* Re: [PATCH/resend] CVS Server: Support reading base and roots from environment
From: Phil Miller @ 2009-12-30 20:12 UTC (permalink / raw)
To: Junio C Hamano; +Cc: Nanako Shiraishi, Git Mailing List
In-Reply-To: <7vpr5wz1hb.fsf@alter.siamese.dyndns.org>
The Gitosis single-account Git/ssh hosting system runs git commands
through git-shell after confirming that the connecting user is
authorized to access the requested repository. This works well for
upload-pack and receive-pack, which take a repository argument through
git-shell. This doesn't work so well for `cvs server', which is passed
through literally, with no arguments. Allowing arguments risks
sneaking in `--export-all', so that restriction should be maintained.
Despite that, passing a repository root is necessary for per-user
access control by the hosting software, and passing a base path
improves usability without weakening security. Thus, git-cvsserver
needs to come up with these values at runtime by some other
means. Since git-shell preserves the environment for other purposes,
the environment can carry these arguments as well.
Thus, modify git-cvsserver to read $GIT_CVSSERVER_{BASE_PATH,ROOT} in
the absence of equivalent command line arguments.
Signed-off-by: Phil Miller <mille121@illinois.edu>
---
I believe this revision addresses all of your comments on the first submission.
Your comment about cramming multiple values into one environment variable made
me realize that more than one simply was unnecessary complexity, since gitosis
expects to authenticate access to a single repository anyway.
I've not documented what GIT_CVSSERVER_BASE_PATH is relative to, because it
behaves identically to the --base-path command line argument. Documenting
what that is relative to is a separate issue.
Documentation/git-cvsserver.txt | 15 +++++++++++++++
git-cvsserver.perl | 22 +++++++++++++++++++++-
2 files changed, 36 insertions(+), 1 deletions(-)
diff --git a/Documentation/git-cvsserver.txt b/Documentation/git-cvsserver.txt
index 99a7c14..fbab295 100644
--- a/Documentation/git-cvsserver.txt
+++ b/Documentation/git-cvsserver.txt
@@ -277,6 +277,21 @@ In `dbdriver` and `dbuser` you can use the
following variables:
If no name can be determined, the
numeric uid is used.
+ENVIRONMENT
+-----------
+
+These variables obviate the need for command-line options in some
+circumstances, allowing easier restricted usage through git-shell.
+
+GIT_CVSSERVER_BASE_PATH takes the place of the argument to --base-path.
+
+GIT_CVSSERVER_ROOT specifies a single-directory whitelist. The
+repository must still be configured to allow access through
+git-cvsserver, as described above.
+
+When these environment variables are set, the corresponding
+command-line arguments may not be used.
+
Eclipse CVS Client Notes
------------------------
diff --git a/git-cvsserver.perl b/git-cvsserver.perl
index 6dc45f5..f5b57b9 100755
--- a/git-cvsserver.perl
+++ b/git-cvsserver.perl
@@ -104,6 +104,7 @@ $log->info("--------------- STARTING -----------------");
my $usage =
"Usage: git cvsserver [options] [pserver|server] [<directory> ...]\n".
" --base-path <path> : Prepend to requested CVSROOT\n".
+ " Can be read from GIT_CVSSERVER_BASE_PATH\n".
" --strict-paths : Don't allow recursing into subdirectories\n".
" --export-all : Don't check for gitcvs.enabled in config\n".
" --version, -V : Print version information and exit\n".
@@ -111,7 +112,8 @@ my $usage =
"\n".
"<directory> ... is a list of allowed directories. If no directories\n".
"are given, all are allowed. This is an additional restriction, gitcvs\n".
- "access still needs to be enabled by the gitcvs.enabled config option.\n";
+ "access still needs to be enabled by the gitcvs.enabled config option.\n".
+ "Alternately, one directory may be specified in GIT_CVSSERVER_ROOT.\n";
my @opts = ( 'help|h|H', 'version|V',
'base-path=s', 'strict-paths', 'export-all' );
@@ -148,6 +150,24 @@ if ($state->{'export-all'} &&
!@{$state->{allowed_roots}}) {
die "--export-all can only be used together with an explicit whitelist\n";
}
+# Environment handling for running under git-shell
+if (exists $ENV{GIT_CVSSERVER_BASE_PATH}) {
+ if ($state->{'base-path'}) {
+ die "Cannot specify base path both ways.\n";
+ }
+ my $base_path = $ENV{GIT_CVSSERVER_BASE_PATH};
+ $state->{'base-path'} = $base_path;
+ $log->debug("Picked up base path '$base_path' from environment.\n");
+}
+if (exists $ENV{GIT_CVSSERVER_ROOT}) {
+ if (@{$state->{allowed_roots}}) {
+ die "Cannot specify roots both ways: @ARGV\n";
+ }
+ my $allowed_root = $ENV{GIT_CVSSERVER_ROOT};
+ $state->{allowed_roots} = [ $allowed_root ];
+ $log->debug("Picked up allowed root '$allowed_root' from environment.\n");
+}
+
# if we are called with a pserver argument,
# deal with the authentication cat before entering the
# main loop
--
debian.1.6.6_rc2.1.7.gc3ed7
^ permalink raw reply related
* [PATCH v4] builtin-push: add --delete as syntactic sugar for :foo
From: Jan Krüger @ 2009-12-30 19:57 UTC (permalink / raw)
To: Junio C Hamano; +Cc: Nanako Shiraishi, Git ML, Sverre Rabbelier
In-Reply-To: <7vvdfoz1hi.fsf@alter.siamese.dyndns.org>
Refspecs without a source side have been reported as confusing by many.
As an alternative, this adds support for commands like:
git push origin --delete somebranch
git push origin --delete tag sometag
Specifically, --delete will prepend a colon to all colon-less refspecs
given on the command line, and will refuse to accept refspecs with
colons to prevent undue confusion.
Signed-off-by: Jan Krüger <jk@jk.gs>
---
Documentation/git-push.txt | 4 ++++
builtin-push.c | 26 +++++++++++++++++++++++---
t/t5516-fetch-push.sh | 26 ++++++++++++++++++++++++++
3 files changed, 53 insertions(+), 3 deletions(-)
Junio C Hamano <gitster@pobox.com> wrote:
> Thanks. From a cursory read, the patch looks good. We would however
> want to have a test that has test_must_fail to protect the error
> codepath from getting broken in the future.
Here you are.
diff --git a/Documentation/git-push.txt b/Documentation/git-push.txt
index 52c0538..e3eb1e8 100644
--- a/Documentation/git-push.txt
+++ b/Documentation/git-push.txt
@@ -91,6 +91,10 @@ nor in any Push line of the corresponding remotes file---see below).
will be tab-separated and sent to stdout instead of stderr. The full
symbolic names of the refs will be given.
+--delete::
+ All listed refs are deleted from the remote repository. This is
+ the same as prefixing all refs with a colon.
