalpha\fR, \fBdigit\fR, \fBprint\fR, \fBupper\fR, \fBblank\fR, \fBgraph\fR, \fBxdigit\fR. \fB[][-!^]\fR - special case \- matches only '\-', ']', '[', '!' or '^'. These characters have no special purpose. \fB[\\[\\]\\\\]\fR - escape syntax. Matches '[', ']' or '\\'. Using the rules above, a file name pattern can be constructed: \&ftp://example.com/some/path/\fB[a-z[:upper:]\\\\].jpeg\fR .RE .PP .SH CALLBACK OPTIONS .IP CURLOPT_WRITEFUNCTION Loading Loading @@ -424,6 +472,43 @@ in 7.20.0) .IP CURLOPT_INTERLEAVEDATA This is the stream that will be passed to \fICURLOPT_INTERLEAVEFUNCTION\fP when interleaved RTP data is received. (Added in 7.20.0) .IP CURLOPT_CHUNK_BGN_FUNCTION Function pointer that should match the following prototype: \fBlong function (const void *transfer_info, void *ptr, int remains)\fR. This function gets called by libcurl before a part of the stream is going to be transferred (if the transfer supports chunks). This callback makes sense only when using the \fICURLOPT_WILDCARDMATCH\fR option for now. The target of transfer_info parameter is a "feature depended" structure. For the FTP wildcard download, the target is curl_fileinfo structure (see \fIcurl/curl.h\fR). The parameter ptr is a pointer given by \fICURLOPT_CHUNK_DATA\fR. The parameter remains contains number of chunks remaining per the transfer. If the feature is not available, the parameter has zero value. Return \fICURL_CHUNK_BGN_FUNC_OK\fR if everything is fine, \fICURL_CHUNK_BGN_FUNC_SKIP\fR if you want to skip the concrete chunk or \fICURL_CHUNK_BGN_FUNC_FAIL\fR to tell libcurl to stop if some error occurred. .IP CURLOPT_CHUNK_END_FUNCTION Function pointer that should match the following prototype: \fBlong function(void *ptr)\fR. This function gets called by libcurl as soon as a part of the stream has been transferred (or skipped). Return \fICURL_CHUNK_END_FUNC_OK\fR if everything is fine or \fBCURL_CHUNK_END_FUNC_FAIL\fR to tell the lib to stop if some error occurred. .IP CURLOPT_CHUNK_DATA Pass a pointer that will be untouched by libcurl and passed as the ptr argument to the \fICURL_CHUNK_BGN_FUNTION\fR and \fICURL_CHUNK_END_FUNTION\fR. .IP CURLOPT_FNMATCH_FUNCTION Function pointer that should match \fBint function(const char *pattern, const char *string)\fR prototype (see \fIcurl/curl.h\fR). It is used internally for the wildcard matching feature. Return \fICURL_FNMATCHFUNC_MATCH\fR if pattern matches the string, \fICURL_FNMATCHFUNC_NOMATCH\fR if not or \fICURL_FNMATCHFUNC_FAIL\fR if an error occurred. .SH ERROR OPTIONS .IP CURLOPT_ERRORBUFFER Pass a char * to a buffer that the libcurl may store human readable error Loading
