git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH] handle http urls with query string ("?foo") correctly

From
David ‘Bombe’ Roden <bombe@pterodactylus.net>
Date
Jun 4, 2008, 23:28 UTC
Message-ID
<200806050128.33467.bombe@pterodactylus.net>
From 1f55b9176248673b78c77ff292fb5c6c988a7f8b Mon Sep 17 00:00:00 2001
From: =?utf-8?q?David=20=E2=80=98Bombe=E2=80=99=20Roden?= <bombe@pterodactylus.net>
Date: Thu, 5 Jun 2008 00:53:56 +0200
Subject: [PATCH] handle http urls with query string ("?foo") correctly

--- I'm a developer for the Freenet Project[1] and tried to get git running with Freenet's http-gateway. For several reasons I have to use URLs like "http://host:8888/USK@<lots-of-stuff>/project.git/4?type=text/plain". This breaks the way git creates http URLs to retrieve. The following patch remedys this situation by inserting the git-generated part of the URL (like "/info/refs") before the query string.

 http-walker.c |   31 ++++++++++++-------------------
 http.c        |   43 ++++++++++++++++++++++++++++++++++---------
 http.h        |    2 ++
 transport.c   |    5 +++--
 4 files changed, 51 insertions(+), 30 deletions(-)
diff --git a/http-walker.c b/http-walker.c
index 99f397e..711e55b 100644
--- a/http-walker.c
+++ b/http-walker.c
@@ -100,7 +100,6 @@ static void start_object_request(struct walker *walker,
 	char *hex = sha1_to_hex(obj_req->sha1);
 	char prevfile[PATH_MAX];
 	char *url;
-	char *posn;
 	int prevlocal;
 	unsigned char prev_buf[PREV_BUF_SIZE];
 	ssize_t prev_read = 0;
@@ -146,16 +145,11 @@ static void start_object_request(struct walker *walker,
 
 	SHA1_Init(&obj_req->c);
 
-	url = xmalloc(strlen(obj_req->repo->base) + 51);
+	char *suffix = xmalloc(51);
 	obj_req->url = xmalloc(strlen(obj_req->repo->base) + 51);
-	strcpy(url, obj_req->repo->base);
-	posn = url + strlen(obj_req->repo->base);
-	strcpy(posn, "/objects/");
-	posn += 9;
-	memcpy(posn, hex, 2);
-	posn += 2;
-	*(posn++) = '/';
-	strcpy(posn, hex + 2);
+	strcpy(suffix, "/objects/");
+	sprintf(suffix + 9, "%c%c/%s", hex[0], hex[1], hex + 2);
+	url = transform_url(obj_req->repo->base, suffix);
 	strcpy(obj_req->url, url);
 
 	/* If a previous temp file is present, process what was already
@@ -384,8 +378,9 @@ static int fetch_index(struct walker *walker, struct alt_base *repo, unsigned ch
 	if (walker->get_verbosely)
 		fprintf(stderr, "Getting index for pack %s\n", hex);
 
-	url = xmalloc(strlen(repo->base) + 64);
-	sprintf(url, "%s/objects/pack/pack-%s.idx", repo->base, hex);
+	char *index_url = xmalloc(64);
+	sprintf(index_url, "/objects/pack/pack-%s.idx", hex);
+	url = transform_url(repo->base, index_url);
 
 	filename = sha1_pack_index_name(sha1);
 	snprintf(tmpfile, sizeof(tmpfile), "%s.temp", filename);
@@ -608,8 +603,7 @@ static void fetch_alternates(struct walker *walker, const char *base)
 	if (walker->get_verbosely)
 		fprintf(stderr, "Getting alternates list for %s\n", base);
 
-	url = xmalloc(strlen(base) + 31);
-	sprintf(url, "%s/objects/info/http-alternates", base);
+	url = transform_url(base, "/objects/info/http-alternates");
 
 	/* Use a callback to process the result, since another request
 	   may fail and need to have alternates loaded before continuing */
@@ -655,8 +649,7 @@ static int fetch_indices(struct walker *walker, struct alt_base *repo)
 	if (walker->get_verbosely)
 		fprintf(stderr, "Getting pack list for %s\n", repo->base);
 
-	url = xmalloc(strlen(repo->base) + 21);
-	sprintf(url, "%s/objects/info/packs", repo->base);
+	url = transform_url(repo->base, "/objects/info/packs");
 
 	slot = get_active_slot();
 	slot->results = &results;
@@ -739,9 +732,9 @@ static int fetch_pack(struct walker *walker, struct alt_base *repo, unsigned cha
 			sha1_to_hex(sha1));
 	}
 