+
--tags::
All refs under `$GIT_DIR/refs/tags` are pushed, in
addition to refspecs explicitly listed on the command
diff --git a/builtin-push.c b/builtin-push.c
index dcfb53f..f7661d2 100644
--- a/builtin-push.c
+++ b/builtin-push.c
@@ -15,6 +15,7 @@ static const char * const push_usage[] = {
};
static int thin;
+static int deleterefs;
static const char *receivepack;
static const char **refspec;
@@ -39,11 +40,24 @@ static void set_refspecs(const char **refs, int nr)
if (nr <= ++i)
die("tag shorthand without <tag>");
len = strlen(refs[i]) + 11;
- tag = xmalloc(len);
- strcpy(tag, "refs/tags/");
+ if (deleterefs) {
+ tag = xmalloc(len+1);
+ strcpy(tag, ":refs/tags/");
+ } else {
+ tag = xmalloc(len);
+ strcpy(tag, "refs/tags/");
+ }
strcat(tag, refs[i]);
ref = tag;
- }
+ } else if (deleterefs && !strchr(ref, ':')) {
+ char *delref;
+ int len = strlen(ref)+1;
+ delref = xmalloc(len);
+ strcpy(delref, ":");
+ strcat(delref, ref);
+ ref = delref;
+ } else if (deleterefs)
+ die("--delete only accepts plain target ref names");
add_refspec(ref);
}
}
@@ -196,6 +210,7 @@ int cmd_push(int argc, const char **argv, const char *prefix)
OPT_BIT( 0 , "all", &flags, "push all refs", TRANSPORT_PUSH_ALL),
OPT_BIT( 0 , "mirror", &flags, "mirror all refs",
(TRANSPORT_PUSH_MIRROR|TRANSPORT_PUSH_FORCE)),
+ OPT_BOOLEAN( 0, "delete", &deleterefs, "delete refs"),
OPT_BOOLEAN( 0 , "tags", &tags, "push tags (can't be used with --all or --mirror)"),
OPT_BIT('n' , "dry-run", &flags, "dry run", TRANSPORT_PUSH_DRY_RUN),
OPT_BIT( 0, "porcelain", &flags, "machine-readable output", TRANSPORT_PUSH_PORCELAIN),
@@ -209,6 +224,11 @@ int cmd_push(int argc, const char **argv, const char *prefix)
git_config(git_default_config, NULL);
argc = parse_options(argc, argv, prefix, options, push_usage, 0);
+ if (deleterefs && (tags || (flags & (TRANSPORT_PUSH_ALL | TRANSPORT_PUSH_MIRROR))))
+ die("--delete is incompatible with --all, --mirror and --tags");
+ if (deleterefs && argc < 2)
+ die("--delete doesn't make sense without any refs");
+
if (tags)
add_refspec("refs/tags/*");
diff --git a/t/t5516-fetch-push.sh b/t/t5516-fetch-push.sh
index 516127b..0f04b2e 100755
--- a/t/t5516-fetch-push.sh
+++ b/t/t5516-fetch-push.sh
@@ -547,6 +547,32 @@ test_expect_success 'allow deleting an invalid remote ref' '
'
+test_expect_success 'allow deleting a ref using --delete' '
+ mk_test heads/master &&
+ (cd testrepo && git config receive.denyDeleteCurrent warn) &&
+ git push testrepo --delete master &&
+ (cd testrepo && test_must_fail git rev-parse --verify refs/heads/master)
+'
+
+test_expect_success 'allow deleting a tag using --delete' '
+ mk_test heads/master &&
+ git tag -a -m dummy_message deltag heads/master &&
+ git push testrepo --tags &&
+ (cd testrepo && git rev-parse --verify -q refs/tags/deltag) &&
+ git push testrepo --delete tag deltag &&
+ (cd testrepo && test_must_fail git rev-parse --verify refs/tags/deltag)
+'
+
+test_expect_success 'push --delete without args aborts' '
+ mk_test heads/master &&
+ test_must_fail git push testrepo --delete
+'
+
+test_expect_success 'push --delete refuses src:dest refspecs' '
+ mk_test heads/master &&
+ test_must_fail git push testrepo --delete master:foo
+'
+
test_expect_success 'warn on push to HEAD of non-bare repository' '
mk_test heads/master
(cd testrepo &&
--
1.6.6.60.gc2ff1
^ permalink raw reply related
* Re: [PATCH v2] Let core.excludesfile default to ~/.gitexcludes.
From: Junio C Hamano @ 2009-12-30 19:49 UTC (permalink / raw)
To: Matthieu Moy; +Cc: Nanako Shiraishi, git
In-Reply-To: <vpqhbr8ttwv.fsf@bauges.imag.fr>
Matthieu Moy <Matthieu.Moy@grenoble-inp.fr> writes:
> Nanako Shiraishi <nanako3@lavabit.com> writes:
>
>> Junio, could you tell us what happened to this thread?
>
> I'm not Junio, but I can try ...
> ...
> So, I dropped the patch until we get closer to a consensus on what the
> default value should be.
Thanks for status updates. Should we start trying to reach a consensus on
how to move forward, or is it not your itch?
It is not _too_ bad to treat ~/.gitconfig specially and support reading
from ~/.$SOMEGITTTYNAME/{excludes,attributes} files. We can also add
support for ~/.$SOMEGITTYNAME/config and try reading from it as well, but
even if we did so I don't think we should stop reading from ~/.gitconfig.
I don't have strong preference myself either way.
^ permalink raw reply
* Re: [PATCH/resend] CVS Server: Support reading base and roots from environment
From: Junio C Hamano @ 2009-12-30 19:49 UTC (permalink / raw)
To: Phil Miller; +Cc: Nanako Shiraishi, Git Mailing List
In-Reply-To: <81f018ac0912300912g25887523g1bdf29fd6e31d011@mail.gmail.com>
Phil Miller <mille121@illinois.edu> writes:
> I never resent with modifications as Junio requested. I've been busy
> with some end-of-year close-out stuff, and actually wanted to
> dramatically simplify part of the patch before the resend. I'll do
> that shortly.
> ...
> The documentation hasn't been updated, but the recent history makes as
> least some usage clear. I'll send more details in a follow-up.
Thanks Phil for status updates (and thanks Nana for prodding).
^ permalink raw reply
* Re: [PATCH v3] builtin-push: add --delete as syntactic sugar for :foo
From: Junio C Hamano @ 2009-12-30 19:49 UTC (permalink / raw)
To: Jan Krüger; +Cc: Nanako Shiraishi, Git ML, Sverre Rabbelier
In-Reply-To: <20091230105244.67f5969e@perceptron>
Jan Krüger <jk@jk.gs> writes:
> Refspecs without a source side have been reported as confusing by many.
> As an alternative, this adds support for commands like:
>
> git push origin --delete somebranch
> git push origin --delete tag sometag
>
> Specifically, --delete will prepend a colon to all colon-less refspecs
> given on the command line, and will refuse to accept refspecs with
> colons to prevent undue confusion.
>
> Signed-off-by: Jan Krüger <jk@jk.gs>
> ---
Thanks. From a cursory read, the patch looks good. We would however want
to have a test that has test_must_fail to protect the error codepath from
getting broken in the future.
^ permalink raw reply
* Re: [PATCH] git-update-index: report(...) now flushes stdout after printing the report line
From: Junio C Hamano @ 2009-12-30 19:46 UTC (permalink / raw)
To: Nanako Shiraishi; +Cc: Sebastian Thiel, git
In-Reply-To: <20091230224122.6117@nanako3.lavabit.com>
Nanako Shiraishi <nanako3@lavabit.com> writes:
> Junio, could you tell us what happened to this thread?
I didn't feel I had enough energy to read the commit log message after
seeing it was badly linewrapped and didn't have a sign-off, so I didn't
read it.
I've read it now; it is unclear from the proposed commit log message how
this fits in the larger picture. Presumably this change is meant to be
useful when driving update-index through --stdin? To see if I got the
intention right, let me try paraphrasing it...
update-index: flush standard output after each action is reported
A scripted Porcelain that runs "git update-index --stdin" might want
to use a bidirectional pipe, while feeding one path at a time and
reading the output from report() every time after feeding a path.
Such a Porcelain would deadlock if the standard output is not flushed
after report().
I don't know if the above is what Sebastian meant, though..
An obvious question, when phrased this way, is "what impact does this
change have for scripted Porcelains that don't use bi-di pipe?" I think
the answer would be "The I/O overhead for flushing would increase", but I
don't know if it would be "... would increase but it is still negligible"
or "... would increase too much to make it noticeably or unusably slow
especially if it feeds hundreds of paths". If it is the latter, this may
need to be controlled by another command line option.
Sebastian, care to redo the justification, make it a bit more readable,
and add your sign-off?
Thanks.
^ permalink raw reply
* [PATCH/resend] CVS Server: Support reading base and roots from environment
From: Phil Miller @ 2009-12-30 19:35 UTC (permalink / raw)
To: Junio C Hamano; +Cc: Git Mailing List, Nanako Shiraishi
In-Reply-To: <7vocmwvmvr.fsf@alter.siamese.dyndns.org>
The Gitosis single-account Git/ssh hosting system runs git commands
through git-shell after confirming that the connecting user is
authorized to access the requested repository. This works well for
upload-pack and receive-pack, which take a repository argument through
git-shell. This doesn't work so well for `cvs server', which is passed
through literally, with no arguments. Allowing arguments risks
sneaking in `--export-all', so that restriction should be maintained.
