All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Rast <trast@student.ethz.ch>
To: <git@vger.kernel.org>
Cc: Jeff King <peff@peff.net>, <avarab@gmail.com>,
	<jstpierre@mecheye.net>, "Shawn O. Pearce" <spearce@spearce.org>,
	Junio C Hamano <gitster@pobox.com>
Subject: [PATCH v2] Do not unquote + into ' ' in URLs
Date: Sat, 24 Jul 2010 16:49:04 +0200	[thread overview]
Message-ID: <ed2d311355fca478f97b82f8d955494509d6b9de.1279982471.git.trast@student.ethz.ch> (raw)
In-Reply-To: <201007240104.25341.trast@student.ethz.ch>

Since 9d2e942 (decode file:// and ssh:// URLs, 2010-05-23) the URL
logic unquotes escaped URLs.  For the %2B type of escape, this is
conformant with RFC 2396.  However, it also unquotes + into a space
character, which is only appropriate for the query strings in HTTP.
This notably broke fetching from the gtk+ repository.

We cannot just remove the corresponding code since the same
url_decode_internal() is also used by the HTTP backend to decode query
parameters.  Introduce a new argument that controls whether the +
decoding happens, and use it only in the (client-side) url_decode().

Reported-by: Jasper St. Pierre <jstpierre@mecheye.net>
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
---

I wrote:
> Junio C Hamano wrote:
> > 
> >   http-backend.c::get_info_refs()
> >    -> http-backend.c::get_parameter()
> >      -> http-backend.c::get_parameters()
> >        -> url.c::url_decode_parameter_value()
> >          -> url.c::url_decode_internal()
> 
> You're right, I forgot about those.  I imagine it would be one of two
> cases:
[...]
> Shawn, can you help with this?

The third case, of course, is:

* It only uses these functions for parameter decoding, which of course
  was correct to begin with.

So after hopefully drinking enough coffee, I made this one.  The catch
is that I'm not entirely clear whether *not* decoding the +
client-side anywhere in the URL is correct for http:// URLs?  If the
client decodes and re-encodes the URL, then the + would be turned into
a %2B on the re-encoding.  Then again maybe UI-facing URLs should
never have a query part at all?


 t/t5601-clone.sh |   10 ++++++++--
 url.c            |   11 ++++++-----
 2 files changed, 14 insertions(+), 7 deletions(-)

diff --git a/t/t5601-clone.sh b/t/t5601-clone.sh
index 8abb71a..4431dfd 100755
--- a/t/t5601-clone.sh
+++ b/t/t5601-clone.sh
@@ -178,8 +178,14 @@ test_expect_success 'clone respects global branch.autosetuprebase' '
 
 test_expect_success 'respect url-encoding of file://' '
 	git init x+y &&
-	test_must_fail git clone "file://$PWD/x+y" xy-url &&
-	git clone "file://$PWD/x%2By" xy-url
+	git clone "file://$PWD/x+y" xy-url-1 &&
+	git clone "file://$PWD/x%2By" xy-url-2
+'
+
+test_expect_success 'do not query-string-decode + in URLs' '
+	rm -rf x+y &&
+	git init "x y" &&
+	test_must_fail git clone "file://$PWD/x+y" xy-no-plus
 '
 
 test_expect_success 'do not respect url-encoding of non-url path' '
diff --git a/url.c b/url.c
index 2306236..cd8f74f 100644
--- a/url.c
+++ b/url.c
@@ -67,7 +67,8 @@ static int url_decode_char(const char *q)
 	return val;
 }
 
-static char *url_decode_internal(const char **query, const char *stop_at, struct strbuf *out)
+static char *url_decode_internal(const char **query, const char *stop_at,
+				 struct strbuf *out, int decode_plus)
 {
 	const char *q = *query;
 
@@ -90,7 +91,7 @@ static char *url_decode_internal(const char **query, const char *stop_at, struct
 			}
 		}
 
-		if (c == '+')
+		if (decode_plus && c == '+')
 			strbuf_addch(out, ' ');
 		else
 			strbuf_addch(out, c);
@@ -110,17 +111,17 @@ char *url_decode(const char *url)
 		strbuf_add(&out, url, colon - url);
 		url = colon;
 	}
-	return url_decode_internal(&url, NULL, &out);
+	return url_decode_internal(&url, NULL, &out, 0);
 }
 
 char *url_decode_parameter_name(const char **query)
 {
 	struct strbuf out = STRBUF_INIT;
-	return url_decode_internal(query, "&=", &out);
+	return url_decode_internal(query, "&=", &out, 1);
 }
 
 char *url_decode_parameter_value(const char **query)
 {
 	struct strbuf out = STRBUF_INIT;
-	return url_decode_internal(query, "&", &out);
+	return url_decode_internal(query, "&", &out, 1);
 }
-- 
1.7.2.278.g76edd.dirty

  reply	other threads:[~2010-07-24 14:49 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-23 13:18 URL decoding changed semantics of + in URLs Thomas Rast
2010-07-23 13:21 ` Thomas Rast
2010-07-23 14:10 ` Ævar Arnfjörð Bjarmason
2010-07-23 14:25   ` Jasper St. Pierre
2010-07-23 21:23   ` [PATCH] Do not unquote + into ' ' " Thomas Rast
2010-07-23 22:20     ` Ævar Arnfjörð Bjarmason
2010-07-23 22:26     ` Junio C Hamano
2010-07-23 23:04       ` Thomas Rast
2010-07-24 14:49         ` Thomas Rast [this message]
2010-07-31 21:18           ` [PATCH v2] " Jasper St. Pierre
2010-07-31 21:33             ` Thomas Rast
2010-08-06 10:46               ` Ralf Ebert
2010-07-26 15:40 ` URL decoding changed semantics of + " Jeff King
2010-07-26 17:57   ` Ævar Arnfjörð Bjarmason
2010-07-26 18:22     ` Jasper St. Pierre
2010-07-26 18:30       ` Matthieu Moy
2010-07-26 18:35       ` Ævar Arnfjörð Bjarmason
2010-07-26 18:44         ` Jasper St. Pierre

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ed2d311355fca478f97b82f8d955494509d6b9de.1279982471.git.trast@student.ethz.ch \
    --to=trast@student.ethz.ch \
    --cc=avarab@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jstpierre@mecheye.net \
    --cc=peff@peff.net \
    --cc=spearce@spearce.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.