Verify Content-Type from smart HTTP servers

Before parsing a suspected smart-HTTP response verify the returned Content-Type matches the standard. This protects a client from attempting to process a payload that smells like a smart-HTTP server response. JGit has been doing this check on all responses since the dawn of time. I mistakenly failed to include it in git-core when smart HTTP was introduced. At the time I didn't know how to get the Content-Type from libcurl. I punted, meant to circle back and fix this, and just plain forgot about it. Signed-off-by: Shawn Pearce <spearce@spearce.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Shawn Pearce committed Jan 31, 2013 at 13:02 UTC 4656bf47fca857df51b5d6f4b7b052192b3b2317
8 files changed +59 -17
http-push.c
+2 -2
@@ -1560,7 +1560,7 @@ static int remote_exists(const char *path)
1560
1561 sprintf(url, "%s%s", repo->url, path);
1562
1563 - switch (http_get_strbuf(url, NULL, 0)) {
1563 + switch (http_get_strbuf(url, NULL, NULL, 0)) {
1564 case HTTP_OK:
1565 ret = 1;
1566 break;
@@ -1584,7 +1584,7 @@ static void fetch_symref(const char *path, char **symref, unsigned char *sha1)
1584 url = xmalloc(strlen(repo->url) + strlen(path) + 1);
1585 sprintf(url, "%s%s", repo->url, path);
1586
1587 - if (http_get_strbuf(url, &buffer, 0) != HTTP_OK)
1587 + if (http_get_strbuf(url, NULL, &buffer, 0) != HTTP_OK)
1588 die("Couldn't get %s for remote symref\n%s", url,
1589 curl_errorstr);
1590 free(url);
http.c
+22 -9
@@ -788,7 +788,8 @@ int handle_curl_result(struct slot_results *results)
788 #define HTTP_REQUEST_STRBUF 0
789 #define HTTP_REQUEST_FILE 1
790
791 -static int http_request(const char *url, void *result, int target, int options)
791 +static int http_request(const char *url, struct strbuf *type,
792 + void *result, int target, int options)
793 {
794 struct active_request_slot *slot;
795 struct slot_results results;
@@ -838,24 +839,36 @@ static int http_request(const char *url, void *result, int target, int options)
839 ret = HTTP_START_FAILED;
840 }
841
842 + if (type) {
843 + char *t;
844 + curl_easy_getinfo(slot->curl, CURLINFO_CONTENT_TYPE, &t);
845 + if (t)
846 + strbuf_addstr(type, t);
847 + }
848 +
849 curl_slist_free_all(headers);
850 strbuf_release(&buf);
851
852 return ret;
853 }
854
847 -static int http_request_reauth(const char *url, void *result, int target,
855 +static int http_request_reauth(const char *url,
856 + struct strbuf *type,
857 + void *result, int target,
858 int options)
859 {
850 - int ret = http_request(url, result, target, options);
860 + int ret = http_request(url, type, result, target, options);
861 if (ret != HTTP_REAUTH)
862 return ret;
853 - return http_request(url, result, target, options);
863 + return http_request(url, type, result, target, options);
864 }
865
856 -int http_get_strbuf(const char *url, struct strbuf *result, int options)
866 +int http_get_strbuf(const char *url,
867 + struct strbuf *type,
868 + struct strbuf *result, int options)
869 {
858 - return http_request_reauth(url, result, HTTP_REQUEST_STRBUF, options);
870 + return http_request_reauth(url, type, result,
871 + HTTP_REQUEST_STRBUF, options);
872 }
873
874 /*
@@ -878,7 +891,7 @@ static int http_get_file(const char *url, const char *filename, int options)
891 goto cleanup;
892 }
893
881 - ret = http_request_reauth(url, result, HTTP_REQUEST_FILE, options);
894 + ret = http_request_reauth(url, NULL, result, HTTP_REQUEST_FILE, options);
895 fclose(result);
896
897 if ((ret == HTTP_OK) && move_temp_to_file(tmpfile.buf, filename))
@@ -904,7 +917,7 @@ int http_fetch_ref(const char *base, struct ref *ref)
917 int ret = -1;
918
919 url = quote_ref_url(base, ref->name);
907 - if (http_get_strbuf(url, &buffer, HTTP_NO_CACHE) == HTTP_OK) {
920 + if (http_get_strbuf(url, NULL, &buffer, HTTP_NO_CACHE) == HTTP_OK) {
921 strbuf_rtrim(&buffer);
922 if (buffer.len == 40)
923 ret = get_sha1_hex(buffer.buf, ref->old_sha1);
@@ -997,7 +1010,7 @@ int http_get_info_packs(const char *base_url, struct packed_git **packs_head)
1010 strbuf_addstr(&buf, "objects/info/packs");
1011 url = strbuf_detach(&buf, NULL);
1012
1000 - ret = http_get_strbuf(url, &buf, HTTP_NO_CACHE);
1013 + ret = http_get_strbuf(url, NULL, &buf, HTTP_NO_CACHE);
1014 if (ret != HTTP_OK)
1015 goto cleanup;
1016
http.h
+1 -1
@@ -132,7 +132,7 @@ extern char *get_remote_object_url(const char *url, const char *hex,
132 *
133 * If the result pointer is NULL, a HTTP HEAD request is made instead of GET.