Despite that, passing a repository root is necessary for per-user
access control by the hosting software, and passing a base path
improves usability without weakening security. Thus, git-cvsserver
needs to come up with these values at runtime by some other
means. Since git-shell preserves the environment for other purposes,
the environment can carry these arguments as well.
Thus, modify git-cvsserver to read $GIT_CVSSERVER_{BASE_PATH,ROOT} in
the absence of equivalent command line arguments.
Signed-off-by: Phil Miller <mille121@illinois.edu>
---
I believe this revision addresses all of your comments on the first submission.
Your comment about cramming multiple values into one environment variable made
me realize that more than one simply was unnecessary complexity, since gitosis
expects to authenticate access to a single repository anyway.
I've not documented what GIT_CVSSERVER_BASE_PATH is relative to, because it
behaves identically to the --base-path command line argument. Documenting
what that is relative to is a separate issue.
Documentation/git-cvsserver.txt | 15 +++++++++++++++
git-cvsserver.perl | 22 +++++++++++++++++++++-
2 files changed, 36 insertions(+), 1 deletions(-)
diff --git a/Documentation/git-cvsserver.txt b/Documentation/git-cvsserver.txt
index 99a7c14..fbab295 100644
--- a/Documentation/git-cvsserver.txt
+++ b/Documentation/git-cvsserver.txt
@@ -277,6 +277,21 @@ In `dbdriver` and `dbuser` you can use the following variables:
If no name can be determined, the
numeric uid is used.
+ENVIRONMENT
+-----------
+
+These variables obviate the need for command-line options in some
+circumstances, allowing easier restricted usage through git-shell.
+
+GIT_CVSSERVER_BASE_PATH takes the place of the argument to --base-path.
+
+GIT_CVSSERVER_ROOT specifies a single-directory whitelist. The
+repository must still be configured to allow access through
+git-cvsserver, as described above.
+
+When these environment variables are set, the corresponding
+command-line arguments may not be used.
+
Eclipse CVS Client Notes
------------------------
diff --git a/git-cvsserver.perl b/git-cvsserver.perl
index 6dc45f5..f5b57b9 100755
--- a/git-cvsserver.perl
+++ b/git-cvsserver.perl
@@ -104,6 +104,7 @@ $log->info("--------------- STARTING -----------------");
my $usage =
"Usage: git cvsserver [options] [pserver|server] [<directory> ...]\n".
" --base-path <path> : Prepend to requested CVSROOT\n".
+ " Can be read from GIT_CVSSERVER_BASE_PATH\n".
" --strict-paths : Don't allow recursing into subdirectories\n".
" --export-all : Don't check for gitcvs.enabled in config\n".
" --version, -V : Print version information and exit\n".
@@ -111,7 +112,8 @@ my $usage =
"\n".
"<directory> ... is a list of allowed directories. If no directories\n".
"are given, all are allowed. This is an additional restriction, gitcvs\n".
- "access still needs to be enabled by the gitcvs.enabled config option.\n";
+ "access still needs to be enabled by the gitcvs.enabled config option.\n".
+ "Alternately, one directory may be specified in GIT_CVSSERVER_ROOT.\n";
my @opts = ( 'help|h|H', 'version|V',
'base-path=s', 'strict-paths', 'export-all' );
@@ -148,6 +150,24 @@ if ($state->{'export-all'} && !@{$state->{allowed_roots}}) {
die "--export-all can only be used together with an explicit whitelist\n";
}
+# Environment handling for running under git-shell
+if (exists $ENV{GIT_CVSSERVER_BASE_PATH}) {
+ if ($state->{'base-path'}) {
+ die "Cannot specify base path both ways.\n";
+ }
+ my $base_path = $ENV{GIT_CVSSERVER_BASE_PATH};
+ $state->{'base-path'} = $base_path;
+ $log->debug("Picked up base path '$base_path' from environment.\n");
+}
+if (exists $ENV{GIT_CVSSERVER_ROOT}) {
+ if (@{$state->{allowed_roots}}) {
+ die "Cannot specify roots both ways: @ARGV\n";
+ }
+ my $allowed_root = $ENV{GIT_CVSSERVER_ROOT};
+ $state->{allowed_roots} = [ $allowed_root ];
+ $log->debug("Picked up allowed root '$allowed_root' from environment.\n");
+}
+
# if we are called with a pserver argument,
# deal with the authentication cat before entering the
# main loop
--
debian.1.6.6_rc2.1.7.gc3ed7
^ permalink raw reply related
* Re: [PATCH] bash completion: add space between branch name and status flags
From: Junio C Hamano @ 2009-12-30 19:27 UTC (permalink / raw)
To: Shawn O. Pearce; +Cc: Nanako Shiraishi, git, Roman Fietze
In-Reply-To: <20091230145751.GE6914@spearce.org>
"Shawn O. Pearce" <spearce@spearce.org> writes:
> Nanako Shiraishi <nanako3@lavabit.com> wrote:
>> Junio, could you tell us what happened to this thread?
>>
>> In response to a patch from Roman Fietze to outline a better way
>> to do it. Nothing happened.
>
> Junio responded with a suggestion on how to improve the patch when
> GIT_PS1_SHOWDIRTYSTATE was not set, but Roman Fietze didn't send
> a revised patch, so it got dropped.
>
> Here is the revised patch, Junio, still think its a good idea?
Thanks for following up.
As I don't use $[wisu] myself, I don't have a very strong opinion either
way, but the user has spent cycles to compute them, so we'd better present
them in a way that is easier to read, I guess.
> diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash
> index fbfa5f2..3c8b6df 100755
> --- a/contrib/completion/git-completion.bash
> +++ b/contrib/completion/git-completion.bash
> @@ -163,10 +163,12 @@ __git_ps1 ()
> fi
> fi
>
> + local f="$w$i$s$u"
> + f="${f:+ $f}$r"
> if [ -n "${1-}" ]; then
> - printf "$1" "$c${b##refs/heads/}$w$i$s$u$r"
> + printf "$1" "$c${b##refs/heads/}$f"
> else
> - printf " (%s)" "$c${b##refs/heads/}$w$i$s$u$r"
> + printf " (%s)" "$c${b##refs/heads/}$f"
> fi
I notice that printf argument look very similar. Maybe we want to do
something like
printf "${1:-" (%s)"}" ...
to avoid duplication?
^ permalink raw reply
* [PATCH RFC 2/2] Smart-http tests: Test http-backend without curl or a webserver
From: Tarmigan Casebolt @ 2009-12-30 18:59 UTC (permalink / raw)
To: git; +Cc: spearce, gitster, Tarmigan Casebolt
In-Reply-To: <1262199542-73876-1-git-send-email-tarmigan+git@gmail.com>
This reuses many of the tests from t5560 but runs those tests without curl
or a webserver. This will hopefully increase the testing coverage for
http-backend because it does not require users to set GIT_TEST_HTTPD.