-	url = xmalloc(strlen(repo->base) + 65);
-	sprintf(url, "%s/objects/pack/pack-%s.pack",
-		repo->base, sha1_to_hex(target->sha1));
+	char *pack_url = xmalloc(65);
+	sprintf(pack_url, "/objects/pack/pack-%s.pack", sha1_to_hex(target->sha1));
+	url = transform_url(repo->base, pack_url);
 
 	filename = sha1_pack_name(target->sha1);
 	snprintf(tmpfile, sizeof(tmpfile), "%s.temp", filename);
diff --git a/http.c b/http.c
index 2a21ccb..e2b764a 100644
--- a/http.c
+++ b/http.c
@@ -590,15 +590,17 @@ static char *quote_ref_url(const char *base, const char *ref)
 	qref = xmalloc(len);
 	memcpy(qref, base, baselen);
 	dp = qref + baselen;
-	*(dp++) = '/';
-	for (cp = ref; (ch = *cp) != 0; cp++) {
-		if (needs_quote(ch)) {
-			*dp++ = '%';
-			*dp++ = hex((ch >> 4) & 0xF);
-			*dp++ = hex(ch & 0xF);
+	if (*ref) {
+		*(dp++) = '/';
+		for (cp = ref; (ch = *cp) != 0; cp++) {
+			if (needs_quote(ch)) {
+				*dp++ = '%';
+				*dp++ = hex((ch >> 4) & 0xF);
+				*dp++ = hex(ch & 0xF);
+			}
+			else
+				*dp++ = ch;
 		}
-		else
-			*dp++ = ch;
 	}
 	*dp = 0;
 
@@ -613,7 +615,10 @@ int http_fetch_ref(const char *base, struct ref *ref)
 	struct slot_results results;
 	int ret;
 
-	url = quote_ref_url(base, ref->name);
+	char *real_ref_name = xmalloc(strlen(ref->name) + 2);
+	sprintf(real_ref_name, "/%s", ref->name);
+	char *real_url = transform_url(base, real_ref_name);
+	url = quote_ref_url(real_url, "");
 	slot = get_active_slot();
 	slot->results = &results;
 	curl_easy_setopt(slot->curl, CURLOPT_FILE, &buffer);
@@ -643,3 +648,23 @@ int http_fetch_ref(const char *base, struct ref *ref)
 	free(url);
 	return ret;
 }
+
+char *transform_url(const char *url, const char *suffix)
+{
+	char *new_url;
+	char *question_mark;
+	ptrdiff_t offset;
+
+	new_url = xmalloc(strlen(url) + strlen(suffix) + 1);
+
+	if ((question_mark = strchr(url, '?'))) {
+		offset = (ptrdiff_t) question_mark - (ptrdiff_t) url;
+		snprintf(new_url, offset + 1, "%s", url);
+		strcat(new_url, suffix);
+		strcat(new_url, url + offset);
+	} else {
+		sprintf(new_url, "%s%s", url, suffix);
+	}
+	return new_url;
+}
+
diff --git a/http.h b/http.h
index a04fc6a..58730b8 100644
--- a/http.h
+++ b/http.h
@@ -107,4 +107,6 @@ static inline int missing__target(int code, int result)
 
 extern int http_fetch_ref(const char *base, struct ref *ref);
 
+extern char *transform_url(const char *url, const char *suffix);
+
 #endif /* HTTP_H */
diff --git a/transport.c b/transport.c
index 3ff8519..b1966d8 100644
--- a/transport.c
+++ b/transport.c
@@ -1,3 +1,4 @@
+#include <stddef.h>
 #include "cache.h"
 #include "transport.h"
 #include "run-command.h"
@@ -449,8 +450,7 @@ static struct ref *get_refs_via_curl(struct transport *transport)
 
 	walker = transport->data;
 
-	refs_url = xmalloc(strlen(transport->url) + 11);
-	sprintf(refs_url, "%s/info/refs", transport->url);
+	refs_url = transform_url(transport->url, "/info/refs");
 
 	slot = get_active_slot();
 	slot->results = &results;
@@ -833,3 +833,4 @@ int transport_disconnect(struct transport *transport)
 	free(transport);
 	return ret;
 }
+

-- 
1.5.6.rc1.6.gc8c52
Next: Johannes Schindelin
Message 1 of 8 in “handle http urls with query string ("?foo") correctly”
  1. handle http urls with query string ("?foo") correctlyDavid ‘Bombe’ Roden, Jun 4, 2008
  2. Johannes SchindelinJun 5, 2008
  3. David ‘Bombe’ RodenJun 5, 2008
  4. Johannes SchindelinJun 5, 2008
  5. Junio C HamanoJun 5, 2008
  6. David ‘Bombe’ RodenJun 5, 2008
  7. David ‘Bombe’ RodenJun 5, 2008
  8. David ‘Bombe’ RodenJun 5, 2008

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.