134 */
135 -int http_get_strbuf(const char *url, struct strbuf *result, int options);
135 +int http_get_strbuf(const char *url, struct strbuf *content_type, struct strbuf *result, int options);
136
137 /*
138 * Prints an error message using error() containing url and curl_errorstr,
remote-curl.c
+12 -5
@@ -92,6 +92,8 @@ static void free_discovery(struct discovery *d)
92
93 static struct discovery* discover_refs(const char *service)
94 {
95 + struct strbuf exp = STRBUF_INIT;
96 + struct strbuf type = STRBUF_INIT;
97 struct strbuf buffer = STRBUF_INIT;
98 struct discovery *last = last_discovery;
99 char *refs_url;
@@ -113,7 +115,7 @@ static struct discovery* discover_refs(const char *service)
115 }
116 refs_url = strbuf_detach(&buffer, NULL);
117
116 - http_ret = http_get_strbuf(refs_url, &buffer, HTTP_NO_CACHE);
118 + http_ret = http_get_strbuf(refs_url, &type, &buffer, HTTP_NO_CACHE);
119 switch (http_ret) {
120 case HTTP_OK:
121 break;
@@ -132,17 +134,20 @@ static struct discovery* discover_refs(const char *service)
134 last->buf_alloc = strbuf_detach(&buffer, &last->len);
135 last->buf = last->buf_alloc;
136
135 - if (maybe_smart && 5 <= last->len && last->buf[4] == '#') {
136 - /* smart HTTP response; validate that the service
137 + strbuf_addf(&exp, "application/x-%s-advertisement", service);
138 + if (maybe_smart &&
139 + (5 <= last->len && last->buf[4] == '#') &&
140 + !strbuf_cmp(&exp, &type)) {
141 + /*
142 + * smart HTTP response; validate that the service
143 * pkt-line matches our request.
144 */
139 - struct strbuf exp = STRBUF_INIT;
140 -
145 if (packet_get_line(&buffer, &last->buf, &last->len) <= 0)
146 die("%s has invalid packet header", refs_url);
147 if (buffer.len && buffer.buf[buffer.len - 1] == '\n')
148 strbuf_setlen(&buffer, buffer.len - 1);
149
150 + strbuf_reset(&exp);
151 strbuf_addf(&exp, "# service=%s", service);
152 if (strbuf_cmp(&exp, &buffer))
153 die("invalid server response; got '%s'", buffer.buf);
@@ -160,6 +165,8 @@ static struct discovery* discover_refs(const char *service)
165 }
166
167 free(refs_url);
168 + strbuf_release(&exp);
169 + strbuf_release(&type);
170 strbuf_release(&buffer);
171 last_discovery = last;
172 return last;
t/lib-httpd.sh
+1
@@ -80,6 +80,7 @@ fi
80 prepare_httpd() {
81 mkdir -p "$HTTPD_DOCUMENT_ROOT_PATH"
82 cp "$TEST_PATH"/passwd "$HTTPD_ROOT_PATH"
83 + cp "$TEST_PATH"/broken-smart-http.sh "$HTTPD_ROOT_PATH"
84
85 ln -s "$LIB_HTTPD_MODULE_PATH" "$HTTPD_ROOT_PATH/modules"
86
t/lib-httpd/apache.conf
+4
@@ -62,9 +62,13 @@ Alias /auth/dumb/ www/auth/dumb/
62 SetEnv GIT_COMMITTER_EMAIL custom@example.com
63 </LocationMatch>
64 ScriptAliasMatch /smart_*[^/]*/(.*) ${GIT_EXEC_PATH}/git-http-backend/$1
65 +ScriptAlias /broken_smart/ broken-smart-http.sh/
66 <Directory ${GIT_EXEC_PATH}>
67 Options FollowSymlinks
68 </Directory>
69 +<Files broken-smart-http.sh>
70 + Options ExecCGI
71 +</Files>
72 <Files ${GIT_EXEC_PATH}/git-http-backend>
73 Options ExecCGI
74 </Files>
t/lib-httpd/broken-smart-http.sh new
+11
@@ -0,0 +1,11 @@
1 +#!/bin/sh
2 +printf "Content-Type: text/%s\n" "html"
3 +echo
4 +printf "%s\n" "001e# service=git-upload-pack"
5 +printf "%s" "0000"
6 +printf "%s%c%s%s\n" \
7 + "00a58681d9f286a48b08f37b3a095330da16689e3693 HEAD" \
8 + 0 \
9 + " include-tag multi_ack_detailed multi_ack ofs-delta" \
10 + " side-band side-band-64k thin-pack no-progress shallow no-done "
11 +printf "%s" "0000"
t/t5551-http-fetch.sh
+6
@@ -157,6 +157,12 @@ test_expect_success 'GIT_SMART_HTTP can disable smart http' '
157 test_must_fail git fetch)
158 '
159
160 +test_expect_success 'invalid Content-Type rejected' '
161 + echo "fatal: invalid content-type text/html" >expect
162 + test_must_fail git clone $HTTPD_URL/broken_smart/repo.git 2>actual
163 + test_cmp expect actual
164 +'
165 +
166 test -n "$GIT_TEST_LONG" && test_set_prereq EXPENSIVE
167
168 test_expect_success EXPENSIVE 'create 50,000 tags in the repo' '