Signed-off-by: Tarmigan Casebolt <tarmigan+git@gmail.com>
---
In addition to the rebase, I fixed a bunch of bugs in this version. I'm
still leaving this as RFC as there may be some more.
t/t5561-http-backend-noserver.sh | 49 ++++++++++++++++++++++---------------
t/t556x_common | 3 ++
2 files changed, 32 insertions(+), 20 deletions(-)
diff --git a/t/t5561-http-backend-noserver.sh b/t/t5561-http-backend-noserver.sh
index 501b328..bda5029 100755
--- a/t/t5561-http-backend-noserver.sh
+++ b/t/t5561-http-backend-noserver.sh
@@ -5,40 +5,49 @@ test_description='test git-http-backend-noserver'
HTTPD_DOCUMENT_ROOT_PATH="$TRASH_DIRECTORY"
+run_backend() {
+ echo "$3"| \
+ QUERY_STRING="${2#*\?}" \
+ GIT_PROJECT_ROOT="$HTTPD_DOCUMENT_ROOT_PATH" \
+ PATH_INFO="${2%%\?*}" \
+ git http-backend >act.out 2>act.err
+}
+
GET() {
- return 0
+ REQUEST_METHOD="GET" \
+ run_backend 0 "/repo.git/$1" &&
+ grep "Status" act.out >act
+ if [ $? -eq 1 ];
+ then
+ printf "Status: 200 OK\r\n" > act
+ fi
+ printf "Status: $2\r\n" > exp &&
+ test_cmp exp act
}
POST() {
- return 0
+ REQUEST_METHOD="POST" \
+ CONTENT_TYPE="application/x-$1-request" \
+ run_backend 0 "/repo.git/$1" "$2" &&
+ grep "Status" act.out >act
+ if [ $? -eq 1 ];
+ then
+ printf "Status: 200 OK\r\n" > act
+ fi
+ printf "Status: $3\r\n" > exp &&
+ test_cmp exp act
}
-logdiv() {
+log_div() {
return 0
}
. "$TEST_DIRECTORY"/t556x_common
-run_backend() {
- REQUEST_METHOD=GET \
- GIT_PROJECT_ROOT="$HTTPD_DOCUMENT_ROOT_PATH" \
- PATH_INFO="$2" \
- git http-backend >act.out 2>act.err
-}
-
-path_info() {
- if test $1 = 0; then
- run_backend "$2"
- else
- test_must_fail run_backend "$2" &&
- echo "fatal: '$2': aliased" >exp.err &&
- test_cmp exp.err act.err
- fi
-}
-
test_expect_success 'http-backend blocks bad PATH_INFO' '
config http.getanyfile true &&
+ REQUEST_METHOD="GET" &&
run_backend 0 /repo.git/HEAD &&
run_backend 1 /repo.git/../HEAD &&
diff --git a/t/t556x_common b/t/t556x_common
index 1b4921c..be024e5 100755
--- a/t/t556x_common
+++ b/t/t556x_common
@@ -50,6 +50,7 @@ get_static_files() {
}
SMART=smart
+export GIT_HTTP_EXPORT_ALL=1
test_expect_success 'direct refs/heads/master not found' '
log_div "refs/heads/master"
GET refs/heads/master "404 Not Found"
@@ -59,6 +60,7 @@ test_expect_success 'static file is ok' '
get_static_files "200 OK"
'
SMART=smart_noexport
+unset GIT_HTTP_EXPORT_ALL
test_expect_success 'no export by default' '
log_div "no git-daemon-export-ok"
get_static_files "404 Not Found"
@@ -71,6 +73,7 @@ test_expect_success 'export if git-daemon-export-ok' '
get_static_files "200 OK"
'
SMART=smart
+export GIT_HTTP_EXPORT_ALL=1
test_expect_success 'static file if http.getanyfile true is ok' '
log_div "getanyfile true"
config http.getanyfile true &&
--
1.6.6
^ permalink raw reply related
* [PATCH RFC 1/2] Smart-http tests: Break test t5560-http-backend into pieces
From: Tarmigan Casebolt @ 2009-12-30 18:59 UTC (permalink / raw)
To: git; +Cc: spearce, gitster, Tarmigan Casebolt
In-Reply-To: <905315640912301009x491f957al839f66de7aba56ed@mail.gmail.com>
This should introduce no functional change in the tests or the amount
of test coverage.
Signed-off-by: Tarmigan Casebolt <tarmigan+git@gmail.com>
---
This is rebased on top of today's pu and the conflicts should be fixed.
t/t5560-http-backend.sh | 148 +-------------------------------------
t/t5561-http-backend-noserver.sh | 52 +++++++++++++
t/t556x_common | 119 ++++++++++++++++++++++++++++++
3 files changed, 172 insertions(+), 147 deletions(-)
create mode 100755 t/t5561-http-backend-noserver.sh
create mode 100755 t/t556x_common
diff --git a/t/t5560-http-backend.sh b/t/t5560-http-backend.sh
index 04a9896..dd844d3 100755
--- a/t/t5560-http-backend.sh
+++ b/t/t5560-http-backend.sh
@@ -12,16 +12,6 @@ LIB_HTTPD_PORT=${LIB_HTTPD_PORT-'5560'}
. "$TEST_DIRECTORY"/lib-httpd.sh
start_httpd
-find_file() {
- cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
- find $1 -type f |
- sed -e 1q
-}
-
-config() {
- git --git-dir="$HTTPD_DOCUMENT_ROOT_PATH/repo.git" config $1 $2
-}
-
GET() {
curl --include "$HTTPD_URL/$SMART/repo.git/$1" >out 2>/dev/null &&
tr '\015' Q <out |
@@ -52,143 +42,7 @@ log_div() {
echo "###" >>"$HTTPD_ROOT_PATH"/access.log
}
-test_expect_success 'setup repository' '
- echo content >file &&
- git add file &&
- git commit -m one &&
-
- mkdir "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
- (cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
- git --bare init &&
- : >objects/info/alternates &&
- : >objects/info/http-alternates
- ) &&
- git remote add public "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
- git push public master:master &&
-
- (cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
- git repack -a -d
- ) &&
-
- echo other >file &&
- git add file &&
- git commit -m two &&
- git push public master:master &&
-
- LOOSE_URL=$(find_file objects/??) &&
- PACK_URL=$(find_file objects/pack/*.pack) &&
- IDX_URL=$(find_file objects/pack/*.idx)
-'
-
-get_static_files() {
- GET HEAD "$1" &&
- GET info/refs "$1" &&
- GET objects/info/packs "$1" &&
- GET objects/info/alternates "$1" &&
- GET objects/info/http-alternates "$1" &&
- GET $LOOSE_URL "$1" &&
- GET $PACK_URL "$1" &&
- GET $IDX_URL "$1"
-}
-
-SMART=smart
-test_expect_success 'direct refs/heads/master not found' '
- log_div "refs/heads/master"
- GET refs/heads/master "404 Not Found"
-'
-test_expect_success 'static file is ok' '
- log_div "getanyfile default"
- get_static_files "200 OK"
-'
-SMART=smart_noexport
-test_expect_success 'no export by default' '
- log_div "no git-daemon-export-ok"
- get_static_files "404 Not Found"
-'
-test_expect_success 'export if git-daemon-export-ok' '
- log_div "git-daemon-export-ok"
- (cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
- touch git-daemon-export-ok
- ) &&
- get_static_files "200 OK"
-'
-SMART=smart
-test_expect_success 'static file if http.getanyfile true is ok' '
- log_div "getanyfile true"
- config http.getanyfile true &&
- get_static_files "200 OK"
-'
-test_expect_success 'static file if http.getanyfile false fails' '
- log_div "getanyfile false"
- config http.getanyfile false &&
- get_static_files "403 Forbidden"
-'
-
-test_expect_success 'http.uploadpack default enabled' '
- log_div "uploadpack default"
- GET info/refs?service=git-upload-pack "200 OK" &&
- POST git-upload-pack 0000 "200 OK"
-'
-test_expect_success 'http.uploadpack true' '
- log_div "uploadpack true"
- config http.uploadpack true &&
- GET info/refs?service=git-upload-pack "200 OK" &&
- POST git-upload-pack 0000 "200 OK"
-'
-test_expect_success 'http.uploadpack false' '
- log_div "uploadpack false"
- config http.uploadpack false &&
- GET info/refs?service=git-upload-pack "403 Forbidden" &&
- POST git-upload-pack 0000 "403 Forbidden"
-'
-
-test_expect_success 'http.receivepack default disabled' '
- log_div "receivepack default"
- GET info/refs?service=git-receive-pack "403 Forbidden" &&
- POST git-receive-pack 0000 "403 Forbidden"
-'
-test_expect_success 'http.receivepack true' '
- log_div "receivepack true"
- config http.receivepack true &&