docs/libcurl/libcurl-errors.3 +10 −0 Original line number Diff line number Diff line Loading @@ -218,6 +218,16 @@ return code is only returned from \fIcurl_easy_recv(3)\fP and Failed to load CRL file (Added in 7.19.0) .IP "CURLE_SSL_ISSUER_ERROR (83)" Issuer check failed (Added in 7.19.0) .IP "CURLE_FTP_PRET_FAILED (84)" PRET command failed .IP "CURLE_RTSP_CSEQ_ERROR (85)" Mismatch of RTSP CSeq numbers. .IP "CURLE_RTSP_SESSION_ERROR (86)" Mismatch of RTSP Session Identifiers. .IP "CURLE_FTP_BAD_FILE_LIST (87)" Unable to parse FTP file list (during FTP wildcard downloading). .IP "CURLE_CHUNK_FAILED (88)" Chunk callback reported error. .IP "CURLE_OBSOLETE*" These error codes will never be returned. They were used in an old libcurl version and are currently unused. Loading
include/curl/curl.h +109 −0 Original line number Diff line number Diff line Loading @@ -198,6 +198,96 @@ typedef size_t (*curl_write_callback)(char *buffer, size_t nitems, void *outstream); /* enumeration of file types */ typedef enum { CURLFILETYPE_FILE = 0, CURLFILETYPE_DIRECTORY, CURLFILETYPE_SYMLINK, CURLFILETYPE_DEVICE_BLOCK, CURLFILETYPE_DEVICE_CHAR, CURLFILETYPE_NAMEDPIPE, CURLFILETYPE_SOCKET, CURLFILETYPE_DOOR, /* is possible only on Sun Solaris now */ CURLFILETYPE_UNKNOWN /* should never occur */ } curlfiletype; #define CURLFINFOFLAG_KNOWN_FILENAME (1<<0) #define CURLFINFOFLAG_KNOWN_FILETYPE (1<<1) #define CURLFINFOFLAG_KNOWN_TIME (1<<2) #define CURLFINFOFLAG_KNOWN_PERM (1<<3) #define CURLFINFOFLAG_KNOWN_UID (1<<4) #define CURLFINFOFLAG_KNOWN_GID (1<<5) #define CURLFINFOFLAG_KNOWN_SIZE (1<<6) #define CURLFINFOFLAG_KNOWN_HLINKCOUNT (1<<7) /* Content of this structure depends on information which is known and is achievable (e.g. by FTP LIST parsing). Please see the url_easy_setopt(3) man page for callbacks returning this structure -- some fields are mandatory, some others are optional. The FLAG field has special meaning. */ struct curl_fileinfo { char *filename; curlfiletype filetype; time_t time; int32_t perm; int uid; int gid; curl_off_t size; long int hardlinks; struct { /* If some of these fields is not NULL, it is a pointer to b_data. */ char *time; char *perm; char *user; char *group; char *target; /* pointer to the target filename of a symlink */ } strings; int32_t flags; /* used internally */ char * b_data; size_t b_size; size_t b_used; }; /* return codes for CURLOPT_CHUNK_BGN_FUNCTION */ #define CURL_CHUNK_BGN_FUNC_OK 0 #define CURL_CHUNK_BGN_FUNC_FAIL 1 /* tell the lib to end the task */ #define CURL_CHUNK_BGN_FUNC_SKIP 2 /* skip this chunk over */ /* if splitting of data transfer is enabled, this callback is called before download of an individual chunk started. Note that parameter "remains" works only for FTP wildcard downloading (for now), otherwise is not used */ typedef long (*curl_chunk_bgn_callback)(const void *transfer_info, void *ptr, int remains); /* return codes for CURLOPT_CHUNK_END_FUNCTION */ #define CURL_CHUNK_END_FUNC_OK 0 #define CURL_CHUNK_END_FUNC_FAIL 1 /* tell the lib to end the task */ /* If splitting of data transfer is enabled this callback is called after download of an individual chunk finished. Note! After this callback was set then it have to be called FOR ALL chunks. Even if downloading of this chunk was skipped in CHUNK_BGN_FUNC. This is the reason why we don't need "transfer_info" parameter in this callback and we are not interested in "remains" parameter too. */ typedef long (*curl_chunk_end_callback)(void *ptr); /* return codes for FNMATCHFUNCTION */ #define CURL_FNMATCHFUNC_MATCH 0 /* string corresponds to the pattern */ #define CURL_FNMATCHFUNC_NOMATCH 1 /* pattern doesn't match the string */ #define CURL_FNMATCHFUNC_FAIL 2 /* an error occurred */ /* callback type for wildcard downloading pattern matching. If the string matches the pattern, return CURL_FNMATCHFUNC_MATCH value, etc. */ typedef int (*curl_fnmatch_callback)(const char *pattern, const char *string); /* These are the return codes for the seek callbacks */ #define CURL_SEEKFUNC_OK 0 #define CURL_SEEKFUNC_FAIL 1 /* fail the entire transfer */ Loading Loading @@ -409,6 +499,8 @@ typedef enum { CURLE_FTP_PRET_FAILED, /* 84 - a PRET command failed */ CURLE_RTSP_CSEQ_ERROR, /* 85 - mismatch of RTSP CSeq numbers */ CURLE_RTSP_SESSION_ERROR, /* 86 - mismatch of RTSP Session Identifiers */ CURLE_FTP_BAD_FILE_LIST, /* 87 - unable to parse FTP file list */ CURLE_CHUNK_FAILED, /* 88 - chunk callback reported error */ CURL_LAST /* never use! */ } CURLcode; Loading Loading @@ -1322,6 +1414,23 @@ typedef enum { /* Let the application define a custom write method for RTP data */ CINIT(INTERLEAVEFUNCTION, FUNCTIONPOINT, 196), /* Turn on wildcard matching */ CINIT(WILDCARDMATCH, LONG, 197), /* Directory matching callback called before downloading of an individual file (chunk) started */ CINIT(CHUNK_BGN_FUNCTION, FUNCTIONPOINT, 198), /* Directory matching callback called after the file (chunk) was downloaded, or skipped */ CINIT(CHUNK_END_FUNCTION, FUNCTIONPOINT, 199), /* Change match (fnmatch-like) callback for wildcard matching */ CINIT(FNMATCH_FUNCTION, FUNCTIONPOINT, 200), /* Let the application define custom chunk data pointer */ CINIT(CHUNK_DATA, OBJECTPOINT, 201), CURLOPT_LASTENTRY /* the last unused */ } CURLoption; Loading
lib/Makefile.inc +2 −0 Original line number Diff line number Diff line Loading @@ -4,6 +4,7 @@ CSOURCES = file.c timeval.c base64.c hostip.c progress.c formdata.c \ cookie.c http.c sendf.c ftp.c url.c dict.c if2ip.c speedcheck.c \ ldap.c ssluse.c version.c getenv.c escape.c mprintf.c telnet.c \ netrc.c getinfo.c transfer.c strequal.c easy.c security.c krb4.c \ curl_fnmatch.c fileinfo.c ftplistparser.c wildcard.c \ krb5.c memdebug.c http_chunks.c strtok.c connect.c llist.c hash.c \ multi.c content_encoding.c share.c http_digest.c md5.c curl_rand.c \ http_negotiate.c http_ntlm.c inet_pton.c strtoofft.c strerror.c \ Loading @@ -18,6 +19,7 @@ HHEADERS = arpa_telnet.h netrc.h file.h timeval.h qssl.h hostip.h \ progress.h formdata.h cookie.h http.h sendf.h ftp.h url.h dict.h \ if2ip.h speedcheck.h urldata.h curl_ldap.h ssluse.h escape.h telnet.h \ getinfo.h strequal.h krb4.h memdebug.h http_chunks.h curl_rand.h \ curl_fnmatch.h wildcard.h fileinfo.h ftplistparser.h \ strtok.h connect.h llist.h hash.h content_encoding.h share.h \ curl_md5.h http_digest.h http_negotiate.h http_ntlm.h inet_pton.h \ strtoofft.h strerror.h inet_ntop.h curlx.h curl_memory.h setup.h \ Loading
lib/curl_fnmatch.c 0 → 100644 +410 −0 Original line number Diff line number Diff line /*************************************************************************** * _ _ ____ _ * Project ___| | | | _ \| | * / __| | | | |_) | | * | (__| |_| | _ <| |___ * \___|\___/|_| \_\_____| * * Copyright (C) 1998 - 2009, Daniel Stenberg, <daniel@haxx.se>, et al. * * This software is licensed as described in the file COPYING, which * you should have received as part of this distribution. The terms * are also available at http://curl.haxx.se/docs/copyright.html. * * You may opt to use, copy, modify, merge, publish, distribute and/or sell * copies of the Software, and permit persons to whom the Software is * furnished to do so, under the terms of the COPYING file. * * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY * KIND, either express or implied. * ***************************************************************************/ #include "curl_fnmatch.h" #include "setup.h" #define CURLFNM_CHARSET_LEN (sizeof(char) * 256) #define CURLFNM_CHSET_SIZE (CURLFNM_CHARSET_LEN + 15) #define CURLFNM_NEGATE CURLFNM_CHARSET_LEN #define CURLFNM_ALNUM (CURLFNM_CHARSET_LEN + 1) #define CURLFNM_DIGIT (CURLFNM_CHARSET_LEN + 2) #define CURLFNM_XDIGIT (CURLFNM_CHARSET_LEN + 3) #define CURLFNM_ALPHA (CURLFNM_CHARSET_LEN + 4) #define CURLFNM_PRINT (CURLFNM_CHARSET_LEN + 5) #define CURLFNM_BLANK (CURLFNM_CHARSET_LEN + 6) #define CURLFNM_LOWER (CURLFNM_CHARSET_LEN + 7) #define CURLFNM_GRAPH (CURLFNM_CHARSET_LEN + 8) #define CURLFNM_SPACE (CURLFNM_CHARSET_LEN + 9) #define CURLFNM_UPPER (CURLFNM_CHARSET_LEN + 10) typedef enum { CURLFNM_LOOP_DEFAULT = 0, CURLFNM_LOOP_BACKSLASH } loop_state; typedef enum { CURLFNM_SCHS_DEFAULT = 0, CURLFNM_SCHS_MAYRANGE, CURLFNM_SCHS_MAYRANGE2, CURLFNM_SCHS_RIGHTBR, CURLFNM_SCHS_RIGHTBRLEFTBR } setcharset_state; typedef enum { CURLFNM_PKW_INIT = 0, CURLFNM_PKW_DDOT } parsekey_state; #define SETCHARSET_OK 1 #define SETCHARSET_FAIL 0 static int parsekeyword(unsigned char **pattern, unsigned char *charset) { parsekey_state state = CURLFNM_PKW_INIT; #define KEYLEN 10 char keyword[KEYLEN] = { 0 }; int found = FALSE; int i; register unsigned char *p = *pattern; for(i = 0; !found; i++) { char c = *p++; if(i >= KEYLEN) return SETCHARSET_FAIL; switch(state) { case CURLFNM_PKW_INIT: if(ISALPHA(c) && ISLOWER(c)) keyword[i] = c; else if(c == ':') state = CURLFNM_PKW_DDOT; else return 0; break; case CURLFNM_PKW_DDOT: if(c == ']') found = TRUE; else return SETCHARSET_FAIL; } } #undef KEYLEN *pattern = p; /* move caller's pattern pointer */ if(strcmp(keyword, "digit") == 0) charset[CURLFNM_DIGIT] = 1; else if(strcmp(keyword, "alnum") == 0) charset[CURLFNM_ALNUM] = 1; else if(strcmp(keyword, "alpha") == 0) charset[CURLFNM_ALPHA] = 1; else if(strcmp(keyword, "xdigit") == 0) charset[CURLFNM_XDIGIT] = 1; else if(strcmp(keyword, "print") == 0) charset[CURLFNM_PRINT] = 1; else if(strcmp(keyword, "graph") == 0) charset[CURLFNM_GRAPH] = 1; else if(strcmp(keyword, "space") == 0) charset[CURLFNM_SPACE] = 1; else if(strcmp(keyword, "blank") == 0) charset[CURLFNM_BLANK] = 1; else if(strcmp(keyword, "upper") == 0) charset[CURLFNM_UPPER] = 1; else if(strcmp(keyword, "lower") == 0) charset[CURLFNM_LOWER] = 1; else return SETCHARSET_FAIL; return SETCHARSET_OK; } /* returns 1 (true) if pattern is OK, 0 if is bad ("p" is pattern pointer) */ static int setcharset(unsigned char **p, unsigned char *charset) { setcharset_state state = CURLFNM_SCHS_DEFAULT; unsigned char rangestart = 0; unsigned char lastchar = 0; bool something_found = FALSE; register unsigned char c; for(;;) { c = **p; switch(state){ case CURLFNM_SCHS_DEFAULT: if(ISALNUM(c)) { /* ASCII value */ rangestart = c; charset[c] = 1; (*p)++; state = CURLFNM_SCHS_MAYRANGE; something_found = TRUE; } else if(c == ']') { if(something_found) return SETCHARSET_OK; else something_found = TRUE; state = CURLFNM_SCHS_RIGHTBR; charset[c] = 1; (*p)++; } else if(c == '[') { char c2 = *((*p)+1); if(c2 == ':') { /* there has to be a keyword */ (*p) += 2; if(parsekeyword(p, charset)) { state = CURLFNM_SCHS_DEFAULT; } else return SETCHARSET_FAIL; } else { charset[c] = 1; (*p)++; } something_found = TRUE; } else if(c == '?' || c == '*') { something_found = TRUE; charset[c] = 1; (*p)++; } else if(c == '^' || c == '!') { if(!something_found) { if(charset[CURLFNM_NEGATE]) { charset[c] = 1; something_found = 1; } else charset[CURLFNM_NEGATE] = 1; /* negate charset */ } else charset[c] = 1; (*p)++; } else if(c == '\\') { c = *(++(*p)); if(ISPRINT((c))) { something_found = TRUE; state = CURLFNM_SCHS_MAYRANGE; charset[c] = 1; rangestart = c; (*p)++; } else return SETCHARSET_FAIL; } else if(c == '\0') { return SETCHARSET_FAIL; } else { charset[c] = 1; (*p)++; something_found = TRUE; } break; case CURLFNM_SCHS_MAYRANGE: if(c == '-'){ charset[c] = 1; (*p)++; lastchar = '-'; state = CURLFNM_SCHS_MAYRANGE2; } else if(c == '[') { state = CURLFNM_SCHS_DEFAULT; } else if(ISALNUM(c)) { charset[c] = 1; (*p)++; } else if(c == '\\') { c = *(++(*p)); if(isprint(c)) { charset[c] = 1; (*p)++; } else return SETCHARSET_FAIL; } else if(c == ']') { return SETCHARSET_OK; } else return SETCHARSET_FAIL; break; case CURLFNM_SCHS_MAYRANGE2: if(c == '\\') { c = *(++(*p)); if(!ISPRINT(c)) return SETCHARSET_FAIL; } if(c == ']') { return SETCHARSET_OK; } else if(c == '\\') { c = *(++(*p)); if(ISPRINT(c)) { charset[c] = 1; state = CURLFNM_SCHS_DEFAULT; (*p)++; } else return SETCHARSET_FAIL; } if(c >= rangestart) { if((ISLOWER(c) && ISLOWER(rangestart)) || (ISDIGIT(c) && ISDIGIT(rangestart)) || (ISUPPER(c) && ISUPPER(rangestart))) { charset[lastchar] = 0; rangestart++; while(rangestart++ <= c) charset[rangestart-1] = 1; (*p)++; state = CURLFNM_SCHS_DEFAULT; } else return SETCHARSET_FAIL; } break; case CURLFNM_SCHS_RIGHTBR: if(c == '[') { state = CURLFNM_SCHS_RIGHTBRLEFTBR; charset[c] = 1; (*p)++; } else if(c == ']') { return SETCHARSET_OK; } else if(c == '\0') { return SETCHARSET_FAIL; } else if(ISPRINT(c)){ charset[c] = 1; (*p)++; state = CURLFNM_SCHS_DEFAULT; } else return SETCHARSET_FAIL; break; case CURLFNM_SCHS_RIGHTBRLEFTBR: if(c == ']') { return SETCHARSET_OK; } else { state = CURLFNM_SCHS_DEFAULT; charset[c] = 1; (*p)++; } break; } } return SETCHARSET_FAIL; } static int loop(const unsigned char *pattern, const unsigned char *string) { loop_state state = CURLFNM_LOOP_DEFAULT; register unsigned char *p = (unsigned char *)pattern; register unsigned char *s = (unsigned char *)string; unsigned char charset[CURLFNM_CHSET_SIZE] = { 0 }; int rc = 0; for (;;) { switch(state) { case CURLFNM_LOOP_DEFAULT: if(*p == '*') { while(*(p+1) == '*') /* eliminate multiple stars */ p++; if(*s == '\0' && *(p+1) == '\0') return CURL_FNMATCH_MATCH; rc = loop(p + 1, s); /* *.txt matches .txt <=> .txt matches .txt */ if(rc == CURL_FNMATCH_MATCH) return CURL_FNMATCH_MATCH; if(*s) /* let the star eat up one character */ s++; else return CURL_FNMATCH_NOMATCH; } else if(*p == '?') { if(ISPRINT(*s)) { s++; p++; } else if(*s == '\0') return CURL_FNMATCH_NOMATCH; else return CURL_FNMATCH_FAIL; /* cannot deal with other character */ } else if(*p == '\0') { if(*s == '\0') return CURL_FNMATCH_MATCH; else return CURL_FNMATCH_NOMATCH; } else if(*p == '\\') { state = CURLFNM_LOOP_BACKSLASH; p++; } else if(*p == '[') { unsigned char *pp = p+1; /* cannot handle with pointer to register */ if(setcharset(&pp, charset)) { bool found = FALSE; if(charset[(unsigned int)*s]) found = TRUE; else if(charset[CURLFNM_ALNUM]) found = ISALNUM(*s); else if(charset[CURLFNM_ALPHA]) found = ISALPHA(*s); else if(charset[CURLFNM_DIGIT]) found = ISDIGIT(*s); else if(charset[CURLFNM_XDIGIT]) found = ISXDIGIT(*s); else if(charset[CURLFNM_PRINT]) found = ISPRINT(*s); else if(charset[CURLFNM_SPACE]) found = ISSPACE(*s); else if(charset[CURLFNM_UPPER]) found = ISUPPER(*s); else if(charset[CURLFNM_LOWER]) found = ISLOWER(*s); else if(charset[CURLFNM_BLANK]) found = ISBLANK(*s); else if(charset[CURLFNM_GRAPH]) found = ISGRAPH(*s); if(charset[CURLFNM_NEGATE]) found = !found; if(found) { p = pp+1; s++; memset(charset, 0, CURLFNM_CHSET_SIZE); } else return CURL_FNMATCH_NOMATCH; } else return CURL_FNMATCH_FAIL; } else { if(*p++ != *s++) return CURL_FNMATCH_NOMATCH; } break; case CURLFNM_LOOP_BACKSLASH: if(ISPRINT(*p)) { if(*p++ == *s++) state = CURLFNM_LOOP_DEFAULT; else return CURL_FNMATCH_NOMATCH; } else return CURL_FNMATCH_FAIL; break; } } } int Curl_fnmatch(const char *pattern, const char *string) { if(!pattern || !string) { return CURL_FNMATCH_FAIL; } return loop((unsigned char *)pattern, (unsigned char *)string); }
