diff options
author | Daniel Stenberg <daniel@haxx.se> | 2021-10-08 20:12:26 +0200 |
---|---|---|
committer | Daniel Stenberg <daniel@haxx.se> | 2021-10-11 17:04:14 +0200 |
commit | 9a8564a920188e49d5bd8c1c8573ddef97f6e03a (patch) | |
tree | ebfc3b77a636822d4d5805553e59c2770d6c3dc2 /lib | |
parent | ab1671cafefcaf19ff6ac00a6cc7b36dfe16adcd (diff) | |
download | curl-9a8564a920188e49d5bd8c1c8573ddef97f6e03a.tar.gz |
urlapi: URL decode percent-encoded host names
The host name is stored decoded and can be encoded when used to extract
the full URL. By default when extracting the URL, the host name will not
be URL encoded to work as similar as possible as before. When not URL
encoding the host name, the '%' character will however still be encoded.
Getting the URL with the CURLU_URLENCODE flag set will percent encode
the host name part.
As a bonus, setting the host name part with curl_url_set() no longer
accepts a name that contains space, CR or LF.
Test 1560 has been extended to verify percent encodings.
Reported-by: Noam Moshe
Reported-by: Sharon Brizinov
Reported-by: Raul Onitza-Klugman
Reported-by: Kirill Efimov
Fixes #7830
Closes #7834
Diffstat (limited to 'lib')
-rw-r--r-- | lib/urlapi.c | 109 |
1 files changed, 90 insertions, 19 deletions
diff --git a/lib/urlapi.c b/lib/urlapi.c index 367ddca79..5c943c52c 100644 --- a/lib/urlapi.c +++ b/lib/urlapi.c @@ -654,7 +654,7 @@ static CURLUcode hostname_check(struct Curl_URL *u, char *hostname) } else { /* letters from the second string is not ok */ - len = strcspn(hostname, " "); + len = strcspn(hostname, " \r\n"); if(hlen != len) /* hostname with bad content */ return CURLUE_MALFORMED_INPUT; @@ -754,6 +754,30 @@ static bool ipv4_normalize(const char *hostname, char *outp, size_t olen) return TRUE; } +/* return strdup'ed version in 'outp', possibly percent decoded */ +static CURLUcode decode_host(char *hostname, char **outp) +{ + char *per = NULL; + if(hostname[0] != '[') + /* only decode if not an ipv6 numerical */ + per = strchr(hostname, '%'); + if(!per) { + *outp = strdup(hostname); + if(!*outp) + return CURLUE_OUT_OF_MEMORY; + } + else { + /* might be encoded */ + size_t dlen; + CURLcode result = Curl_urldecode(NULL, hostname, 0, + outp, &dlen, REJECT_CTRL); + if(result) + return CURLUE_MALFORMED_INPUT; + } + + return CURLUE_OK; +} + static CURLUcode seturl(const char *url, CURLU *u, unsigned int flags) { char *path; @@ -1029,20 +1053,22 @@ static CURLUcode seturl(const char *url, CURLU *u, unsigned int flags) if(0 == strlen(hostname) && (flags & CURLU_NO_AUTHORITY)) { /* Skip hostname check, it's allowed to be empty. */ + u->host = strdup(""); } else { - result = hostname_check(u, hostname); - if(result) - return result; + if(ipv4_normalize(hostname, normalized_ipv4, sizeof(normalized_ipv4))) + u->host = strdup(normalized_ipv4); + else { + result = decode_host(hostname, &u->host); + if(result) + return result; + result = hostname_check(u, u->host); + if(result) + return result; + } } - - if(ipv4_normalize(hostname, normalized_ipv4, sizeof(normalized_ipv4))) - u->host = strdup(normalized_ipv4); - else - u->host = strdup(hostname); if(!u->host) return CURLUE_OUT_OF_MEMORY; - if((flags & CURLU_GUESS_SCHEME) && !schemep) { /* legacy curl-style guess based on host name */ if(checkprefix("ftp.", hostname)) @@ -1137,6 +1163,7 @@ CURLUcode curl_url_get(CURLU *u, CURLUPart what, CURLUcode ifmissing = CURLUE_UNKNOWN_PART; char portbuf[7]; bool urldecode = (flags & CURLU_URLDECODE)?1:0; + bool urlencode = (flags & CURLU_URLENCODE)?1:0; bool plusdecode = FALSE; (void)flags; if(!u) @@ -1254,16 +1281,55 @@ CURLUcode curl_url_get(CURLU *u, CURLUPart what, if(h && !(h->flags & PROTOPT_URLOPTIONS)) options = NULL; - if((u->host[0] == '[') && u->zoneid) { - /* make it '[ host %25 zoneid ]' */ - size_t hostlen = strlen(u->host); - size_t alen = hostlen + 3 + strlen(u->zoneid) + 1; - allochost = malloc(alen); + if(u->host[0] == '[') { + if(u->zoneid) { + /* make it '[ host %25 zoneid ]' */ + size_t hostlen = strlen(u->host); + size_t alen = hostlen + 3 + strlen(u->zoneid) + 1; + allochost = malloc(alen); + if(!allochost) + return CURLUE_OUT_OF_MEMORY; + memcpy(allochost, u->host, hostlen - 1); + msnprintf(&allochost[hostlen - 1], alen - hostlen + 1, + "%%25%s]", u->zoneid); + } + } + else if(urlencode) { + int hostlen = (int)strlen(u->host); + allochost = curl_easy_escape(NULL, u->host, hostlen); if(!allochost) return CURLUE_OUT_OF_MEMORY; - memcpy(allochost, u->host, hostlen - 1); - msnprintf(&allochost[hostlen - 1], alen - hostlen + 1, - "%%25%s]", u->zoneid); + } + else { + /* only encode '%' in output host name */ + char *host = u->host; + size_t pcount = 0; + /* first, count number of percents present in the name */ + while(*host) { + if(*host == '%') + pcount++; + host++; + } + /* if there were percents, encode the host name */ + if(pcount) { + size_t hostlen = strlen(u->host); + size_t alen = hostlen + 2 * pcount + 1; + char *o = allochost = malloc(alen); + if(!allochost) + return CURLUE_OUT_OF_MEMORY; + + host = u->host; + while(*host) { + if(*host == '%') { + memcpy(o, "%25", 3); + o += 3; + host++; + continue; + } + *o++ = *host++; + } + *o = '\0'; + } } url = aprintf("%s://%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s", @@ -1405,10 +1471,15 @@ CURLUcode curl_url_set(CURLU *u, CURLUPart what, case CURLUPART_OPTIONS: storep = &u->options; break; - case CURLUPART_HOST: + case CURLUPART_HOST: { + size_t len = strcspn(part, " \r\n"); + if(strlen(part) != len) + /* hostname with bad content */ + return CURLUE_MALFORMED_INPUT; storep = &u->host; Curl_safefree(u->zoneid); break; + } case CURLUPART_ZONEID: storep = &u->zoneid; break; |