- GET info/refs?service=git-receive-pack "200 OK" &&
- POST git-receive-pack 0000 "200 OK"
-'
-test_expect_success 'http.receivepack false' '
- log_div "receivepack false"
- config http.receivepack false &&
- GET info/refs?service=git-receive-pack "403 Forbidden" &&
- POST git-receive-pack 0000 "403 Forbidden"
-'
-run_backend() {
- REQUEST_METHOD=GET \
- GIT_PROJECT_ROOT="$HTTPD_DOCUMENT_ROOT_PATH" \
- PATH_INFO="$2" \
- git http-backend >act.out 2>act.err
-}
-
-path_info() {
- if test $1 = 0; then
- run_backend "$2"
- else
- test_must_fail run_backend "$2" &&
- echo "fatal: '$2': aliased" >exp.err &&
- test_cmp exp.err act.err
- fi
-}
-
-test_expect_success 'http-backend blocks bad PATH_INFO' '
- config http.getanyfile true &&
-
- run_backend 0 /repo.git/HEAD &&
-
- run_backend 1 /repo.git/../HEAD &&
- run_backend 1 /../etc/passwd &&
- run_backend 1 ../etc/passwd &&
- run_backend 1 /etc//passwd &&
- run_backend 1 /etc/./passwd &&
- run_backend 1 /etc/.../passwd &&
- run_backend 1 //domain/data.txt
-'
+. "$TEST_DIRECTORY"/t556x_common
cat >exp <<EOF
diff --git a/t/t5561-http-backend-noserver.sh b/t/t5561-http-backend-noserver.sh
new file mode 100755
index 0000000..501b328
--- /dev/null
+++ b/t/t5561-http-backend-noserver.sh
@@ -0,0 +1,52 @@
+#!/bin/sh
+
+test_description='test git-http-backend-noserver'
+. ./test-lib.sh
+
+HTTPD_DOCUMENT_ROOT_PATH="$TRASH_DIRECTORY"
+
+GET() {
+ return 0
+}
+
+POST() {
+ return 0
+}
+
+logdiv() {
+ return 0
+}
+
+. "$TEST_DIRECTORY"/t556x_common
+
+run_backend() {
+ REQUEST_METHOD=GET \
+ GIT_PROJECT_ROOT="$HTTPD_DOCUMENT_ROOT_PATH" \
+ PATH_INFO="$2" \
+ git http-backend >act.out 2>act.err
+}
+
+path_info() {
+ if test $1 = 0; then
+ run_backend "$2"
+ else
+ test_must_fail run_backend "$2" &&
+ echo "fatal: '$2': aliased" >exp.err &&
+ test_cmp exp.err act.err
+ fi
+}
+
+test_expect_success 'http-backend blocks bad PATH_INFO' '
+ config http.getanyfile true &&
+
+ run_backend 0 /repo.git/HEAD &&
+
+ run_backend 1 /repo.git/../HEAD &&
+ run_backend 1 /../etc/passwd &&
+ run_backend 1 ../etc/passwd &&
+ run_backend 1 /etc//passwd &&
+ run_backend 1 /etc/./passwd &&
+ run_backend 1 /etc/.../passwd &&
+ run_backend 1 //domain/data.txt
+'
+test_done
diff --git a/t/t556x_common b/t/t556x_common
new file mode 100755
index 0000000..1b4921c
--- /dev/null
+++ b/t/t556x_common
@@ -0,0 +1,119 @@
+#!/bin/sh
+
+find_file() {
+ cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
+ find $1 -type f |
+ sed -e 1q
+}
+
+config() {
+ git --git-dir="$HTTPD_DOCUMENT_ROOT_PATH/repo.git" config $1 $2
+}
+
+test_expect_success 'setup repository' '
+ echo content >file &&
+ git add file &&
+ git commit -m one &&
+
+ mkdir "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
+ (cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
+ git --bare init &&
+ : >objects/info/alternates &&
+ : >objects/info/http-alternates
+ ) &&
+ git remote add public "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
+ git push public master:master &&
+
+ (cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
+ git repack -a -d
+ ) &&
+
+ echo other >file &&
+ git add file &&
+ git commit -m two &&
+ git push public master:master &&
+
+ LOOSE_URL=$(find_file objects/??) &&
+ PACK_URL=$(find_file objects/pack/*.pack) &&
+ IDX_URL=$(find_file objects/pack/*.idx)
+'
+
+get_static_files() {
+ GET HEAD "$1" &&
+ GET info/refs "$1" &&
+ GET objects/info/packs "$1" &&
+ GET objects/info/alternates "$1" &&
+ GET objects/info/http-alternates "$1" &&
+ GET $LOOSE_URL "$1" &&
+ GET $PACK_URL "$1" &&
+ GET $IDX_URL "$1"
+}
+
+SMART=smart
+test_expect_success 'direct refs/heads/master not found' '
+ log_div "refs/heads/master"
+ GET refs/heads/master "404 Not Found"
+'
+test_expect_success 'static file is ok' '
+ log_div "getanyfile default"
+ get_static_files "200 OK"
+'
+SMART=smart_noexport
+test_expect_success 'no export by default' '
+ log_div "no git-daemon-export-ok"
+ get_static_files "404 Not Found"
+'
+test_expect_success 'export if git-daemon-export-ok' '
+ log_div "git-daemon-export-ok"
+ (cd "$HTTPD_DOCUMENT_ROOT_PATH/repo.git" &&
+ touch git-daemon-export-ok
+ ) &&
+ get_static_files "200 OK"
+'
+SMART=smart
+test_expect_success 'static file if http.getanyfile true is ok' '
+ log_div "getanyfile true"
+ config http.getanyfile true &&
+ get_static_files "200 OK"
+'
+test_expect_success 'static file if http.getanyfile false fails' '
+ log_div "getanyfile false"
+ config http.getanyfile false &&
+ get_static_files "403 Forbidden"
+'
+
+test_expect_success 'http.uploadpack default enabled' '
+ log_div "uploadpack default"
+ GET info/refs?service=git-upload-pack "200 OK" &&
+ POST git-upload-pack 0000 "200 OK"
+'
+test_expect_success 'http.uploadpack true' '
+ log_div "uploadpack true"
+ config http.uploadpack true &&
+ GET info/refs?service=git-upload-pack "200 OK" &&
+ POST git-upload-pack 0000 "200 OK"
+'
+test_expect_success 'http.uploadpack false' '
+ log_div "uploadpack false"
+ config http.uploadpack false &&
+ GET info/refs?service=git-upload-pack "403 Forbidden" &&
+ POST git-upload-pack 0000 "403 Forbidden"
+'
+
+test_expect_success 'http.receivepack default disabled' '
+ log_div "receivepack default"
+ GET info/refs?service=git-receive-pack "403 Forbidden" &&
+ POST git-receive-pack 0000 "403 Forbidden"
+'
+test_expect_success 'http.receivepack true' '
+ log_div "receivepack true"
+ config http.receivepack true &&
+ GET info/refs?service=git-receive-pack "200 OK" &&
+ POST git-receive-pack 0000 "200 OK"
+'
+test_expect_success 'http.receivepack false' '
+ log_div "receivepack false"
+ config http.receivepack false &&
+ GET info/refs?service=git-receive-pack "403 Forbidden" &&
+ POST git-receive-pack 0000 "403 Forbidden"
+'
--
1.6.6
^ permalink raw reply related
* Re: Minor bug in bash completion
From: Sylvain Rabot @ 2009-12-30 17:59 UTC (permalink / raw)
To: git
In-Reply-To: <20091230112222.GA493@neumann>
On Wed, 2009-12-30 at 12:22 +0100, SZEDER Gábor wrote:
> Hi Sylvain,
>
>
> On Tue, Dec 29, 2009 at 03:36:58PM +0100, Sylvain RABOT wrote:
> > I found a bug in the git bash completion.
> > It occurs when I press tab to complete branch name when I want to pull
> > from the origin.
> > Instead of completing the branch name it prompts me directly for my
> > password on the origin remote.
>
> I don't think it's a bug. The completion script should offer the
> currently available refs in the remote repository after a 'git pull
> <remote> <TAB>'. In order to do that it contacts the remote
> repository for the list of refs available there. Depending on the
> access method, it might need to authenticate, in your case via ssh.
> To silence the password prompts you should change your ssh
> configuration to use key-based authentication when logging in to the
> remote repository's server (just google for 'ssh login without
> password').
>
>
> Best,
> Gábor
Haaa,
I understand now, tough, I thought that the list of origin's branches
where stored somewhere in my local repository and that I needed to do a
"git remote update origin" to get the current origin branches list.
Thanks for your answer.
Regards.
--
Sylvain Rabot <sylvain@abstraction.fr>
^ permalink raw reply
* Re: [PATCH RFC 1/2] Smart-http tests: Break test t5560-http-backend into pieces
From: Tarmigan @ 2009-12-30 18:09 UTC (permalink / raw)
To: Junio C Hamano; +Cc: Shawn O . Pearce, git
In-Reply-To: <7veimc2vq0.fsf@alter.siamese.dyndns.org>
On Wed, Dec 30, 2009 at 12:54 PM, Junio C Hamano <gitster@pobox.com> wrote:
> Tarmigan Casebolt <tarmigan+git@gmail.com> writes:
>
>> This should introduce no functional change in the tests or the amount
>> of test coverage.
>>
>> Signed-off-by: Tarmigan Casebolt <tarmigan+git@gmail.com>
>
> This seems to crash rather badly with your own "Smart-http: check if
> repository is OK to export before serving it".
Yes, they were both separately based on 'master' a few days ago.
If you think the goal (more about that is in the commit log of 2/2) is
worthwhile, I am happy to rebase on top of pu and resend.
One reason it's labeled RFC is that I'm not very confident in my
ability to write portable shell script. It works for me with bash,
but I'm not completely confident that is would work on ksh or dash.
So it would be nice if you could specifically take a look at the new
POST() and GET() and see if you notice anything obviously wrong there.
Thanks,
Tarmigan
^ permalink raw reply
* Re: Minor bug in bash completion
From: Junio C Hamano @ 2009-12-30 18:00 UTC (permalink / raw)
To: SZEDER Gábor; +Cc: Sylvain RABOT, spearce, git
In-Reply-To: <20091230112222.GA493@neumann>
SZEDER Gábor <szeder@ira.uka.de> writes:
> On Tue, Dec 29, 2009 at 03:36:58PM +0100, Sylvain RABOT wrote:
>> I found a bug in the git bash completion.
>> It occurs when I press tab to complete branch name when I want to pull
>> from the origin.
>> Instead of completing the branch name it prompts me directly for my
>> password on the origin remote.
>
> I don't think it's a bug. The completion script should offer the
> currently available refs in the remote repository after a 'git pull
> <remote> <TAB>'. In order to do that it contacts the remote
> repository for the list of refs available there.
It is definitely a non-bug since the user _asked_ to complete and the
information necessary to do the job the user _asked_ to do is only
accessible over the network connection.
I however do think this is an annoyance when it happens to you by
accident. I wonder if it is worth to have a way for users to tell
completion to skip the ls-remote invocation (hence not getting branches
completed) depending on the type of remote.
^ permalink raw reply
* Re: [PATCH] reset: unbreak hard resets with GIT_WORK_TREE
From: Junio C Hamano @ 2009-12-30 17:54 UTC (permalink / raw)
To: Jeff King
Cc: Junio C Hamano, Fyn Fynn, Nguyen Thai Ngoc Duy, Nanako Shiraishi,
git
In-Reply-To: <20091230084702.GA19649@coredump.intra.peff.net>
Jeff King <peff@peff.net> writes:
> Commit 952dfc6 tried to tighten the safety valves for doing
> a "reset --hard" in a bare repository or outside the work
> tree, but accidentally broke the case for GIT_WORK_TREE.
> This patch unbreaks it.
>
> Most git commands which need a work tree simply use
> NEED_WORK_TREE in git.c to die before they get to their
> cmd_* function. Reset, however, only needs a work tree in
> some cases, and so must handle the work tree itself. The
> error that 952dfc6 made was to simply forbid certain
> operations if the work tree was not set up; instead, we need
> to do the same thing that NEED_WORK_TREE does, which is to
> call setup_work_tree(). We no longer have to worry about dying
> in the non-worktree case, as setup_work_tree handles that
> for us.
Sounds very sane to me. Thanks.
^ permalink raw reply
* Re: [PATCH RFC 1/2] Smart-http tests: Break test t5560-http-backend into pieces
From: Junio C Hamano @ 2009-12-30 17:54 UTC (permalink / raw)
To: Tarmigan Casebolt; +Cc: Shawn O . Pearce, git
In-Reply-To: <1262037899-16786-1-git-send-email-tarmigan+git@gmail.com>
Tarmigan Casebolt <tarmigan+git@gmail.com> writes:
> This should introduce no functional change in the tests or the amount
> of test coverage.
>
> Signed-off-by: Tarmigan Casebolt <tarmigan+git@gmail.com>
This seems to crash rather badly with your own "Smart-http: check if
repository is OK to export before serving it".
^ permalink raw reply
* Re: [PATCH/resend] CVS Server: Support reading base and roots from environment
From: Phil Miller @ 2009-12-30 17:12 UTC (permalink / raw)
To: Nanako Shiraishi; +Cc: Junio C Hamano, Git Mailing List
In-Reply-To: <20091230224108.6117@nanako3.lavabit.com>
On Wed, Dec 30, 2009 at 07:41, Nanako Shiraishi <nanako3@lavabit.com> wrote:
> Junio, could you tell us what happened to this thread?
>
> Phil Miller's patch to help gitosis installation. No response from
> git-cvsserver users.
I never resent with modifications as Junio requested. I've been busy
with some end-of-year close-out stuff, and actually wanted to
dramatically simplify part of the patch before the resend. I'll do
that shortly.
Since Gitosis wants to do repository-level access control before
running any Git commands, I figured out a CVSROOT format that would
let it do this. Given that, there's only one repository that an
invocation of git-cvsserver should be allowed to access, and so the
issue of "list separator" and so forth is moot. The resultant CVS
command line looks something like
cvs -d ":ext;CVS_SERVER=cvs
'/path/to/module':charmgit/path/to/module.git" co -d module
Note that I have had to modify Gitosis to make this, and various other
things, work. I had wanted to clean up my work before announcing it,
but if it would help you now, it is publicly available from github
now:
git://github.com/PhilMiller/gitosis.git
The documentation hasn't been updated, but the recent history makes as
least some usage clear. I'll send more details in a follow-up.
Phil
^ permalink raw reply
* [PATCH] builtin-config: add --path option doing ~ and ~user expansion.
From: Matthieu Moy @ 2009-12-30 16:51 UTC (permalink / raw)
To: git, gitster; +Cc: Matthieu Moy
395de250 (Expand ~ and ~user in core.excludesfile, commit.template)
introduced a C function git_config_pathname, doing ~/ and ~user/
expansion. This patch makes the feature available to scripts with 'git
config --get --path'.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
---
Documentation/git-config.txt | 14 +++++++++++---
builtin-config.c | 20 +++++++++++++++++++-
t/t1300-repo-config.sh | 28 ++++++++++++++++++++++++++++
3 files changed, 58 insertions(+), 4 deletions(-)
diff --git a/Documentation/git-config.txt b/Documentation/git-config.txt
index f68b198..2632928 100644
--- a/Documentation/git-config.txt
+++ b/Documentation/git-config.txt
@@ -37,11 +37,12 @@ existing values that match the regexp are updated or unset. If
you want to handle the lines that do *not* match the regex, just
prepend a single exclamation mark in front (see also <<EXAMPLES>>).
-The type specifier can be either '--int' or '--bool', which will make
+The type specifier can be either '--int' or '--bool', to make
'git-config' ensure that the variable(s) are of the given type and
convert the value to the canonical form (simple decimal number for int,
-a "true" or "false" string for bool). If no type specifier is passed,
-no checks or transformations are performed on the value.
+a "true" or "false" string for bool), or '--path', which does some
+path expansion (see '--path' below). If no type specifier is passed, no
+checks or transformations are performed on the value.
The file-option can be one of '--system', '--global' or '--file'
which specify where the values will be read from or written to.
@@ -136,6 +137,13 @@ See also <<FILES>>.
'git-config' will ensure that the output matches the format of
either --bool or --int, as described above.
+--path::
+ 'git-config' will expand leading '{tilde}' to the value of
+ '$HOME', and '{tilde}user' to the home directory for the
+ specified user. This option has no effect when setting the
+ value (but you can use 'git config bla {tilde}/' from the
+ command line to let your shell do the expansion).
+
-z::
--null::
For all options that output values and/or keys, always
diff --git a/builtin-config.c b/builtin-config.c
index a2d656e..2e3ef91 100644
--- a/builtin-config.c
+++ b/builtin-config.c
@@ -45,6 +45,7 @@ static int end_null;
#define TYPE_BOOL (1<<0)
#define TYPE_INT (1<<1)
#define TYPE_BOOL_OR_INT (1<<2)
+#define TYPE_PATH (1<<3)
static struct option builtin_config_options[] = {
OPT_GROUP("Config file location"),
@@ -69,6 +70,7 @@ static struct option builtin_config_options[] = {
OPT_BIT(0, "bool", &types, "value is \"true\" or \"false\"", TYPE_BOOL),
OPT_BIT(0, "int", &types, "value is decimal number", TYPE_INT),
OPT_BIT(0, "bool-or-int", &types, "value is --bool or --int", TYPE_BOOL_OR_INT),
+ OPT_BIT(0, "path", &types, "value is a path (file or directory name)", TYPE_PATH),
OPT_GROUP("Other"),
OPT_BOOLEAN('z', "null", &end_null, "terminate values with NUL byte"),
OPT_END(),
@@ -94,6 +96,7 @@ static int show_config(const char *key_, const char *value_, void *cb)
{
char value[256];
const char *vptr = value;
+ int must_free_vptr = 0;
int dup_error = 0;
if (!use_key_regexp && strcmp(key_, key))
@@ -123,6 +126,9 @@ static int show_config(const char *key_, const char *value_, void *cb)
vptr = v ? "true" : "false";
else
sprintf(value, "%d", v);
+ } else if (types == TYPE_PATH) {
+ git_config_pathname(&vptr, key_, value_);
+ must_free_vptr = 1;
}
else
vptr = value_?value_:"";
@@ -133,6 +139,12 @@ static int show_config(const char *key_, const char *value_, void *cb)
}
else
printf("%s%c", vptr, term);
+ if (must_free_vptr)
+ /* If vptr must be freed, it's a pointer to a
+ * dynamically allocated buffer, it's safe to cast to
+ * const.
+ */
+ free((char *)vptr);
return 0;
}
@@ -215,7 +227,13 @@ static char *normalize_value(const char *key, const char *value)
if (!value)
return NULL;
- if (types == 0)
+ if (types == 0 || types == TYPE_PATH)
+ /*
+ * We don't do normalization for TYPE_PATH here: If
+ * the path is like ~/foobar/, we prefer to store
+ * "~/foobar/" in the config file, and to expand the ~
+ * when retrieving the value.
+ */
normalized = xstrdup(value);
else {
normalized = xmalloc(64);
diff --git a/t/t1300-repo-config.sh b/t/t1300-repo-config.sh
index 83b7294..f89d7e9 100755
--- a/t/t1300-repo-config.sh
+++ b/t/t1300-repo-config.sh
@@ -683,6 +683,34 @@ test_expect_success 'set --bool-or-int' '
rm .git/config
+cat >expect <<\EOF
+[path]
+ home = ~/
+ normal = /dev/null
+ trailingtilde = foo~
+EOF
+
+test_expect_success 'set --path' '
+ git config --path path.home "~/" &&
+ git config --path path.normal "/dev/null" &&
+ git config --path path.trailingtilde "foo~" &&
+ test_cmp expect .git/config'
+
+cat >expect <<EOF
+$HOME/
+/dev/null
+foo~
+EOF
+
+test_expect_success 'get --path' '
+ git config --get --path path.home > result &&
+ git config --get --path path.normal >> result &&
+ git config --get --path path.trailingtilde >> result &&
+ test_cmp expect result
+'
+
+rm .git/config
+
git config quote.leading " test"
git config quote.ending "test "
git config quote.semicolon "test;test"
--
1.6.6.71.g392d.dirty
^ permalink raw reply related
* GNU patch: 2.6.1 release (mostly bugfixes)
From: Andreas Gruenbacher @ 2009-12-30 16:30 UTC (permalink / raw)
To: info-gnu; +Cc: git
Some glitches in release 2.6 of GNU patch made a new release 2.6.1 necessary,
available from:
ftp://ftp.gnu.org/gnu/patch/
The previous release was on 13 November. NEWS since then:
* Support for diff3(1) style merges which show the old, original, and new
lines of a conflict has been added (--merge=diff3). The default still is
the merge(1) format (--merge or --merge=merge).
* Bug and portability fixes.
The source code repository, a bug-patch@gnu.org mailing list archive, and the
project's bug tracker are available at:
http://savannah.gnu.org/projects/patch
Please email bugs or suggestions to <bug-patch@gnu.org>. Check the bug
tracker for a list of known issues.
Thanks,
Andreas
^ permalink raw reply
* [PATCH] fast-import: Document author/committer/tagger name is optional
From: Shawn O. Pearce @ 2009-12-30 15:03 UTC (permalink / raw)
To: David Reiss, Junio C Hamano; +Cc: git@vger.kernel.org
In-Reply-To: <4B3AED66.3030803@facebook.com>
The fast-import parser does not validate that the author, committer
or tagger name component contains both a name and an email address.
Therefore the name component has always been optional. Correct the
documentation to match the implementation.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
---
David Reiss <dreiss@facebook.com> wrote:
> >> author <somename> 1261454209 +0000
> >> committer <somename> 1261454209 +0000
> > a foreign system where the data might not reasonably exist.
> But shouldn't there still be an extra space? One to separate "author"
> from the empty name, and one to separate the empty name from the email?
> If not, then I think this change should be made. (I couldn't find any
> authoritative documentation on what constitutes a valid commit object.)
Yes, we should do this.
Documentation/git-fast-import.txt | 6 +++---
fast-import.c | 6 +++---
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/Documentation/git-fast-import.txt b/Documentation/git-fast-import.txt
index 288032c..e6d364f 100644
--- a/Documentation/git-fast-import.txt
+++ b/Documentation/git-fast-import.txt
@@ -311,8 +311,8 @@ change to the project.
....
'commit' SP <ref> LF
mark?
- ('author' SP <name> SP LT <email> GT SP <when> LF)?
- 'committer' SP <name> SP LT <email> GT SP <when> LF
+ ('author' (SP <name>)? SP LT <email> GT SP <when> LF)?
+ 'committer' (SP <name>)? SP LT <email> GT SP <when> LF
data
('from' SP <committish> LF)?
('merge' SP <committish> LF)?
@@ -657,7 +657,7 @@ lightweight (non-annotated) tags see the `reset` command below.
....
'tag' SP <name> LF
'from' SP <committish> LF
- 'tagger' SP <name> SP LT <email> GT SP <when> LF
+ 'tagger' (SP <name>)? SP LT <email> GT SP <when> LF
data
....
diff --git a/fast-import.c b/fast-import.c
index dd3c99d..cd87049 100644
--- a/fast-import.c
+++ b/fast-import.c
@@ -19,8 +19,8 @@ Format of STDIN stream:
new_commit ::= 'commit' sp ref_str lf
mark?
- ('author' sp name sp '<' email '>' sp when lf)?
- 'committer' sp name sp '<' email '>' sp when lf
+ ('author' (sp name)? sp '<' email '>' sp when lf)?
+ 'committer' (sp name)? sp '<' email '>' sp when lf
commit_msg
('from' sp committish lf)?
('merge' sp committish lf)*
@@ -47,7 +47,7 @@ Format of STDIN stream:
new_tag ::= 'tag' sp tag_str lf
'from' sp committish lf
- ('tagger' sp name sp '<' email '>' sp when lf)?
+ ('tagger' (sp name)? sp '<' email '>' sp when lf)?
tag_msg;
tag_msg ::= data;
--
1.6.6.307.gba67
--
Shawn.
^ permalink raw reply related
* Re: [PATCH] bash completion: add space between branch name and status flags
From: Shawn O. Pearce @ 2009-12-30 14:57 UTC (permalink / raw)
To: Nanako Shiraishi; +Cc: Junio C Hamano, git, Roman Fietze
In-Reply-To: <20091230224129.6117@nanako3.lavabit.com>
Nanako Shiraishi <nanako3@lavabit.com> wrote:
> Junio, could you tell us what happened to this thread?
>
> In response to a patch from Roman Fietze to outline a better way
> to do it. Nothing happened.
Junio responded with a suggestion on how to improve the patch when
GIT_PS1_SHOWDIRTYSTATE was not set, but Roman Fietze didn't send
a revised patch, so it got dropped.
Here is the revised patch, Junio, still think its a good idea?
--8<--
From: Roman Fietze <roman.fietze@telemotive.de>
Subject: [PATCH] bash completion: add space between branch name and status flags
Improve the readability of the bash prompt by adding a space between
the branch name and the status flags (dirty, stash, untracked).
Signed-off-by: Roman Fietze <roman.fietze@telemotive.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
---
contrib/completion/git-completion.bash | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash
index fbfa5f2..3c8b6df 100755
--- a/contrib/completion/git-completion.bash
+++ b/contrib/completion/git-completion.bash
@@ -163,10 +163,12 @@ __git_ps1 ()
fi
fi
+ local f="$w$i$s$u"
+ f="${f:+ $f}$r"
if [ -n "${1-}" ]; then
- printf "$1" "$c${b##refs/heads/}$w$i$s$u$r"
+ printf "$1" "$c${b##refs/heads/}$f"
else
- printf " (%s)" "$c${b##refs/heads/}$w$i$s$u$r"
+ printf " (%s)" "$c${b##refs/heads/}$f"
fi
fi
}
--
1.6.6.307.gba67
--
Shawn.
^ permalink raw reply related
* [PATCH] SubmittingPatches: hints to know the status of a submitted patch.
From: Matthieu Moy @ 2009-12-30 14:51 UTC (permalink / raw)
To: git, gitster; +Cc: Matthieu Moy
"What happened to my patch" is pretty much a FAQ on the Git mailing list,
it deserves a few paragraphs in SubmittingPatches...
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
---
(I had already sent this patch as a draft RFC, admitedly rather well
hidden within another thread. Let's retry a bit more explicitly ;-) )
Documentation/SubmittingPatches | 14 ++++++++++++++
1 files changed, 14 insertions(+), 0 deletions(-)
diff --git a/Documentation/SubmittingPatches b/Documentation/SubmittingPatches
index 76fc84d..c686f86 100644
--- a/Documentation/SubmittingPatches
+++ b/Documentation/SubmittingPatches
@@ -280,6 +280,20 @@ people play with it without having to pick up and apply the patch to
their trees themselves.
------------------------------------------------
+Know the status of your patch after submission
+
+* You can use Git itself to find out when your patch is merged in
+ master. 'git pull --rebase' will automatically skip already-applied
+ patches, and will let you know. This works only if you rebase on top
+ of the branch in which your patch has been merged (i.e. it will not
+ tell you if your patch is merged in pu if you rebase on top of
+ master).
+
+* Read the git mailing list, the maintainer regularly posts messages
+ entitled "What's cooking in git.git" and "What's in git.git" giving
+ the status of various proposed changes.
+
+------------------------------------------------
MUA specific hints
Some of patches I receive or pick up from the list share common
--
1.6.6.71.gcc720.dirty
^ permalink raw reply related
* [PATCH] branch: die explicitly why when calling "git branch [-a|-r] branchname".
From: Matthieu Moy @ 2009-12-30 14:45 UTC (permalink / raw)
To: git, gitster; +Cc: Matthieu Moy
The -a and -r options used to be silently ignored in such a command.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
---
builtin-branch.c | 7 +++++--
1 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin-branch.c b/builtin-branch.c
index c87e63b..8aef2bb 100644
--- a/builtin-branch.c
+++ b/builtin-branch.c
@@ -638,10 +638,13 @@ int cmd_branch(int argc, const char **argv, const char *prefix)
rename_branch(head, argv[0], rename > 1);
else if (rename && (argc == 2))
rename_branch(argv[0], argv[1], rename > 1);
- else if (argc <= 2)
+ else if (argc <= 2) {
+ if (kinds != REF_LOCAL_BRANCH) {
+ die("-a and -r options to 'git branch' do not make sense with a branch name");
+ }
create_branch(head, argv[0], (argc == 2) ? argv[1] : head,
force_create, reflog, track);
- else
+ } else
usage_with_options(builtin_branch_usage, options);
return 0;
--
1.6.6.71.gcc720.dirty
^ permalink raw reply related
* Re: [PATCH v2] Let core.excludesfile default to ~/.gitexcludes.
From: Matthieu Moy @ 2009-12-30 14:31 UTC (permalink / raw)
To: Nanako Shiraishi; +Cc: Junio C Hamano, git
In-Reply-To: <20091230224135.6117@nanako3.lavabit.com>
Nanako Shiraishi <nanako3@lavabit.com> writes:
> Junio, could you tell us what happened to this thread?
I'm not Junio, but I can try ...
Setting a default for core.excludesfile implies that changing it later
is hard. In particular, _if_ we chose one day to move from multiple
$HOME/.gitsomething files to $HOME/something-else/multiple-files
($XDG_CONFIG_HOME and $HOME/.gitglobal/{config,excludes,...} have been
suggested, as well as a $GIT_HOME variable), then the transition is
made harder by having a default.
So, while I think having a default is good, chosing the default
shouldn't be taken lightly.
I personally like the idea of a user config _directory_ with multiple
files in it, but I'm not clear on exactly what it should be, how the
transition plan would be, ...
So, I dropped the patch until we get closer to a consensus on what the
default value should be.
--
Matthieu Moy
http://www-verimag.imag.fr/~moy/
^ permalink raw reply
* [PATCH] grep: do not do external grep on skip-worktree entries
From: Nguyễn Thái Ngọc Duy @ 2009-12-30 14:11 UTC (permalink / raw)
To: git; +Cc: Nguyễn Thái Ngọc Duy
Skip-worktree entries are not on disk. There is no reason to call
external grep in such cases.
A trace message is also added to aid debugging external grep cases.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
builtin-grep.c | 19 ++++++++++++++++++-
1 files changed, 18 insertions(+), 1 deletions(-)
diff --git a/builtin-grep.c b/builtin-grep.c
index d582232..d49c637 100644
--- a/builtin-grep.c
+++ b/builtin-grep.c
@@ -488,17 +488,34 @@ static int grep_cache(struct grep_opt *opt, const char **paths, int cached,
read_cache();
#if !NO_EXTERNAL_GREP
+ if (cached)
+ external_grep_allowed = 0;
+ if (external_grep_allowed) {
+ for (nr = 0; nr < active_nr; nr++) {
+ struct cache_entry *ce = active_cache[nr];
+ if (!S_ISREG(ce->ce_mode))
+ continue;
+ if (!pathspec_matches(paths, ce->name, opt->max_depth))
+ continue;
+ if (ce_skip_worktree(ce)) {
+ external_grep_allowed = 0;
+ break;
+ }
+ }
+ }
/*
* Use the external "grep" command for the case where
* we grep through the checked-out files. It tends to
* be a lot more optimized
*/
- if (!cached && external_grep_allowed) {
+ if (external_grep_allowed) {
hit = external_grep(opt, paths, cached);
if (hit >= 0)
return hit;
hit = 0;
}
+ else
+ trace_printf("grep: external grep not used\n");
#endif
for (nr = 0; nr < active_nr; nr++) {
--
1.6.6.315.g1a406
^ permalink raw reply related
* Re: [PATCH] git-update-index: report(...) now flushes stdout after printing the report line
From: Sebastian Thiel @ 2009-12-30 13:56 UTC (permalink / raw)
To: git
In-Reply-To: <loom.20091119T221732-624@post.gmane.org>
I'd like to add that since version 1.6.5, non-tty's do not receive any progress
information anymore. The patch causing this says it wants to, in short words,
unify the push and fetch handling regarding the way progress messages are sent.
Now third-party wrappers, such as git-python, are unable to provide any progress
information anymore for possibly lengthy operations.
This is why I clearly recommend to add some kind of a "progress-force" flag that
turns progress messages on again for send-pack and receive-pack.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox