date:20130604

On Tue, Jun 4, 2013 at 1:26 PM, Junio C Hamano  wrote:
> Nguyễn Thái Ngọc Duy  writes:
>
>>  > Have you ever tested this?
>>  >
>>  > Once log_pack_access goes to NULL (e.g. when it sees the empty
>>  > string it was initialized to), this new test will happily
>>  > dereference NULL.
>>
>>  My bad. I did test when GIT_TRACE_PACK_ACCESS was set, not when it
>>  was set to an empty string. All cases tested now.
>>
>> @@ -1956,6 +1958,14 @@ static void write_pack_access_log(struct packed_git 
>> *p, off_t obj_offset)
>>  {
>>   static FILE *log_file;
>>
>> + if (!*log_pack_access) {
>
> The first time, we will see the static empty string and come into
> this block...
>
>> + log_pack_access = getenv("GIT_TRACE_PACK_ACCESS");
>> + if (log_pack_access && !*log_pack_access)
>> + log_pack_access = NULL;
>> + if (!log_pack_access)
>> + return;
>> + }
>
> This may
>
>  (1) not find the env-var, in which case log_pack_access becomes
>  NULL here, and the function returns.
>
>  (2) find the env-var to be an empty string, in which case
>  log_pack_access becomes NULL in the next statement, and the
>  function returns.
>
>  (3) find the env-var to be a non-empty string, and the function
>  does not return.
>
> And the next time around, (3) above may work fine; the first if()
> will fail and we do not come in.  But in either (1) or (2), don't
> you keep checking the environment every time you come here?
>
> Oh, no, it is worse than that.  Either case you set the variable to
> NULL, so the very first "if (!*log_pack_access)" would dereference
> NULL.

You found it out already. The only caller does this

if (log_pack_access) write_pack_access_log(p, obj_offset);

so in (1) and (2), write_pack_access_log() will never be called again.
Originally I had "log_pack_access && !*log_pack_access" in the first
if(), but I dropped it because of the caller's condition. It's a bit
fragile though. Imagine a new callsite is added.. (to be continued)

> Why not start from NULL:
>
> static const char *log_pack_access;
>
> treat that NULL as "unknown" state, and avoid running getenv over
> and over again by treating non-NULL value as "known"?  Perhaps like
> this?
>
> if (!log_pack_access) {
> /* First time: is env set? */
> log_pack_access = getenv("GIT_TRACE_PACK_ACCESS");
> if (!log_pack_access)
> log_pack_access = "";

or set log_pack_access to SomePredefinedButUselessString here. I
wanted to do this but couldn't because of global variable
initialization.

> }
> /* Now GIT_TRACE_PACK_ACCESS is known */
> if (!*log_pack_access)
> return; /* not wanted */
>
> As you can no longer rely on "config is read before we do anything
> else" by switching to lazy env checking, your guard at the caller
> needs to be updated, if you want to reduce unneeded call-return
> overhead:
>
> if (!log_pack_access || *log_pack_access)
> write_pack_access_log(...);

and this turns to "if (!log_.. || log != SomePrede...)", no more
peeking into log_pack_access.

>
> But the guard is purely for performance, not correctness, because
> the function now does the "return no-op if we know we are not
> wanted" itself.

(continued here) ..so yeah this looks better.

> Also you no longer need to have an extern variable in environment.c

will fix.
--
Duy
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] Fix `git svn rebase` if top-level HEAD directory exist

On Tue, Jun 04, 2013 at 11:32:56AM +0400, ojab wrote:

> Oh hai!

You can haz patch?

> I have a svn repo with the top-level directory named HEAD and `git
> svn rebase [HEAD] [--]` fails with
> >$ git svn rebase
> >fatal: ambiguous argument 'HEAD': both revision and filename
> >Use '--' to separate paths from revisions, like this:
> >'git  [...] -- [...]'
> >rev-list --first-parent --pretty=medium HEAD: command returned error: 128

This rationale should probably go in the commit message.

> From 522cbc8b8a7c4f2ab4268551a550585753164677 Mon Sep 17 00:00:00 2001
> From: ojab 
> Date: Tue, 4 Jun 2013 11:28:16 +0400
> Subject: [PATCH] Fix `git svn rebase` if top-level HEAD directory exist

We prefer patches to be inline in the email; these lines can be
dropped, as they are picked up from your email headers.

> Signed-off-by: ojab 

Do you mind providing a real name? The point of Signed-off-by is for
licensing and attribution.

> diff --git a/git-svn.perl b/git-svn.perl
> index d070de0..e35a66a 100755
> --- a/git-svn.perl
> +++ b/git-svn.perl
> @@ -1932,7 +1932,7 @@ sub cmt_sha2rev_batch {
>  sub working_head_info {
>   my ($head, $refs) = @_;
>   my @args = qw/rev-list --first-parent --pretty=medium/;
> - my ($fh, $ctx) = command_output_pipe(@args, $head);
> + my ($fh, $ctx) = command_output_pipe(@args, $head, "--");

Looks obviously correct to me. I did a quick grep, and there is one
other spot that probably should get the same treatment:

diff --git a/git-svn.perl b/git-svn.perl
index d070de0..07797ad 100755
--- a/git-svn.perl
+++ b/git-svn.perl
@@ -831,7 +831,7 @@ sub cmd_dcommit {
 sub cmd_dcommit {
my $head = shift;
command_noisy(qw/update-index --refresh/);
-   git_cmd_try { command_oneline(qw/diff-index --quiet HEAD/) }
+   git_cmd_try { command_oneline(qw/diff-index --quiet HEAD --/) }
'Cannot dcommit with a dirty index.  Commit your changes first, 
'
. "or stash them with `git stash'.\n";
$head ||= 'HEAD';

Feel free to squash it in if you re-roll your patch. There are a few
other spots that feed full sha1s. They are probably less likely to
trigger, but perhaps should be protected, too, just in case.

-Peff
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 0/6] git send-email suppress-cc=self fixes

This includes bugfixes related to handling of --suppress-cc=self
flag. Tests are also included.

Changes from v2:
- add a new test, split patches differently add code comments
 to address comments by Junio
- rename example addresses in tests from redhat.com to example.com
Changes from v1:
- tweak coding style in tests to address comments by Junio


Michael S. Tsirkin (6):
  send-email: fix suppress-cc=self on cccmd
  t/send-email: test suppress-cc=self on cccmd
  send-email: make --suppress-cc=self sanitize input
  t/send-email: add test with quoted sender
  t/send-email: test suppress-cc=self with non-ascii
  test-send-email: test for pre-sanitized self name

 git-send-email.perl   | 23 +++
 t/t9001-send-email.sh | 34 +-
 2 files changed, 48 insertions(+), 9 deletions(-)

-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 2/6] t/send-email: test suppress-cc=self on cccmd

Check that suppress-cc=self works when applied
to output of cccmd.

Signed-off-by: Michael S. Tsirkin 
---
 t/t9001-send-email.sh | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/t/t9001-send-email.sh b/t/t9001-send-email.sh
index e1a7f3e..f81e5f5 100755
--- a/t/t9001-send-email.sh
+++ b/t/t9001-send-email.sh
@@ -204,13 +204,15 @@ test_suppress_self_unquoted () {
 
unquoted-$3
 
+   cccmd--$1 <$2>
+
Cc: $1 <$2>
Signed-off-by: $1 <$2>
EOF
 }
 
 test_expect_success $PREREQ 'self name is suppressed' "
-   test_suppress_self_unquoted 'A U Thor' 'aut...@redhat.com' \
+   test_suppress_self_unquoted 'A U Thor' 'aut...@example.com' \
'self_name_suppressed'
 "
 
-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 3/6] send-email: make --suppress-cc=self sanitize input

--suppress-cc=self fails to filter sender address in many cases where it
needs to be sanitized in some way, for example quoted:
"A U. Thor" 
To fix, make send-email sanitize both sender and the address it is
compared against.

Signed-off-by: Michael S. Tsirkin 
---
 git-send-email.perl | 23 +++
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/git-send-email.perl b/git-send-email.perl
index f366baa..3b828f6 100755
--- a/git-send-email.perl
+++ b/git-send-email.perl
@@ -760,6 +760,11 @@ if (!defined $sender) {
$sender = $repoauthor || $repocommitter || '';
 }
 
+# $sender could be an already sanitized address
+# (e.g. sendemail.from could be manually sanitized by user).
+# But it's a no-op to run sanitize_address on an already sanitized address.
+$sender = sanitize_address($sender);
+
 my $prompting = 0;
 if (!@initial_to && !defined $to_cmd) {
my $to = ask("Who should the emails be sent to (if any)? ",
@@ -1113,10 +1118,9 @@ sub send_message {
if ($cc ne '') {
$ccline = "\nCc: $cc";
}
-   my $sanitized_sender = sanitize_address($sender);
make_message_id() unless defined($message_id);
 
-   my $header = "From: $sanitized_sender
+   my $header = "From: $sender
 To: $to${ccline}
 Subject: $subject
 Date: $date
@@ -1133,7 +1137,7 @@ X-Mailer: git-send-email $gitversion
}
 
my @sendmail_parameters = ('-i', @recipients);
-   my $raw_from = $sanitized_sender;
+   my $raw_from = $sender;
if (defined $envelope_sender && $envelope_sender ne "auto") {
$raw_from = $envelope_sender;
}
@@ -1308,8 +1312,9 @@ foreach my $t (@files) {
}
elsif (/^From:\s+(.*)$/i) {
($author, $author_encoding) = 
unquote_rfc2047($1);
+   my $sauthor = sanitize_address($author);
next if $suppress_cc{'author'};
-   next if $suppress_cc{'self'} and $author eq 
$sender;
+   next if $suppress_cc{'self'} and $sauthor eq 
$sender;
printf("(mbox) Adding cc: %s from line '%s'\n",
$1, $_) unless $quiet;
push @cc, $1;
@@ -1323,7 +1328,9 @@ foreach my $t (@files) {
}
elsif (/^Cc:\s+(.*)$/i) {
foreach my $addr (parse_address_line($1)) {
-   if (unquote_rfc2047($addr) eq $sender) {
+   my $qaddr = unquote_rfc2047($addr);
+   my $saddr = sanitize_address($qaddr);
+   if ($saddr eq $sender) {
next if ($suppress_cc{'self'});
} else {
next if ($suppress_cc{'cc'});
@@ -1370,7 +1377,8 @@ foreach my $t (@files) {
chomp;
my ($what, $c) = ($1, $2);
chomp $c;
-   if ($c eq $sender) {
+   my $sc = sanitize_address($c);
+   if ($sc eq $sender) {
next if ($suppress_cc{'self'});
} else {
next if $suppress_cc{'sob'} and $what =~ 
/Signed-off-by/i;
@@ -1454,7 +1462,6 @@ foreach my $t (@files) {
 sub recipients_cmd {
my ($prefix, $what, $cmd, $file) = @_;
 
-   my $sanitized_sender = sanitize_address($sender);
my @addresses = ();
open my $fh, "-|", "$cmd \Q$file\E"
or die "($prefix) Could not execute '$cmd'";
@@ -1462,7 +1469,7 @@ sub recipients_cmd {
$address =~ s/^\s*//g;
$address =~ s/\s*$//g;
$address = sanitize_address($address);
-   next if ($address eq $sanitized_sender and 
$suppress_cc{'self'});
+   next if ($address eq $sender and $suppress_cc{'self'});
push @addresses, $address;
printf("($prefix) Adding %s: %s from: '%s'\n",
   $what, $address, $cmd) unless $quiet;
-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 5/6] t/send-email: test suppress-cc=self with non-ascii

test suppress-cc=self when sender is non-acsii

Signed-off-by: Michael S. Tsirkin 
---
 t/t9001-send-email.sh | 5 +
 1 file changed, 5 insertions(+)

diff --git a/t/t9001-send-email.sh b/t/t9001-send-email.sh
index 6c0f715..0d50fa7 100755
--- a/t/t9001-send-email.sh
+++ b/t/t9001-send-email.sh
@@ -236,6 +236,11 @@ test_expect_success $PREREQ 'self name with dot is 
suppressed' "
'self_name_dot_suppressed'
 "
 
+test_expect_success $PREREQ 'non-ascii self name is suppressed' "
+   test_suppress_self_quoted 'Füñný Nâmé' 'odd_?=m...@example.com' \
+   'non_ascii_self_suppressed'
+"
+
 test_expect_success $PREREQ 'Show all headers' '
git send-email \
--dry-run \
-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 6/6] test-send-email: test for pre-sanitized self name

Users can sanitize from address manually.
Verify that these are suppressed properly.

Signed-off-by: Michael S. Tsirkin 
---
 t/t9001-send-email.sh | 5 +
 1 file changed, 5 insertions(+)

diff --git a/t/t9001-send-email.sh b/t/t9001-send-email.sh
index 0d50fa7..38f407d 100755
--- a/t/t9001-send-email.sh
+++ b/t/t9001-send-email.sh
@@ -241,6 +241,11 @@ test_expect_success $PREREQ 'non-ascii self name is 
suppressed' "
'non_ascii_self_suppressed'
 "
 
+test_expect_success $PREREQ 'sanitized self name is suppressed' "
+   test_suppress_self_unquoted '\"A U. Thor\"' 'aut...@example.com' \
+   'self_name_sanitized_suppressed'
+"
+
 test_expect_success $PREREQ 'Show all headers' '
git send-email \
--dry-run \
-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 1/6] send-email: fix suppress-cc=self on cccmd

When cccmd is used, old-style suppress-from filter
is applied by the newer suppress-cc=self isn't.
Fix this up.

Signed-off-by: Michael S. Tsirkin 
---

This makes one line a bit too long, but a follow-up patch
fixes this up.

 git-send-email.perl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/git-send-email.perl b/git-send-email.perl
index bd13cc8..f366baa 100755
--- a/git-send-email.perl
+++ b/git-send-email.perl
@@ -1462,7 +1462,7 @@ sub recipients_cmd {
$address =~ s/^\s*//g;
$address =~ s/\s*$//g;
$address = sanitize_address($address);
-   next if ($address eq $sanitized_sender and $suppress_from);
+   next if ($address eq $sanitized_sender and 
$suppress_cc{'self'});
push @addresses, $address;
printf("($prefix) Adding %s: %s from: '%s'\n",
   $what, $address, $cmd) unless $quiet;
-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 4/6] t/send-email: add test with quoted sender

add test where sender address needs to be quoted.
Make sure --suppress-cc=self works well in this case.

Signed-off-by: Michael S. Tsirkin 
---
 t/t9001-send-email.sh | 20 
 1 file changed, 20 insertions(+)

diff --git a/t/t9001-send-email.sh b/t/t9001-send-email.sh
index f81e5f5..6c0f715 100755
--- a/t/t9001-send-email.sh
+++ b/t/t9001-send-email.sh
@@ -211,11 +211,31 @@ test_suppress_self_unquoted () {
EOF
 }
 
+test_suppress_self_quoted () {
+   test_suppress_self "$1" "$2" "quoted-$3" <<-EOF
+   test suppress-cc.self quoted-$3 with name $1 email $2
+
+   quoted-$3
+
+   cccmd--"$1" <$2>
+
+   Cc: $1 <$2>
+   Cc: "$1" <$2>
+   Signed-off-by: $1 <$2>
+   Signed-off-by: "$1" <$2>
+   EOF
+}
+
 test_expect_success $PREREQ 'self name is suppressed' "
test_suppress_self_unquoted 'A U Thor' 'aut...@example.com' \
'self_name_suppressed'
 "
 
+test_expect_success $PREREQ 'self name with dot is suppressed' "
+   test_suppress_self_quoted 'A U. Thor' 'aut...@example.com' \
+   'self_name_dot_suppressed'
+"
+
 test_expect_success $PREREQ 'Show all headers' '
git send-email \
--dry-run \
-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Re: What's cooking in git.git (May 2013, #09; Wed, 29)

2013-06-04 Thread John Keeping

On Tue, Jun 04, 2013 at 03:29:51PM +1000, Heiko Voigt wrote:
> On Mon, Jun 03, 2013 at 11:23:41PM +0100, John Keeping wrote:
> > > Sorry, I should have been more specific here. I saw that you did some
> > > changes to make "submodule add" do the right thing with relative paths,
> > > but the following change to t7406 does not work like I believe it
> > > should but instead makes the test fail:
> > > ---8<-
> > > diff --git a/t/t7406-submodule-update.sh b/t/t7406-submodule-update.sh
> > > index a4ffea0..9766b9e 100755
> > > --- a/t/t7406-submodule-update.sh
> > > +++ b/t/t7406-submodule-update.sh
> > > @@ -559,7 +559,9 @@ test_expect_success 'add different submodules to the 
> > > same pa
> > >  test_expect_success 'submodule add places git-dir in superprojects 
> > > git-dir' '
> > > (cd super &&
> > >  mkdir deeper &&
> > > -git submodule add ../submodule deeper/submodule &&
> > > +(cd deeper &&
> > > + git submodule add ../../submodule submodule
> > > +) &&
> > >  (cd deeper/submodule &&
> > >   git log > ../../expected
> > >  ) &&
> > > ---8<-
> > 
> > Ah, ok.  I think this case is problematic because the repository
> > argument is either relative to "remote.origin.url" or to the top of the
> > working tree if there is no "origin" remote.  I wonder if we should just
> > die when a relative path is given for the repository and we're not at
> > the top of the working tree.
> 
> Why not behave as if we are at the top of the working tree for relative
> paths? If there is an origin remote thats fine. If there is no origin
> remote you could warn that the path used is taken relative from the root
> of the superproject during add. What do you think?

That's what the patch currently queued on "pu" does, which Jens wants to
change, isn't it?
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v2 5/8] sequencer: run post-rewrite hook

2013-06-04 Thread Thomas Rast

Junio C Hamano  writes:

> Felipe Contreras  writes:
>
>> +static void run_rewrite_hook(void)
>> +{
>> +struct strbuf buf = STRBUF_INIT;
>> +struct child_process proc;
>> +const char *argv[3];
>> +int code, i;
>> +
>> +argv[0] = find_hook("post-rewrite");
>> +if (!argv[0])
>> +return;
>> +
>> +argv[1] = "rebase";
>> +argv[2] = NULL;
>
> When the end-user action is "git cherry-pick A..B", shouldn't
> the rewrite hook be called with "cherry-pick" not "rebase" as its
> first argument?

The whole function was copied from builtin/commit.c except for that
string.  So it should be refactored anyway, with the "cherry-pick" being
an argument.

-- 
Thomas Rast
trast@{inf,student}.ethz.ch
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Git status reports untracked on tracked files

2013-06-04 Thread Andrey Kiyanovsky

I have tried Git 1.8.3 for Windows. Case is fixed. Thank you very much!

2013/6/4 Jeff King :
> On Wed, May 29, 2013 at 11:40:56AM +0300, Andrey Kiyanovsky wrote:
>
>> Git version 1.8.1.2. for Windows
>>
>> Git config:
>>
>> [core]
>>   repositoryformatversion = 0
>>   filemode = false
>>   bare = false
>>   logallrefupdates = true
>>   symlinks = false
>>   ignorecase = true
>>   hideDotFiles = dotGitOnly
>>   compression = 1
>
> In the past there have been some problems with status listings of
> untracked files when core.ignorecase is in use. I fixed some cases with
> a commit that went into v1.7.8, but some problems remained. Karsten
> Blees (cc'd) did some work that went into git v1.8.1.6, but I do not
> know off-hand if it would fix your case or not.
>
> Can you try with a more recent version of git?
>
> -Peff
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

SNI (SSL virtual hosts)

2013-06-04 Thread Janusz Harkot

I was trying to to a push some repo over https and after few unsuccessful tries 
I've managed to find a problem - multiple virtual SSL servers on one IP address…

Strange was, that initial communication was OK (http GET), but when there was 
http POST - git reported error (incorrect certificate).
The only workaround was to disable certificate verification.

My question is: does git support SNI on the https? If so - are there 
(undocumented) options to make it work?



Thanks!
Janusz  

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Verbessern Sie den persönlichen Geschmack an lebensnotwendigen Gütern

2013-06-04 Thread Nanken

Leichtathletik Nike Air Jordan
begann, weil Leinwand
besten und Gummisohlen Schuhe oder Stiefel mit zu gesehen worden, weil
Turnschuhe. US Rubber die Firma Keds verwendet, um die primären sehr
Turnschuhe im Jahr 1917 zu vermarkten. Generell Turnschuhe über eine
erhebliche Menge an Entscheidungsträger aufgebaut wurden nur tatsächlichen
Ist Sportschuhen in Acht auf dem Markt, während mit United sagen ideal bis
vor die bisher aufgrund 1950er Jahren jeder einzelne und jedes Mal, wenn ein
Mitmensch in der Hochschulbildung mit Oregon Track Gruppe über die
Unterlegenheit der Turnschuhe für die Arbeit gesprochen und zusätzlich
Ausbildung im Zusammenhang .
Im Jahr 1972, Ritter eine Werbe-und Marketing-Trick, die es einfacher für
bestimmte Nike in Bezug auf den Plan erstellt genutzt. Das Motto der Werbung
darauf hin, dass 4 von 7 Athleten, um die Anforderungen zu erfüllen, auch
unter Ausnutzung laufen Vereinigten sagt Olympischen Tests Ausdruck, dass
die Schuhe.
Die meisten der sportlich Produkt wurde entwickelt, um Komfort für die
Läufer zu versorgen. Allerdings ist die Herausforderung, wie eine richtige
Paar von diesen unter all den oberen Ende Malerei Produkte wählen? Die
Antwort ist einfach. Wählen Sie einen Satz, der die Fuß-Typ passt .
Sollten Sie sich entscheiden wollen entsprechende Fuß-Zubehör für den Zweck
des fließenden wählen, dann sind die Chancen sollten Sie versuchen, zu Fuß
in Sport am Nachmittag oder Abend möglicherweise kaufen da für die Dauer des
Tages Zeit weg Füße haben Neigung zu schwellen. Für mit gültiges Maß aller
Ihre Fuß-Typ, werden Sie wollen immer bestimmen die Art Fuß während der
Wartezeit.
In einer Art Nike Air Max 1 Sale
von Sport
in der Tat gibt es tatsächlich Schuhe für mehrere Ebenen im Zusammenhang
gewährt mit Wettbewerb und auch Körper Arten. Die Zunahme im Bereich der
weiblichen Athleten kürzlich Jahren motivierte Unternehmen bekommen hat
Studien und Versorgung Schuhe für ihre speziellen Anforderungen.
Darüber hinaus mit dem Test sicher, dass mit den echten Werten im
Zusammenhang stehen und zusätzlich der Nike Kostenlose three.0 V4 Männer
Fahrzeug Verkäufern. Ideen, überprüfen Sie ein paar Lesung
Einzel-Assessments an der Spitze von Mitteln, die helfen soll die Prüfung
eines einzigartigen Erlebnissen inmitten der Händler. Einzelheiten zu diesem
Ballett Schuhe Produkts wurden möglicherweise für den Nike erhoben werden
traditionellen Web-Adresse.
In der Tat gibt es keine innere Kurve für den Fall der flachen Fuß, während
eine hohe Bogenspannung ist deutlich sichtbar in dem Fall eines stark
gewölbten Fuß. Provider mit Sport-Zubehör im Zusammenhang mit Recht diese
Variationen in Fuß Arten gelobt und auch Malerei Produkte entsprechend für
mehrere Fußtypen.
Schuhe Enthusiasten aus Ländern weltweit genannt AIR Jordan Schuhe, am Ende
wird die "Sammler Turnschuhe". "Die größten Golfball shoes" wurde angewandt,
um AIR Jordan XX3, die das Herz, dass Männer Nike Schuhe günstig
und Frauen würden ihre Augen auf zu
beheben hat zu beschreiben. Es endete als bekannt, die eine Menge Leute
haben durch die Geschichte von "23" wurde daher um wunderbare Heimkehr
wieder lassen, Nike 706 Raum könnte "Casting-Legende", um Menschen
aufrichtig zu begehen.

-
nike schuhe herren
nike schuhe online
nike schuhe günstig
--
View this message in context:
http://git.661346.n2.nabble.com/Verbessern-Sie-den-personlichen-Geschmack-an-lebensnotwendigen-Gutern-tp7587859.html
Sent from the git mailing list archive at Nabble.com.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Nike - die Eleganz von innen heraus

2013-06-04 Thread Nanken

Eine größere  nike blazer low
   Anzahl von
Personen in der Erde nutzen irgendeine Art von Turnschuhen täglich. Es gibt
eine große Auswahl mit Bezug zur Auswahl. Die historische Vergangenheit von
Turnschuhen hat eine große Vielzahl von Änderungen in den Arten und Arten,
die getragen werden, gesehen. Ändern besitzen vor allem der Fall für die
Leichtathletik-Schuhe oder Stiefel. Es gibt unglaublich viele, die sie
fühlen können stoßen ist-Shops, die nur ihnen.
Ob eine Einzelperson sind ein Pro Runner oder vielleicht ein Morgen Jogger,
dem oberen Ende Betriebssystem Sportschuhen immer im Einzelfall zu ziehen,
da das Paar verbessern die Menge Komfort beim fließt. Athletisch Fuß für die
Ausübung kommen, um großartig zu irgendeiner Art von Athlet praktisch egal,
ob Ihr Mann ist in Marathon teilnehmen oder möglicherweise irgendeine Art
von kurzfristig zu unterstützen. Wenn im Falle eines Sportlers ist äußerst
angenehm, wenn fließenden, dass pr ihre sexuelle Leistung könnte automatisch
deutlich verbessern. Der sportliche Fuß trägt dazu beitragen, bei der
Vermehrung Leistung Menge, wie sie riesige Komfort für die Sportler bieten.
Aber können diese Produkte auch verschlechtern funktionalen
Performance-Level, wenn irgendeine Art von Athlet Einkäufe eine Reihe von
falschen Produkte für Sie oder Ihn.
Der Anstieg der Relaxationszeit in den frühen zwanzigsten Jahrhunderts fiel
mit Produktionsverfahren, die für wachsenden Palette von Turnschuhen
ermöglichen würde. Die folgenden Jahrzehnten führte zu einem starken Anstieg
in der Qualität und specialitzation der Schuhe für die Anzeige Bemühungen.
Eine individuelle Zusammenhang mit hundert vor vielen Jahren würde sich
gestresst fühlen, was genau ist heute zugänglich in den Markt.
Die  nike free 3.0 v3
   Entscheidung über
die besten Turnschuhe zum Joggen stützt sich auf welche Art von Fuß haben
Sie. Ihre beste beste Freundin kann wie ein und zusätzlich können Sie es
hassen. Zahlreiche Einzelhändler können heute ermutigen, zu kommen und es
ihnen ermöglichen, testen Sie Ihre Gangart. Dies ist sicherlich ein großer
Schritt, egal in welchem Sie kaufen die Schuhe oder Stiefel im Wesentlichen.
Ein großes Paar Schuhe sollten bequem letzten rund 500 Meilen.
Die Notwendigkeit für eine Boot verschiedener fiel zusammen mit der
fortschreitenden tech und Kenntnisse der Körper Techniker. Verschiedene
Sportarten haben unterschiedliche Bedürfnisse in Bezug auf ihre Schuhe. Ein
Schuh für das Spiel des Tennis heute erstellt werden zunehmend ganz anders,
wenn sie die für Basketball-Spieler verglichen. Tatsächlich gibt es
unterschiedliche Anforderungen für die Verwaltung und auch materielle
Auswirkungen Highlights.
Um die Technologien von Nike TN verstehen, müssen wir allerersten Blick in
ersten Kreationen von Nike: Nike Air, Air Max und Focus-Air. Nike Air war
die erste Luft-Technologie im Jahr 1979 von Nike für die höchstmögliche
Dämpfung entwickelt. Der unter Druck stehende Kraftstoff in einem flexiblen
Urethan Kunststoff Pille unter der Ferse positioniert, Vorfuß oder
vielleicht beides zu geringeren Auswirkungen auf den Fuß. Nach entwickelte
Nike ist Air Max Reihe zusammen mit dem sichtbaren Luft Vorrichtung
innerhalb der Ferse. Das Boot landete geschaffen, um die Hauptlast der hart,
häufige Einflüsse widerstehen. Früher oder später wird eine feinere und auch
robuste Polsterung System, Zoom Air, geschaffen worden war, um eine gute
Dämpfung zu liefern. Später hatte der TN oder möglicherweise Tuned
Air-Technologie auf den Markt als eine Strategie mit einzelnen Schoten unter
dem Stiefel gemeint Einflüsse an jedem der Bereiche des Fußes absorbieren
Zusammenhang gelehrt.
Es besitzt  nike blazer high
   ein
Ausgangspunkt 1mm leichter diskriminieren der Nike Kostenlose fünf., Könnten
die Menschen nicht erklären, das ist an sich, obwohl es eine nach oben geht
zu einem großen Teil wollen Bewegung in den konstanten Mechanik und wie der
Kurier scheint beim Betrieb. Dies ist sicherlich voll, weil es die Variante
mit Nike Free four.0 die bounder zusammen mit den Zehen erzeugt damit eine
größere Übung soweit nichts ändert Ihre Beine.



-
nike schuhe herren 
nike schuhe online 
nike schuhe günstig 
--
View this message in context: 
http://git.661346.n2.nabble.com/Nike-die-Eleganz-von-innen-heraus-tp7587860.html
Sent from the git mailing list archive at Nabble.com.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: SNI (SSL virtual hosts)

2013-06-04 Thread Daniel Stenberg


On Tue, 4 Jun 2013, Janusz Harkot wrote:

Strange was, that initial communication was OK (http GET), but when there 
was http POST - git reported error (incorrect certificate). The only 
workaround was to disable certificate verification.


My question is: does git support SNI on the https? If so - are there 
(undocumented) options to make it work?


It does. git uses libcurl for the HTTPS parts and it has support SNI for a 
long time, assuming you built libcurl with a TLS library that handles it.


Which libcurl version and SSL backend is this? (curl -V usually tells)

If you made it working by disabling certificate verification then it sounds as 
if SNI might still have worked and the problem was rahter something else, as 
without SNI you can't do name-based virtual hosting over HTTPS - but perhaps 
you wanted to communicate with the "default" server on that IP?


--

 / daniel.haxx.se
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

fetch delta resolution vs. checkout (was: java zlib woes)

2013-06-04 Thread Andreas Krey

On Mon, 27 May 2013 06:11:46 +, Andreas Krey wrote:
...
> 
> I now have a full test case (involving a generated repo just shy of 1GB)
> that will reproduce that hang. Will look up the existing jgit bug to
> report there.

On https://bugs.eclipse.org/bugs/show_bug.cgi?id=394078

A question: The delta decoding. If I understand correctly,
git and jgit do verify the packfile content after fetching/cloning,
and need to resolve any deltified files in the pack.

And when checking out a commit it needs this to again for the
files that are being checked out?

Because we now have the phenomenon that the packfile is fetched
ok, but a checkout then hangs (100%) CPU on one of the large files,
and on one that should, according to core.bigfilethreshold, not
even be deltified.

(Setting core.bigfilethreshold to 20m in the source repo (C git)
gets jgit to no longer hang in the fetch/delta resolution phase.
And it doesn't look like jgit would repack the pack file, and
uses it as it was received plus 20 bytes at the end.)

Andreas

-- 
"Totally trivial. Famous last words."
From: Linus Torvalds 
Date: Fri, 22 Jan 2010 07:29:21 -0800
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: SNI (SSL virtual hosts)

2013-06-04 Thread Janusz Harkot

> It does. git uses libcurl for the HTTPS parts and it has support SNI for a 
> long time, assuming you built libcurl with a TLS library that handles it.
> 
> Which libcurl version and SSL backend is this? (curl -V usually tells)
$ curl -V
curl 7.24.0 (x86_64-apple-darwin12.0) libcurl/7.24.0 OpenSSL/0.9.8r zlib/1.2.5
Protocols: dict file ftp ftps gopher http https imap imaps ldap ldaps pop3 
pop3s rtsp smtp smtps telnet tftp 
Features: AsynchDNS GSS-Negotiate IPv6 Largefile NTLM NTLM_WB SSL libz 

$ otool -L /usr/local/bin/git
/usr/local/bin/git:
/usr/lib/libz.1.dylib (compatibility version 1.0.0, current version 1.2.5)
/usr/lib/libiconv.2.dylib (compatibility version 7.0.0, current version 7.0.0)
/usr/local/opt/openssl/lib/libcrypto.1.0.0.dylib (compatibility version 1.0.0, 
current version 1.0.0)
/usr/local/opt/openssl/lib/libssl.1.0.0.dylib (compatibility version 1.0.0, 
current version 1.0.0)
/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 
169.3.0)



> If you made it working by disabling certificate verification then it sounds 
> as 
> if SNI might still have worked and the problem was rahter something else, as 
> without SNI you can't do name-based virtual hosting over HTTPS - but perhaps 
> you wanted to communicate with the "default" server on that IP?

here is a log (with GIT_CURL_VERBOSE=1)

https://gist.github.com/anonymous/8f6533a755ae5c710c75 

Initial connection is correct (line 10 - shows that it reads correct 
certificate),
 but then subsequent call to the server (line 68) shows that the defat server 
certificate is used.

It looks like the second call was without hostname (?).

Thanks!
Janusz

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH] git-remote-mediawiki: use git.pm functions for credentials

2013-06-04 Thread benoit . person

From: Benoit Person 

In 52dce6d, a new credential function was added to git.pm, based on
git-remote-mediawiki's functions. The logical follow-up is to use
those functions in git-remote-mediawiki.

Signed-off-by: Benoit Person 
Signed-off-by: Matthieu Moy 
---
 contrib/mw-to-git/git-remote-mediawiki.perl | 66 -
 1 file changed, 9 insertions(+), 57 deletions(-)

diff --git a/contrib/mw-to-git/git-remote-mediawiki.perl 
b/contrib/mw-to-git/git-remote-mediawiki.perl
index 9c14c1f..9fb281e 100755
--- a/contrib/mw-to-git/git-remote-mediawiki.perl
+++ b/contrib/mw-to-git/git-remote-mediawiki.perl
@@ -13,6 +13,7 @@
 
 use strict;
 use MediaWiki::API;
+use Git;
 use DateTime::Format::ISO8601;
 
 # By default, use UTF-8 to communicate with Git and the user
@@ -156,57 +157,6 @@ while () {
 
 ## Functions ##
 
-## credential API management (generic functions)
-
-sub credential_read {
-   my %credential;
-   my $reader = shift;
-   my $op = shift;
-   while (<$reader>) {
-   my ($key, $value) = /([^=]*)=(.*)/;
-   if (not defined $key) {
-   die "ERROR receiving response from git credential 
$op:\n$_\n";
-   }
-   $credential{$key} = $value;
-   }
-   return %credential;
-}
-
-sub credential_write {
-   my $credential = shift;
-   my $writer = shift;
-   # url overwrites other fields, so it must come first
-   print $writer "url=$credential->{url}\n" if exists $credential->{url};
-   while (my ($key, $value) = each(%$credential) ) {
-   if (length $value && $key ne 'url') {
-   print $writer "$key=$value\n";
-   }
-   }
-}
-
-sub credential_run {
-   my $op = shift;
-   my $credential = shift;
-   my $pid = open2(my $reader, my $writer, "git credential $op");
-   credential_write($credential, $writer);
-   print $writer "\n";
-   close($writer);
-
-   if ($op eq "fill") {
-   %$credential = credential_read($reader, $op);
-   } else {
-   if (<$reader>) {
-   die "ERROR while running git credential $op:\n$_";
-   }
-   }
-   close($reader);
-   waitpid($pid, 0);
-   my $child_exit_status = $? >> 8;
-   if ($child_exit_status != 0) {
-   die "'git credential $op' failed with code $child_exit_status.";
-   }
-}
-
 # MediaWiki API instance, created lazily.
 my $mediawiki;
 
@@ -217,22 +167,24 @@ sub mw_connect_maybe {
$mediawiki = MediaWiki::API->new;
$mediawiki->{config}->{api_url} = "$url/api.php";
if ($wiki_login) {
-   my %credential = (url => $url);
-   $credential{username} = $wiki_login;
-   $credential{password} = $wiki_passwd;
-   credential_run("fill", \%credential);
+   my %credential = (
+   'url' => $url,
+   'username' => $wiki_login,
+   'password' => $wiki_passwd
+   );
+   Git::credential \%credential;
my $request = {lgname => $credential{username},
   lgpassword => $credential{password},
   lgdomain => $wiki_domain};
if ($mediawiki->login($request)) {
-   credential_run("approve", \%credential);
+   Git::credential \%credential, 'approve';
print STDERR "Logged in mediawiki user 
\"$credential{username}\".\n";
} else {
print STDERR "Failed to log in mediawiki user 
\"$credential{username}\" on $url\n";
print STDERR "  (error " .
$mediawiki->{error}->{code} . ': ' .
$mediawiki->{error}->{details} . ")\n";
-   credential_run("reject", \%credential);
+   Git::credential \%credential, 'reject';
exit 1;
}
}
-- 
1.8.3.rc3.7.gc2f33ed.dirty

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Re: Re: What's cooking in git.git (May 2013, #09; Wed, 29)

2013-06-04 Thread Heiko Voigt

On Tue, Jun 04, 2013 at 09:10:45AM +0100, John Keeping wrote:
> On Tue, Jun 04, 2013 at 03:29:51PM +1000, Heiko Voigt wrote:
> > On Mon, Jun 03, 2013 at 11:23:41PM +0100, John Keeping wrote:
> > > > Sorry, I should have been more specific here. I saw that you did some
> > > > changes to make "submodule add" do the right thing with relative paths,
> > > > but the following change to t7406 does not work like I believe it
> > > > should but instead makes the test fail:
> > > > ---8<-
> > > > diff --git a/t/t7406-submodule-update.sh b/t/t7406-submodule-update.sh
> > > > index a4ffea0..9766b9e 100755
> > > > --- a/t/t7406-submodule-update.sh
> > > > +++ b/t/t7406-submodule-update.sh
> > > > @@ -559,7 +559,9 @@ test_expect_success 'add different submodules to 
> > > > the same pa
> > > >  test_expect_success 'submodule add places git-dir in superprojects 
> > > > git-dir' '
> > > > (cd super &&
> > > >  mkdir deeper &&
> > > > -git submodule add ../submodule deeper/submodule &&
> > > > +(cd deeper &&
> > > > + git submodule add ../../submodule submodule
> > > > +) &&
> > > >  (cd deeper/submodule &&
> > > >   git log > ../../expected
> > > >  ) &&
> > > > ---8<-
> > > 
> > > Ah, ok.  I think this case is problematic because the repository
> > > argument is either relative to "remote.origin.url" or to the top of the
> > > working tree if there is no "origin" remote.  I wonder if we should just
> > > die when a relative path is given for the repository and we're not at
> > > the top of the working tree.
> > 
> > Why not behave as if we are at the top of the working tree for relative
> > paths? If there is an origin remote thats fine. If there is no origin
> > remote you could warn that the path used is taken relative from the root
> > of the superproject during add. What do you think?
> 
> That's what the patch currently queued on "pu" does, which Jens wants to
> change, isn't it?

True I did not realize this when reading it the first time. But I think
we should still not die when in a subdirectory. After all this series is
trying to archive that the submodule command works in subdirectories
seamlessly right? So you probably want to translate a relative path
without "origin" remote given from a subdirectory to the superproject
level and use that. Then you do not have to die.

Cheers Heiko
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: SNI (SSL virtual hosts)

2013-06-04 Thread Daniel Stenberg


On Tue, 4 Jun 2013, Janusz Harkot wrote:


Which libcurl version and SSL backend is this? (curl -V usually tells)

$ curl -V
curl 7.24.0 (x86_64-apple-darwin12.0) libcurl/7.24.0 OpenSSL/0.9.8r 
zlib/1.2.5


From what I can tell, that OpenSSL version supports SNI fine and libcurl has 

supported it since 7.18.1.


here is a log (with GIT_CURL_VERBOSE=1)

https://gist.github.com/anonymous/8f6533a755ae5c710c75

Initial connection is correct (line 10 - shows that it reads correct 
certificate), but then subsequent call to the server (line 68) shows that 
the defat server certificate is used.


It looks like the second call was without hostname (?).


What makes you suggest that's what's happening? Sure, if it would've sent no 
or the wrong host name it would probably have that effect.


Any chance you can snoop on the network and the SSL handshake to see who's to 
blame? I can't but to think that is is a very common use case!


--

 / daniel.haxx.se
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 00/15] Towards a more awesome git-branch

Hi,

Duy and I have been working on this topic for some time now.  Here's a
review candidate.  Duy did most of the chunky work, and I mostly
did review/documentation.  The key patches are:

[5/15]  is a brilliant patch that made this entire thing possible.
[10/15] is another brilliant patch to auto-calculate widths.

Duy is currently writing code to factor out -v[v] and adding --sort,
--count to git-branch, but I'm having a lot of difficulty working with
a series of this size.  I'm leaning towards getting this merged before
tackling branch (who wants to review a 40-part series?).  Especially
after [14/15] and [15/15], git for-each-ref should be usable in its
own right.  I currently have:

[pretty]
hot = %C(red)%(HEAD)%C(reset) 
%C(green)%(refname:short)%C(reset)%(upstream:trackshort)
[alias]
hot = for-each-ref --pretty=hot --count 10 --sort='-committerdate' refs/heads

Which is really useful and manageable until we port these options to
branch and get nice configuration variables.

Nguyễn Thái Ngọc Duy (8):
  for-each-ref, quote: convert *_quote_print -> *_quote_buf
  for-each-ref: don't print out elements directly
  pretty: extend pretty_print_context with callback
  pretty: allow passing NULL commit to format_commit_message()
  for-each-ref: get --pretty using format_commit_message
  for-each-ref: teach verify_format() about pretty's syntax
  for-each-ref: introduce format specifier %>(*) and %<(*)
  for-each-ref: improve responsiveness of %(upstream:track)

Ramkumar Ramachandra (7):
  tar-tree: remove dependency on sq_quote_print()
  quote: remove sq_quote_print()
  pretty: limit recursion in format_commit_one()
  for-each-ref: introduce %(HEAD) marker
  for-each-ref: introduce %(upstream:track[short])
  pretty: introduce get_pretty_userformat
  for-each-ref: use get_pretty_userformat in --pretty

 Documentation/git-for-each-ref.txt |  42 +-
 builtin/for-each-ref.c | 274 ++---
 builtin/tar-tree.c |  11 +-
 commit.h   |   9 ++
 pretty.c   |  77 ++-
 quote.c|  61 +++--
 quote.h|   8 +-
 7 files changed, 372 insertions(+), 110 deletions(-)

-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 01/15] for-each-ref, quote: convert _quote_print -> _quote_buf

From: Nguyễn Thái Ngọc Duy 

for-each-ref.c:print_value() currently prints values to stdout
immediately using {sq|perl|python|tcl}_quote_print, giving us no
opportunity to do any further processing.  In preparation for getting
print_value() to accept an additional strbuf argument to write to,
convert the *_quote_print functions and callers to *_quote_buf.

[rr: commit message, minor modifications]

Signed-off-by: Ramkumar Ramachandra 
---
 builtin/for-each-ref.c | 13 +
 quote.c| 44 ++--
 quote.h|  6 +++---
 3 files changed, 34 insertions(+), 29 deletions(-)

diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index 7f059c3..1d4083c 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -867,24 +867,29 @@ static void sort_refs(struct ref_sort *sort, struct 
refinfo **refs, int num_refs
 static void print_value(struct refinfo *ref, int atom, int quote_style)
 {
struct atom_value *v;
+   struct strbuf sb = STRBUF_INIT;
get_value(ref, atom, &v);
switch (quote_style) {
case QUOTE_NONE:
fputs(v->s, stdout);
break;
case QUOTE_SHELL:
-   sq_quote_print(stdout, v->s);
+   sq_quote_buf(&sb, v->s);
break;
case QUOTE_PERL:
-   perl_quote_print(stdout, v->s);
+   perl_quote_buf(&sb, v->s);
break;
case QUOTE_PYTHON:
-   python_quote_print(stdout, v->s);
+   python_quote_buf(&sb, v->s);
break;
case QUOTE_TCL:
-   tcl_quote_print(stdout, v->s);
+   tcl_quote_buf(&sb, v->s);
break;
}
+   if (quote_style != QUOTE_NONE) {
+   fputs(sb.buf, stdout);
+   strbuf_release(&sb);
+   }
 }
 
 static int hex1(char ch)
diff --git a/quote.c b/quote.c
index 911229f..8c294df 100644
--- a/quote.c
+++ b/quote.c
@@ -463,72 +463,72 @@ int unquote_c_style(struct strbuf *sb, const char 
*quoted, const char **endp)
 
 /* quoting as a string literal for other languages */
 
-void perl_quote_print(FILE *stream, const char *src)
+void perl_quote_buf(struct strbuf *sb, const char *src)
 {
const char sq = '\'';
const char bq = '\\';
char c;
 
-   fputc(sq, stream);
+   strbuf_addch(sb, sq);
while ((c = *src++)) {
if (c == sq || c == bq)
-   fputc(bq, stream);
-   fputc(c, stream);
+   strbuf_addch(sb, bq);
+   strbuf_addch(sb, c);
}
-   fputc(sq, stream);
+   strbuf_addch(sb, sq);
 }
 
-void python_quote_print(FILE *stream, const char *src)
+void python_quote_buf(struct strbuf *sb, const char *src)
 {
const char sq = '\'';
const char bq = '\\';
const char nl = '\n';
char c;
 
-   fputc(sq, stream);
+   strbuf_addch(sb, sq);
while ((c = *src++)) {
if (c == nl) {
-   fputc(bq, stream);
-   fputc('n', stream);
+   strbuf_addch(sb, bq);
+   strbuf_addch(sb, 'n');
continue;
}
if (c == sq || c == bq)
-   fputc(bq, stream);
-   fputc(c, stream);
+   strbuf_addch(sb, bq);
+   strbuf_addch(sb, c);
}
-   fputc(sq, stream);
+   strbuf_addch(sb, sq);
 }
 
-void tcl_quote_print(FILE *stream, const char *src)
+void tcl_quote_buf(struct strbuf *sb, const char *src)
 {
char c;
 
-   fputc('"', stream);
+   strbuf_addch(sb, '"');
while ((c = *src++)) {
switch (c) {
case '[': case ']':
case '{': case '}':
case '$': case '\\': case '"':
-   fputc('\\', stream);
+   strbuf_addch(sb, '\\');
default:
-   fputc(c, stream);
+   strbuf_addch(sb, c);
break;
case '\f':
-   fputs("\\f", stream);
+   strbuf_addstr(sb, "\\f");
break;
case '\r':
-   fputs("\\r", stream);
+   strbuf_addstr(sb, "\\r");
break;
case '\n':
-   fputs("\\n", stream);
+   strbuf_addstr(sb, "\\n");
break;
case '\t':
-   fputs("\\t", stream);
+   strbuf_addstr(sb, "\\t");
break;
case '\v':
-   fputs("\\v", stream);
+   strbuf_addstr(sb, "\\v");
break;
}
}
-   fputc('"', stream

[PATCH 09/15] for-each-ref: teach verify_format() about pretty's syntax

From: Nguyễn Thái Ngọc Duy 

Pretty format accepts either ' ', '+' or '-' after '%' and before the
placeholder name to modify certain behaviors. Teach verify_format()
about this so that it finds atom "upstream" in, for example,
'% (upstream)'. This is important because verify_format populates
used_atom, which get_value() and populate_value() later rely on.

Signed-off-by: Ramkumar Ramachandra 
---
 builtin/for-each-ref.c | 15 +--
 pretty.c   |  4 
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index 576a882..f8a3880 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -150,7 +150,7 @@ static int parse_atom(const char *atom, const char *ep)
 /*
  * In a format string, find the next occurrence of %(atom).
  */
-static const char *find_next(const char *cp)
+static const char *find_next(const char *cp, int pretty)
 {
while (*cp) {
if (*cp == '%') {
@@ -160,6 +160,9 @@ static const char *find_next(const char *cp)
 */
if (cp[1] == '(')
return cp;
+   else if (pretty && cp[1] && cp[2] == '(' &&
+strchr(" +-", cp[1])) /* see 
format_commit_item() */
+   return cp + 1;
else if (cp[1] == '%')
cp++; /* skip over two % */
/* otherwise this is a singleton, literal % */
@@ -173,10 +176,10 @@ static const char *find_next(const char *cp)
  * Make sure the format string is well formed, and parse out
  * the used atoms.
  */
-static int verify_format(const char *format)
+static int verify_format(const char *format, int pretty)
 {
const char *cp, *sp;
-   for (cp = format; *cp && (sp = find_next(cp)); ) {
+   for (cp = format; *cp && (sp = find_next(cp, pretty)); ) {
const char *ep = strchr(sp, ')');
if (!ep)
return error("malformed format string %s", sp);
@@ -935,7 +938,7 @@ static void show_ref(struct strbuf *sb, struct refinfo 
*info,
 {
const char *cp, *sp, *ep;
 
-   for (cp = format; *cp && (sp = find_next(cp)); cp = ep + 1) {
+   for (cp = format; *cp && (sp = find_next(cp, 0)); cp = ep + 1) {
ep = strchr(sp, ')');
if (cp < sp)
emit(sb, cp, sp);
@@ -1093,8 +1096,8 @@ int cmd_for_each_ref(int argc, const char **argv, const 
char *prefix)
}
if (format != default_format && pretty)
die("--format and --pretty cannot be used together");
-   if ((pretty && verify_format(pretty)) ||
-   (!pretty && verify_format(format)))
+   if ((pretty && verify_format(pretty, 1)) ||
+   (!pretty && verify_format(format, 0)))
usage_with_options(for_each_ref_usage, opts);
 
if (!sort)
diff --git a/pretty.c b/pretty.c
index 816aa32..28c0a72 100644
--- a/pretty.c
+++ b/pretty.c
@@ -1439,6 +1439,10 @@ static size_t format_commit_item(struct strbuf *sb, /* 
in UTF-8 */
ADD_SP_BEFORE_NON_EMPTY
} magic = NO_MAGIC;
 
+   /*
+* Note: any modification in what placeholder[0] contains
+* should be redone in builtin/for-each-ref.c:find_next().
+*/
switch (placeholder[0]) {
case '-':
magic = DEL_LF_BEFORE_EMPTY;
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 06/15] pretty: limit recursion in format_commit_one()

To make sure that a pretty_ctx->format substitution doesn't result in an
infinite recursion, change the prototype of format_commit_one() to
accept one last argument: no_recurse.  So, a single substitution by
format() must yield a result that can be parsed by format_commit_one()
without the help of format().

Signed-off-by: Ramkumar Ramachandra 
---
 pretty.c | 11 ++-
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/pretty.c b/pretty.c
index 095e5ba..0063f2d 100644
--- a/pretty.c
+++ b/pretty.c
@@ -1061,7 +1061,8 @@ static size_t parse_padding_placeholder(struct strbuf *sb,
 
 static size_t format_commit_one(struct strbuf *sb, /* in UTF-8 */
const char *placeholder,
-   void *context)
+   void *context,
+   int no_recurse)
 {
struct format_commit_context *c = context;
const struct commit *commit = c->commit;
@@ -1069,7 +1070,7 @@ static size_t format_commit_one(struct strbuf *sb, /* in 
UTF-8 */
struct commit_list *p;
int h1, h2;
 
-   if (c->pretty_ctx->format) {
+   if (!no_recurse && c->pretty_ctx->format) {
struct strbuf subst = STRBUF_INIT;
int ret = c->pretty_ctx->format(sb, placeholder, context,
c->pretty_ctx->user_data,
@@ -1083,7 +1084,7 @@ static size_t format_commit_one(struct strbuf *sb, /* in 
UTF-8 */
 * ret: the length of the original string
 * before substitution.
 */
-   ret = format_commit_one(sb, subst.buf, context) ? ret : 
0;
+   ret = format_commit_one(sb, subst.buf, context, 1) ? 
ret : 0;
strbuf_release(&subst);
return ret;
} else if (ret)
@@ -1332,7 +1333,7 @@ static size_t format_and_pad_commit(struct strbuf *sb, /* 
in UTF-8 */
}
while (1) {
int modifier = *placeholder == 'C';
-   int consumed = format_commit_one(&local_sb, placeholder, c);
+   int consumed = format_commit_one(&local_sb, placeholder, c, 0);
total_consumed += consumed;
 
if (!modifier)
@@ -1452,7 +1453,7 @@ static size_t format_commit_item(struct strbuf *sb, /* in 
UTF-8 */
if (((struct format_commit_context *)context)->flush_type != no_flush)
consumed = format_and_pad_commit(sb, placeholder, context);
else
-   consumed = format_commit_one(sb, placeholder, context);
+   consumed = format_commit_one(sb, placeholder, context, 0);
if (magic == NO_MAGIC)
return consumed;
 
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 07/15] pretty: allow passing NULL commit to format_commit_message()

From: Nguyễn Thái Ngọc Duy 

The new formatter, for-each-ref, may use non-commit placeholders only.
While it could audit the format line and warn/exclude commit
placeholders, that's a lot more work than simply ignore them.
Unrecognized placeholders are displayed as-is, pretty obvious that they
are not handled.

Signed-off-by: Ramkumar Ramachandra 
---
 pretty.c | 16 
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/pretty.c b/pretty.c
index 0063f2d..816aa32 100644
--- a/pretty.c
+++ b/pretty.c
@@ -1156,6 +1156,9 @@ static size_t format_commit_one(struct strbuf *sb, /* in 
UTF-8 */
}
 
/* these depend on the commit */
+   if (!commit)
+   return 0;
+
if (!commit->object.parsed)
parse_object(commit->object.sha1);
 
@@ -1276,6 +1279,9 @@ static size_t format_commit_one(struct strbuf *sb, /* in 
UTF-8 */
}
 
 
+   if (!c->message)
+   return 0;
+
/* For the rest we have to parse the commit header. */
if (!c->commit_header_parsed)
parse_commit_header(c);
@@ -1510,9 +1516,10 @@ void format_commit_message(const struct commit *commit,
context.commit = commit;
context.pretty_ctx = pretty_ctx;
context.wrap_start = sb->len;
-   context.message = logmsg_reencode(commit,
- &context.commit_encoding,
- output_enc);
+   if (commit)
+   context.message = logmsg_reencode(commit,
+ &context.commit_encoding,
+ output_enc);
 
strbuf_expand(sb, format, format_commit_item, &context);
rewrap_message_tail(sb, &context, 0, 0, 0);
@@ -1535,7 +1542,8 @@ void format_commit_message(const struct commit *commit,
}
 
free(context.commit_encoding);
-   logmsg_free(context.message, commit);
+   if (commit)
+   logmsg_free(context.message, commit);
free(context.signature_check.gpg_output);
free(context.signature_check.signer);
 }
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 15/15] for-each-ref: use get_pretty_userformat in --pretty

Use get_pretty_userformat() to interpret the --pretty string.  This
means that leading you can now reference a format specified in a pretty.*
configuration variable as an argument to 'git for-each-ref --pretty='.
There are two caveats:

1. A leading "format:" or "tformat:" is automatically stripped and
   ignored.  Separator semantics are not configurable (yet).

2. No built-in formats are available.  The ones specified in
   pretty-formats (oneline, short etc) don't make sense when displaying
   refs anyway.

Signed-off-by: Ramkumar Ramachandra 
---
 Documentation/git-for-each-ref.txt |  3 +++
 builtin/for-each-ref.c | 16 +---
 2 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/Documentation/git-for-each-ref.txt 
b/Documentation/git-for-each-ref.txt
index 05ff7ba..7f3cba5 100644
--- a/Documentation/git-for-each-ref.txt
+++ b/Documentation/git-for-each-ref.txt
@@ -60,6 +60,9 @@ calculated.
 +
 Caveats:
 
+0. No built-in formats from PRETTY FORMATS (like oneline, short) are
+   available.
+
 1. Many of the placeholders in "PRETTY FORMATS" are designed to work
specifically on commit objects: when non-commit objects are
supplied, those placeholders won't work.
diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index edaa6b2..c00ab05 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -1146,7 +1146,7 @@ int cmd_for_each_ref(int argc, const char **argv, const 
char *prefix)
int num_refs;
const char *default_format = "%(objectname) %(objecttype)\t%(refname)";
const char *format = default_format;
-   const char *pretty = NULL;
+   const char *pretty_raw = NULL, *pretty_userformat = NULL;
struct ref_sort *sort = NULL, **sort_tail = &sort;
int maxcount = 0, quote_style = 0;
struct refinfo **refs;
@@ -1165,13 +1165,15 @@ int cmd_for_each_ref(int argc, const char **argv, const 
char *prefix)
OPT_GROUP(""),
OPT_INTEGER( 0 , "count", &maxcount, N_("show only  matched 
refs")),
OPT_STRING(  0 , "format", &format, N_("format"), N_("format to 
use for the output")),
-   OPT_STRING(  0 , "pretty", &pretty, N_("format"), 
N_("alternative format to use for the output")),
+   OPT_STRING(  0 , "pretty", &pretty_raw, N_("format"), 
N_("alternative format to use for the output")),
OPT_CALLBACK(0 , "sort", sort_tail, N_("key"),
N_("field name to sort on"), &opt_parse_sort),
OPT_END(),
};
 
parse_options(argc, argv, prefix, opts, for_each_ref_usage, 0);
+   if (pretty_raw)
+   pretty_userformat = get_pretty_userformat(pretty_raw);
if (maxcount < 0) {
error("invalid --count argument: `%d'", maxcount);
usage_with_options(for_each_ref_usage, opts);
@@ -1180,10 +1182,10 @@ int cmd_for_each_ref(int argc, const char **argv, const 
char *prefix)
error("more than one quoting style?");
usage_with_options(for_each_ref_usage, opts);
}
-   if (format != default_format && pretty)
+   if (format != default_format && pretty_userformat)
die("--format and --pretty cannot be used together");
-   if ((pretty && verify_format(pretty, 1)) ||
-   (!pretty && verify_format(format, 0)))
+   if ((pretty_userformat && verify_format(pretty_userformat, 1)) ||
+   (!pretty_userformat && verify_format(format, 0)))
usage_with_options(for_each_ref_usage, opts);
 
if (!sort)
@@ -1204,8 +1206,8 @@ int cmd_for_each_ref(int argc, const char **argv, const 
char *prefix)
if (!maxcount || num_refs < maxcount)
maxcount = num_refs;
 
-   if (pretty)
-   show_pretty_refs(refs, maxcount, pretty, quote_style);
+   if (pretty_userformat)
+   show_pretty_refs(refs, maxcount, pretty_userformat, 
quote_style);
else
show_refs(refs, maxcount, format, quote_style);
return 0;
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 03/15] tar-tree: remove dependency on sq_quote_print()

Currently, there is exactly one caller of sq_quote_print(), namely
cmd_tar_tree().  In the interest of removing sq_quote_print() and
simplification, replace it with an equivalent call to sq_quote_argv().
No functional changes intended.

Signed-off-by: Ramkumar Ramachandra 
---
 builtin/tar-tree.c | 11 +--
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/builtin/tar-tree.c b/builtin/tar-tree.c
index 3f1e701..ba3ffe6 100644
--- a/builtin/tar-tree.c
+++ b/builtin/tar-tree.c
@@ -26,8 +26,8 @@ int cmd_tar_tree(int argc, const char **argv, const char 
*prefix)
 * $0 tree-ish basedir ==>
 *  git archive --format-tar --prefix=basedir tree-ish
 */
-   int i;
const char **nargv = xcalloc(sizeof(*nargv), argc + 3);
+   struct strbuf sb = STRBUF_INIT;
char *basedir_arg;
int nargc = 0;
 
@@ -65,11 +65,10 @@ int cmd_tar_tree(int argc, const char **argv, const char 
*prefix)
fprintf(stderr,
"*** \"git tar-tree\" is now deprecated.\n"
"*** Running \"git archive\" instead.\n***");
-   for (i = 0; i < nargc; i++) {
-   fputc(' ', stderr);
-   sq_quote_print(stderr, nargv[i]);
-   }
-   fputc('\n', stderr);
+   sq_quote_argv(&sb, nargv, 0);
+   strbuf_addch(&sb, '\n');
+   fputs(sb.buf, stderr);
+   strbuf_release(&sb);
return cmd_archive(nargc, nargv, prefix);
 }
 
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 14/15] pretty: introduce get_pretty_userformat

This helper function is intended to be used by callers implementing
--pretty themselves; it parses pretty.* configuration variables
recursively and hands the user-defined format back to the caller.  No
builtins are supported, as CMT_FMT_* are really only useful when
displaying commits.  Callers might like to define their own builtins in
the future.

Signed-off-by: Ramkumar Ramachandra 
---
 commit.h |  1 +
 pretty.c | 25 +
 2 files changed, 26 insertions(+)

diff --git a/commit.h b/commit.h
index 04bd935..48424c9 100644
--- a/commit.h
+++ b/commit.h
@@ -113,6 +113,7 @@ extern char *logmsg_reencode(const struct commit *commit,
 const char *output_encoding);
 extern void logmsg_free(char *msg, const struct commit *commit);
 extern void get_commit_format(const char *arg, struct rev_info *);
+extern const char *get_pretty_userformat(const char *arg);
 extern const char *format_subject(struct strbuf *sb, const char *msg,
  const char *line_separator);
 extern void userformat_find_requirements(const char *fmt, struct 
userformat_want *w);
diff --git a/pretty.c b/pretty.c
index 28c0a72..70e4e44 100644
--- a/pretty.c
+++ b/pretty.c
@@ -174,6 +174,31 @@ void get_commit_format(const char *arg, struct rev_info 
*rev)
 }
 
 /*
+ * Function to parse --pretty string, lookup pretty.* configuration
+ * variables and return the format string, assuming no builtin
+ * formats.  Not limited to commits, unlike get_commit_format().
+ */
+const char *get_pretty_userformat(const char *arg)
+{
+   struct cmt_fmt_map *commit_format;
+
+   if (!arg || !*arg)
+   return NULL;
+
+   if (!prefixcmp(arg, "format:") || !prefixcmp(arg, "tformat:"))
+   return xstrdup(strchr(arg, ':' + 1));
+
+   if (strchr(arg, '%'))
+   return xstrdup(arg);
+
+   commit_format = find_commit_format(arg);
+   if (!commit_format || commit_format->format != CMIT_FMT_USERFORMAT)
+   die("invalid --pretty format: %s", arg);
+
+   return xstrdup(commit_format->user_format);
+}
+
+/*
  * Generic support for pretty-printing the header
  */
 static int get_one_line(const char *msg)
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 04/15] quote: remove sq_quote_print()

Remove sq_quote_print() since it has no callers.  A nicer alternative
sq_quote_buf() exists: its callers aren't forced to print immediately.

For historical context, sq_quote_print() was first introduced in
575ba9d6 (GIT_TRACE: show which built-in/external commands are executed,
2006-06-25) for the purpose of printing argv for $GIT_TRACE.  Today, we
achieve this using trace_argv_printf() -> sq_quote_argv() ->
sq_quote_buf(), which ultimately fills in a strbuf.

Signed-off-by: Ramkumar Ramachandra 
---
 quote.c | 17 -
 quote.h |  2 --
 2 files changed, 19 deletions(-)

diff --git a/quote.c b/quote.c
index 8c294df..778b39a 100644
--- a/quote.c
+++ b/quote.c
@@ -42,23 +42,6 @@ void sq_quote_buf(struct strbuf *dst, const char *src)
free(to_free);
 }
 
-void sq_quote_print(FILE *stream, const char *src)
-{
-   char c;
-
-   fputc('\'', stream);
-   while ((c = *src++)) {
-   if (need_bs_quote(c)) {
-   fputs("'\\", stream);
-   fputc(c, stream);
-   fputc('\'', stream);
-   } else {
-   fputc(c, stream);
-   }
-   }
-   fputc('\'', stream);
-}
-
 void sq_quote_argv(struct strbuf *dst, const char** argv, size_t maxlen)
 {
int i;
diff --git a/quote.h b/quote.h
index ed06df5..251f3cc 100644
--- a/quote.h
+++ b/quote.h
@@ -27,8 +27,6 @@ struct strbuf;
  * excluding the final null regardless of the buffer size.
  */
 
-extern void sq_quote_print(FILE *stream, const char *src);
-
 extern void sq_quote_buf(struct strbuf *, const char *src);
 extern void sq_quote_argv(struct strbuf *, const char **argv, size_t maxlen);
 
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 12/15] for-each-ref: introduce %(upstream:track[short])

Introduce %(upstream:track) to display "[ahead M, behind N]" and
%(upstream:trackshort) to display "=", ">", "<", or "<>"
appropriately (inspired by the contrib/completion/git-prompt.sh).

Now you can use the following format in for-each-ref:

  %C(green)%(refname:short)%C(reset)%(upstream:trackshort)

to display refs with terse tracking information.

Note that :track and :trackshort only works with upstream, and errors
out when used with anything else.

Signed-off-by: Ramkumar Ramachandra 
---
 Documentation/git-for-each-ref.txt |  6 +-
 builtin/for-each-ref.c | 42 --
 2 files changed, 45 insertions(+), 3 deletions(-)

diff --git a/Documentation/git-for-each-ref.txt 
b/Documentation/git-for-each-ref.txt
index d529296..05ff7ba 100644
--- a/Documentation/git-for-each-ref.txt
+++ b/Documentation/git-for-each-ref.txt
@@ -118,7 +118,11 @@ objectname::
 upstream::
The name of a local ref which can be considered ``upstream''
from the displayed ref. Respects `:short` in the same way as
-   `refname` above.
+   `refname` above.  Additionally respects `:track` to show
+   "[ahead N, behind M]" and `:trackshort` to show the terse
+   version (like the prompt) ">", "<", "<>", or "=".  Has no
+   effect if the ref does not have tracking information
+   associated with it.
 
 HEAD::
Useful to indicate the currently checked out branch.  Is '*'
diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index b0b8236..adc965e 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -628,6 +628,7 @@ static void populate_value(struct refinfo *ref)
int eaten, i;
unsigned long size;
const unsigned char *tagged;
+   int upstream_present = 0;
 
ref->value = xcalloc(sizeof(struct atom_value), used_atom_cnt);
 
@@ -645,6 +646,7 @@ static void populate_value(struct refinfo *ref)
int deref = 0;
const char *refname;
const char *formatp;
+   struct branch *branch;
 
if (*name == '*') {
deref = 1;
@@ -656,7 +658,6 @@ static void populate_value(struct refinfo *ref)
else if (!prefixcmp(name, "symref"))
refname = ref->symref ? ref->symref : "";
else if (!prefixcmp(name, "upstream")) {
-   struct branch *branch;
/* only local branches may have an upstream */
if (prefixcmp(ref->refname, "refs/heads/"))
continue;
@@ -666,6 +667,7 @@ static void populate_value(struct refinfo *ref)
!branch->merge[0]->dst)
continue;
refname = branch->merge[0]->dst;
+   upstream_present = 1;
}
else if (!strcmp(name, "flag")) {
char buf[256], *cp = buf;
@@ -683,6 +685,7 @@ static void populate_value(struct refinfo *ref)
} else if (!strcmp(name, "HEAD")) {
const char *head;
unsigned char sha1[20];
+
head = resolve_ref_unsafe("HEAD", sha1, 1, NULL);
if (!strcmp(ref->refname, head))
v->s = "*";
@@ -695,11 +698,46 @@ static void populate_value(struct refinfo *ref)
formatp = strchr(name, ':');
/* look for "short" refname format */
if (formatp) {
+   int num_ours, num_theirs;
+
formatp++;
if (!strcmp(formatp, "short"))
refname = shorten_unambiguous_ref(refname,
  warn_ambiguous_refs);
-   else
+   else if (!strcmp(formatp, "track") &&
+   !prefixcmp(name, "upstream")) {
+   char buf[40];
+
+   if (!upstream_present)
+   continue;
+   if (!stat_tracking_info(branch, &num_ours, 
&num_theirs))
+   v->s = "";
+   else if (!num_ours) {
+   sprintf(buf, "[behind %d]", num_theirs);
+   v->s = xstrdup(buf);
+   } else if (!num_theirs) {
+   sprintf(buf, "[ahead %d]", num_ours);
+   v->s = xstrdup(buf);
+   } else {
+   sprintf(buf, "[ahead %d, behind %d]",
+   num_ours, num_theirs);
+   v->s = xstrdup(buf);
+

[PATCH 08/15] for-each-ref: get --pretty using format_commit_message

From: Nguyễn Thái Ngọc Duy 

--format is very limited in its capabilities.  Introduce --pretty, which
extends the existing --format with pretty-formats.  In --pretty:

 - Existing --format %(atom) is available. They also accept some pretty
   magic.  For example, you can use "% (atom)" to only display a leading
   space if the atom produces something.

 - %ab to display a hex character 0xab is not available as it may
   conflict with other pretty's placeholders.  Use %xab instead.

 - Many pretty placeholders are designed to work on commits.  While some
   of them should work on tags too, they don't (yet).

 - Unsupported atoms cause for-each-ref to exit early and report.
   Unsupported pretty placeholders are displayed as-is.

 - Pretty placeholders can not be used as a sorting criteria.

--format is considered deprecated. If the user hits a bug specific in
--format code, they are advised to migrate to --pretty.

[rr: documentation]

Signed-off-by: Ramkumar Ramachandra 
---
 Documentation/git-for-each-ref.txt | 22 -
 builtin/for-each-ref.c | 67 --
 2 files changed, 85 insertions(+), 4 deletions(-)

diff --git a/Documentation/git-for-each-ref.txt 
b/Documentation/git-for-each-ref.txt
index f2e08d1..6135812 100644
--- a/Documentation/git-for-each-ref.txt
+++ b/Documentation/git-for-each-ref.txt
@@ -9,7 +9,8 @@ SYNOPSIS
 
 [verse]
 'git for-each-ref' [--count=] [--shell|--perl|--python|--tcl]
-  [(--sort=)...] [--format=] [...]
+  [(--sort=)...] [--format=|--pretty=]
+  [...]
 
 DESCRIPTION
 ---
@@ -47,6 +48,25 @@ OPTIONS
`xx`; for example `%00` interpolates to `\0` (NUL),
`%09` to `\t` (TAB) and `%0a` to `\n` (LF).
 
+::
+   A format string with supporting placeholders described in the
+   "PRETTY FORMATS" section in linkgit:git-log[1].  Additionally
+   supports placeholders from ``
+   (i.e. `%[*](fieldname)`).
++
+Caveats:
+
+1. Many of the placeholders in "PRETTY FORMATS" are designed to work
+   specifically on commit objects: when non-commit objects are
+   supplied, those placeholders won't work.
+
+2. Does not interpolate `%ab` (where `ab` are hex digits) with the
+   corresponding hex code.  To print a byte from a hex code, use
+   `%xab` (from pretty-formats) instead.
+
+3. Only the placeholders inherited from `` will respect
+   quoting settings.
+
 ...::
If one or more patterns are given, only refs are shown that
match against at least one pattern, either using fnmatch(3) or
diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index e2d6c5a..576a882 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -962,6 +962,58 @@ static void show_refs(struct refinfo **refs, int maxcount,
strbuf_release(&sb);
 }
 
+struct format_one_atom_context {
+   struct refinfo *info;
+   int quote_style;
+};
+
+static size_t format_one_atom(struct strbuf *sb, const char *placeholder,
+ void *format_context, void *user_data,
+ struct strbuf *subst)
+{
+   struct format_one_atom_context *ctx = user_data;
+   const char *ep;
+
+   if (*placeholder == '%') {
+   strbuf_addch(sb, '%');
+   return 1;
+   }
+
+   if (*placeholder != '(')
+   return 0;
+
+   ep = strchr(placeholder + 1, ')');
+   if (!ep)
+   return 0;
+   print_value(sb, ctx->info, parse_atom(placeholder + 1, ep),
+   ctx->quote_style);
+   return ep + 1 - placeholder;
+}
+
+static void show_pretty_refs(struct refinfo **refs, int maxcount,
+const char *format, int quote_style)
+{
+   struct pretty_print_context ctx = {0};
+   struct format_one_atom_context fctx;
+   struct strbuf sb = STRBUF_INIT;
+   int i;
+
+   ctx.format = format_one_atom;
+   ctx.user_data = &fctx;
+   fctx.quote_style = quote_style;
+   for (i = 0; i < maxcount; i++) {
+   struct commit *commit = NULL;
+   fctx.info = refs[i];
+   if (sha1_object_info(refs[i]->objectname, NULL) == OBJ_COMMIT)
+   commit = lookup_commit(refs[i]->objectname);
+   strbuf_reset(&sb);
+   format_commit_message(commit, format, &sb, &ctx);
+   strbuf_addch(&sb, '\n');
+   fputs(sb.buf, stdout);
+   }
+   strbuf_release(&sb);
+}
+
 static struct ref_sort *default_sort(void)
 {
static const char cstr_name[] = "refname";
@@ -1003,7 +1055,9 @@ static char const * const for_each_ref_usage[] = {
 int cmd_for_each_ref(int argc, const char **argv, const char *prefix)
 {
int num_refs;
-   const char *format = "%(objectname) %(objecttype)\t%(refname)";
+   const char *default_format = "%(objectname) %(objecttype)\t%(refname)";
+   const char *format = default_for

[PATCH 13/15] for-each-ref: improve responsiveness of %(upstream:track)

From: Nguyễn Thái Ngọc Duy 

Before anything is printed, for-each-ref sorts all refs first.  As
part of the sorting, populate_value() is called to fill the values in
for all atoms/placeholders per entry. By the time sort_refs() is done,
pretty much all data is already retrieved.

This works fine when data can be cheaply retrieved before
%(upstream:track) comes into the picture. It may take a noticeable
amount of time to process %(upstream:track) for each entry. All
entries add up and make --format='%(refname)%(upstream:track)' seem
hung for a few seconds, then display everything at once.

Improve the responsiveness by only processing the one atom (*) at a
time so that processing one atom for all entries (e.g. sorting) won't
cause much delay (unless you choose a "heavy" atom to process).

(*) This is not entirely correct. If you sort by an atom that needs
object database access, then it will fill all atoms that need odb.
Which is not a bad thing. We don't want to access odb once at sorting
phase and again at display phase.

Signed-off-by: Ramkumar Ramachandra 
---
 builtin/for-each-ref.c | 49 +
 1 file changed, 25 insertions(+), 24 deletions(-)

diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index adc965e..edaa6b2 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -565,20 +565,6 @@ static void grab_sub_body_contents(struct atom_value *val, 
int deref, struct obj
 }
 
 /*
- * We want to have empty print-string for field requests
- * that do not apply (e.g. "authordate" for a tag object)
- */
-static void fill_missing_values(struct atom_value *val)
-{
-   int i;
-   for (i = 0; i < used_atom_cnt; i++) {
-   struct atom_value *v = &val[i];
-   if (v->s == NULL)
-   v->s = "";
-   }
-}
-
-/*
  * val is a list of atom_value to hold returned values.  Extract
  * the values for atoms in used_atom array out of (obj, buf, sz).
  * when deref is false, (obj, buf, sz) is the object that is
@@ -621,7 +607,7 @@ static inline char *copy_advance(char *dst, const char *src)
 /*
  * Parse the object referred by ref, and grab needed value.
  */
-static void populate_value(struct refinfo *ref)
+static void populate_value(struct refinfo *ref, int only_this_atom)
 {
void *buf;
struct object *obj;
@@ -630,13 +616,15 @@ static void populate_value(struct refinfo *ref)
const unsigned char *tagged;
int upstream_present = 0;
 
-   ref->value = xcalloc(sizeof(struct atom_value), used_atom_cnt);
+   if (!ref->value) {
+   ref->value = xcalloc(sizeof(struct atom_value), used_atom_cnt);
 
-   if (need_symref && (ref->flag & REF_ISSYMREF) && !ref->symref) {
-   unsigned char unused1[20];
-   ref->symref = resolve_refdup(ref->refname, unused1, 1, NULL);
-   if (!ref->symref)
-   ref->symref = "";
+   if (need_symref && (ref->flag & REF_ISSYMREF) && !ref->symref) {
+   unsigned char unused1[20];
+   ref->symref = resolve_refdup(ref->refname, unused1, 1, 
NULL);
+   if (!ref->symref)
+   ref->symref = "";
+   }
}
 
/* Fill in specials first */
@@ -648,6 +636,9 @@ static void populate_value(struct refinfo *ref)
const char *formatp;
struct branch *branch;
 
+   if (only_this_atom != -1 && only_this_atom != i)
+   continue;
+
if (*name == '*') {
deref = 1;
name++;
@@ -754,6 +745,10 @@ static void populate_value(struct refinfo *ref)
 
for (i = 0; i < used_atom_cnt; i++) {
struct atom_value *v = &ref->value[i];
+
+   if (only_this_atom != -1 && only_this_atom != i)
+   continue;
+
if (v->s == NULL)
goto need_obj;
}
@@ -809,9 +804,15 @@ static void populate_value(struct refinfo *ref)
  */
 static void get_value(struct refinfo *ref, int atom, struct atom_value **v)
 {
-   if (!ref->value) {
-   populate_value(ref);
-   fill_missing_values(ref->value);
+   if (!ref->value || !ref->value[atom].s) {
+   populate_value(ref, atom);
+   /*
+* We want to have empty print-string for field
+* requests that do not apply (e.g. "authordate" for a
+* tag object)
+*/
+   if (!ref->value[atom].s)
+   ref->value[atom].s = "";
}
*v = &ref->value[atom];
 }
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 10/15] for-each-ref: introduce format specifier %>() and %<()

From: Nguyễn Thái Ngọc Duy 

Pretty placeholders %>(N) and %<(N) require a user provided width N,
which makes sense because the commit chain could be really long and the
user only needs to look at a few at the top, going to the end just to
calculate the best width wastes CPU cycles.

for-each-ref is different; the display set is small, and we display them
all at once. We even support sorting, which goes through all display
items anyway.  This patch introduces new %>(*) and %<(*), which are
supposed to be followed immediately by %(fieldname) (i.e. original
for-each-ref specifiers, not ones coming from pretty.c). They calculate
the best width for the %(fieldname), ignoring ansi escape sequences if
any.

[rr: documentation]

Signed-off-by: Ramkumar Ramachandra 
---
 Documentation/git-for-each-ref.txt |  7 +++
 builtin/for-each-ref.c | 38 ++
 2 files changed, 45 insertions(+)

diff --git a/Documentation/git-for-each-ref.txt 
b/Documentation/git-for-each-ref.txt
index 6135812..7d6db7f 100644
--- a/Documentation/git-for-each-ref.txt
+++ b/Documentation/git-for-each-ref.txt
@@ -47,6 +47,10 @@ OPTIONS
are hex digits interpolates to character with hex code
`xx`; for example `%00` interpolates to `\0` (NUL),
`%09` to `\t` (TAB) and `%0a` to `\n` (LF).
++
+Placeholders `%<(*)` and `%>(*)` work like `%<()` and `%>()`
+respectively, except that the width of the next placeholder is
+calculated.
 
 ::
A format string with supporting placeholders described in the
@@ -67,6 +71,9 @@ Caveats:
 3. Only the placeholders inherited from `` will respect
quoting settings.
 
+3. Only the placeholders inherited from `` will work with the
+   alignment placeholders `%<(*)` and '%>(*)`.
+
 ...::
If one or more patterns are given, only refs are shown that
match against at least one pattern, either using fnmatch(3) or
diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index f8a3880..053a622 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -9,6 +9,7 @@
 #include "quote.h"
 #include "parse-options.h"
 #include "remote.h"
+#include "utf8.h"
 
 /* Quoting styles */
 #define QUOTE_NONE 0
@@ -966,10 +967,30 @@ static void show_refs(struct refinfo **refs, int maxcount,
 }
 
 struct format_one_atom_context {
+   struct refinfo **refs;
+   int maxcount;
+
struct refinfo *info;
int quote_style;
 };
 
+static unsigned int get_atom_width(struct format_one_atom_context *ctx,
+  const char *start, const char *end)
+{
+   struct strbuf sb = STRBUF_INIT;
+   int i, atom = parse_atom(start, end);
+   unsigned int len = 0, sb_len;
+   for (i = 0; i < ctx->maxcount; i++) {
+   print_value(&sb, ctx->refs[i], atom, ctx->quote_style);
+   sb_len = utf8_strnwidth(sb.buf, sb.len, 1);
+   if (sb_len > len)
+   len = sb_len;
+   strbuf_reset(&sb);
+   }
+   strbuf_release(&sb);
+   return len;
+}
+
 static size_t format_one_atom(struct strbuf *sb, const char *placeholder,
  void *format_context, void *user_data,
  struct strbuf *subst)
@@ -982,6 +1003,21 @@ static size_t format_one_atom(struct strbuf *sb, const 
char *placeholder,
return 1;
}
 
+   /*
+* Substitute %>(*)%(atom) and friends with real width.
+*/
+   if (*placeholder == '>' || *placeholder == '<') {
+   const char *star = placeholder + 1;
+   if (!prefixcmp(star, "(*)%(") &&
+   ((ep = strchr(star + strlen("(*)%("), ')')) != NULL)) {
+   star++;
+   strbuf_addf(subst, "%c(%u)",
+   *placeholder,
+   get_atom_width(ctx, star + strlen("*)%("), 
ep));
+   return 1 + strlen("(*)");
+   }
+   }
+
if (*placeholder != '(')
return 0;
 
@@ -1003,6 +1039,8 @@ static void show_pretty_refs(struct refinfo **refs, int 
maxcount,
 
ctx.format = format_one_atom;
ctx.user_data = &fctx;
+   fctx.refs = refs;
+   fctx.maxcount = maxcount;
fctx.quote_style = quote_style;
for (i = 0; i < maxcount; i++) {
struct commit *commit = NULL;
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 02/15] for-each-ref: don't print out elements directly

From: Nguyễn Thái Ngọc Duy 

Currently, the entire callchain starting from show_ref() parses and
prints immediately.  This inflexibility limits our ability to extend the
parser.  So, convert the entire callchain to accept a strbuf argument to
write to.  Also introduce a show_refs() helper that calls show_ref() in
a loop to avoid cluttering up cmd_for_each_ref() with the task of
initializing/freeing the strbuf.

[rr: commit message]

Signed-off-by: Ramkumar Ramachandra 
---
 builtin/for-each-ref.c | 55 --
 1 file changed, 35 insertions(+), 20 deletions(-)

diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index 1d4083c..e2d6c5a 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -864,31 +864,31 @@ static void sort_refs(struct ref_sort *sort, struct 
refinfo **refs, int num_refs
qsort(refs, num_refs, sizeof(struct refinfo *), compare_refs);
 }
 
-static void print_value(struct refinfo *ref, int atom, int quote_style)
+static void print_value(struct strbuf *sb, struct refinfo *ref,
+   int atom, int quote_style)
 {
struct atom_value *v;
-   struct strbuf sb = STRBUF_INIT;
get_value(ref, atom, &v);
switch (quote_style) {
case QUOTE_NONE:
-   fputs(v->s, stdout);
+   strbuf_addstr(sb, v->s);
break;
case QUOTE_SHELL:
-   sq_quote_buf(&sb, v->s);
+   sq_quote_buf(sb, v->s);
break;
case QUOTE_PERL:
-   perl_quote_buf(&sb, v->s);
+   perl_quote_buf(sb, v->s);
break;
case QUOTE_PYTHON:
-   python_quote_buf(&sb, v->s);
+   python_quote_buf(sb, v->s);
break;
case QUOTE_TCL:
-   tcl_quote_buf(&sb, v->s);
+   tcl_quote_buf(sb, v->s);
break;
}
if (quote_style != QUOTE_NONE) {
-   fputs(sb.buf, stdout);
-   strbuf_release(&sb);
+   fputs(sb->buf, stdout);
+   strbuf_release(sb);
}
 }
 
@@ -910,7 +910,7 @@ static int hex2(const char *cp)
return -1;
 }
 
-static void emit(const char *cp, const char *ep)
+static void emit(struct strbuf *sb, const char *cp, const char *ep)
 {
while (*cp && (!ep || cp < ep)) {
if (*cp == '%') {
@@ -919,32 +919,47 @@ static void emit(const char *cp, const char *ep)
else {
int ch = hex2(cp + 1);
if (0 <= ch) {
-   putchar(ch);
+   strbuf_addch(sb, ch);
cp += 3;
continue;
}
}
}
-   putchar(*cp);
+   strbuf_addch(sb, *cp);
cp++;
}
 }
 
-static void show_ref(struct refinfo *info, const char *format, int quote_style)
+static void show_ref(struct strbuf *sb, struct refinfo *info,
+const char *format, int quote_style)
 {
const char *cp, *sp, *ep;
 
for (cp = format; *cp && (sp = find_next(cp)); cp = ep + 1) {
ep = strchr(sp, ')');
if (cp < sp)
-   emit(cp, sp);
-   print_value(info, parse_atom(sp + 2, ep), quote_style);
+   emit(sb, cp, sp);
+   print_value(sb, info, parse_atom(sp + 2, ep), quote_style);
}
if (*cp) {
sp = cp + strlen(cp);
-   emit(cp, sp);
+   emit(sb, cp, sp);
}
-   putchar('\n');
+   strbuf_addch(sb, '\n');
+}
+
+static void show_refs(struct refinfo **refs, int maxcount,
+ const char *format, int quote_style)
+{
+   struct strbuf sb = STRBUF_INIT;
+   int i;
+
+   for (i = 0; i < maxcount; i++) {
+   strbuf_reset(&sb);
+   show_ref(&sb, refs[i], format, quote_style);
+   fputs(sb.buf, stdout);
+   }
+   strbuf_release(&sb);
 }
 
 static struct ref_sort *default_sort(void)
@@ -987,7 +1002,7 @@ static char const * const for_each_ref_usage[] = {
 
 int cmd_for_each_ref(int argc, const char **argv, const char *prefix)
 {
-   int i, num_refs;
+   int num_refs;
const char *format = "%(objectname) %(objecttype)\t%(refname)";
struct ref_sort *sort = NULL, **sort_tail = &sort;
int maxcount = 0, quote_style = 0;
@@ -1041,7 +1056,7 @@ int cmd_for_each_ref(int argc, const char **argv, const 
char *prefix)
 
if (!maxcount || num_refs < maxcount)
maxcount = num_refs;
-   for (i = 0; i < maxcount; i++)
-   show_ref(refs[i], format, quote_style);
+
+   show_refs(refs, maxcount, format, quote_style);
return 0;
 }
--

[PATCH 05/15] pretty: extend pretty_print_context with callback

From: Nguyễn Thái Ngọc Duy 

The struct pretty_print_context contains the context in which the
placeholders in format_commit_one() should be parsed.  Although
format_commit_one() primarily acts as a parser, there is no way for a
caller to plug in custom callbacks.  Now, callers can:

1. Parse a custom placeholder that is not supported by
   format_commit_one(), and act on it independently of the pretty
   machinery.

2. Parse a custom placeholder to substitute the custom placeholder with
   a placeholder that format_commit_one() understands.  This is
   especially useful for supporting %>(*), where * is substituted with a
   length computed by the caller.

To support these two usecases, the interface for the function looks
like:

typedef size_t (*format_message_fn)(struct strbuf *sb,
const char *placeholder,
void *format_context,
void *user_data,
struct strbuf *placeholder_subst)

It is exactly like format_commit_one(), except that there are two
additional fields: user_data (to pass custom data to the callback), and
placeholder_subst (to set the substitution).  The callback should return
the length of the original string parsed, and optionally set
placeholder_subst.

[rr: commit message, minor modifications]

Signed-off-by: Ramkumar Ramachandra 
---
 commit.h |  8 
 pretty.c | 25 +
 2 files changed, 33 insertions(+)

diff --git a/commit.h b/commit.h
index 67bd509..04bd935 100644
--- a/commit.h
+++ b/commit.h
@@ -78,6 +78,12 @@ enum cmit_fmt {
CMIT_FMT_UNSPECIFIED
 };
 
+typedef size_t (*format_message_fn)(struct strbuf *sb,
+   const char *placeholder,
+   void *format_context,
+   void *user_data,
+   struct strbuf *placeholder_subst);
+
 struct pretty_print_context {
enum cmit_fmt fmt;
int abbrev;
@@ -92,6 +98,8 @@ struct pretty_print_context {
const char *output_encoding;
struct string_list *mailmap;
int color;
+   format_message_fn format;
+   void *user_data;
 };
 
 struct userformat_want {
diff --git a/pretty.c b/pretty.c
index 9e43154..095e5ba 100644
--- a/pretty.c
+++ b/pretty.c
@@ -1069,6 +1069,31 @@ static size_t format_commit_one(struct strbuf *sb, /* in 
UTF-8 */
struct commit_list *p;
int h1, h2;
 
+   if (c->pretty_ctx->format) {
+   struct strbuf subst = STRBUF_INIT;
+   int ret = c->pretty_ctx->format(sb, placeholder, context,
+   c->pretty_ctx->user_data,
+   &subst);
+   if (ret && subst.len) {
+   /*
+* Something was parsed by format(), and a
+* placeholder-substitution was set.
+* Recursion is required to override the
+* return value of format_commit_one() with
+* ret: the length of the original string
+* before substitution.
+*/
+   ret = format_commit_one(sb, subst.buf, context) ? ret : 
0;
+   strbuf_release(&subst);
+   return ret;
+   } else if (ret)
+   /*
+* Something was parsed by format(), but
+* no placeholder-substitution was set.
+*/
+   return ret;
+   }
+
/* these are independent of the commit */
switch (placeholder[0]) {
case 'C':
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 11/15] for-each-ref: introduce %(HEAD) marker

'git branch' shows which branch you are currently on with an '*', but
'git for-each-ref' misses this feature.  So, extend the format with
%(HEAD) to do exactly the same thing.

Now you can use the following format in for-each-ref:

  %C(red)%(HEAD)%C(reset) %C(green)%(refname:short)%C(reset)

to display a red asterisk next to the current ref.

Signed-off-by: Ramkumar Ramachandra 
---
 Documentation/git-for-each-ref.txt |  4 
 builtin/for-each-ref.c | 13 +++--
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/Documentation/git-for-each-ref.txt 
b/Documentation/git-for-each-ref.txt
index 7d6db7f..d529296 100644
--- a/Documentation/git-for-each-ref.txt
+++ b/Documentation/git-for-each-ref.txt
@@ -120,6 +120,10 @@ upstream::
from the displayed ref. Respects `:short` in the same way as
`refname` above.
 
+HEAD::
+   Useful to indicate the currently checked out branch.  Is '*'
+   if HEAD points to the current ref, and ' ' otherwise.
+
 In addition to the above, for commit and tag objects, the header
 field names (`tree`, `parent`, `object`, `type`, and `tag`) can
 be used to specify the value in the header field.
diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
index 053a622..b0b8236 100644
--- a/builtin/for-each-ref.c
+++ b/builtin/for-each-ref.c
@@ -76,6 +76,7 @@ static struct {
{ "upstream" },
{ "symref" },
{ "flag" },
+   { "HEAD" },
 };
 
 /*
@@ -679,8 +680,16 @@ static void populate_value(struct refinfo *ref)
v->s = xstrdup(buf + 1);
}
continue;
-   }
-   else
+   } else if (!strcmp(name, "HEAD")) {
+   const char *head;
+   unsigned char sha1[20];
+   head = resolve_ref_unsafe("HEAD", sha1, 1, NULL);
+   if (!strcmp(ref->refname, head))
+   v->s = "*";
+   else
+   v->s = " ";
+   continue;
+   } else
continue;
 
formatp = strchr(name, ':');
-- 
1.8.3.GIT

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Re: Re: What's cooking in git.git (May 2013, #09; Wed, 29)

2013-06-04 Thread John Keeping

On Tue, Jun 04, 2013 at 09:17:17PM +1000, Heiko Voigt wrote:
> On Tue, Jun 04, 2013 at 09:10:45AM +0100, John Keeping wrote:
> > On Tue, Jun 04, 2013 at 03:29:51PM +1000, Heiko Voigt wrote:
> > > On Mon, Jun 03, 2013 at 11:23:41PM +0100, John Keeping wrote:
> > > > > Sorry, I should have been more specific here. I saw that you did some
> > > > > changes to make "submodule add" do the right thing with relative 
> > > > > paths,
> > > > > but the following change to t7406 does not work like I believe it
> > > > > should but instead makes the test fail:
> > > > > ---8<-
> > > > > diff --git a/t/t7406-submodule-update.sh b/t/t7406-submodule-update.sh
> > > > > index a4ffea0..9766b9e 100755
> > > > > --- a/t/t7406-submodule-update.sh
> > > > > +++ b/t/t7406-submodule-update.sh
> > > > > @@ -559,7 +559,9 @@ test_expect_success 'add different submodules to 
> > > > > the same pa
> > > > >  test_expect_success 'submodule add places git-dir in superprojects 
> > > > > git-dir' '
> > > > > (cd super &&
> > > > >  mkdir deeper &&
> > > > > -git submodule add ../submodule deeper/submodule &&
> > > > > +(cd deeper &&
> > > > > + git submodule add ../../submodule submodule
> > > > > +) &&
> > > > >  (cd deeper/submodule &&
> > > > >   git log > ../../expected
> > > > >  ) &&
> > > > > ---8<-
> > > > 
> > > > Ah, ok.  I think this case is problematic because the repository
> > > > argument is either relative to "remote.origin.url" or to the top of the
> > > > working tree if there is no "origin" remote.  I wonder if we should just
> > > > die when a relative path is given for the repository and we're not at
> > > > the top of the working tree.
> > > 
> > > Why not behave as if we are at the top of the working tree for relative
> > > paths? If there is an origin remote thats fine. If there is no origin
> > > remote you could warn that the path used is taken relative from the root
> > > of the superproject during add. What do you think?
> > 
> > That's what the patch currently queued on "pu" does, which Jens wants to
> > change, isn't it?
> 
> True I did not realize this when reading it the first time. But I think
> we should still not die when in a subdirectory. After all this series is
> trying to archive that the submodule command works in subdirectories
> seamlessly right? So you probably want to translate a relative path
> without "origin" remote given from a subdirectory to the superproject
> level and use that. Then you do not have to die.

The problem is that sometimes you do want to adjust the path and
sometimes you don't.  Reading git-submodule(1), it says:

 This may be either an absolute URL, or (if it begins with ./ or
 ../), the location relative to the superproject’s origin
 repository.
 [snip]
 If the superproject doesn’t have an origin configured the
 superproject is its own authoritative upstream and the current
 working directory is used instead.

So I think it's quite reasonable to have a server layout that looks like
this:

project
|- libs
|  |- libA
|  `- libB
|- core.git

and with only core.git on your local system do:

cd core/libs
git submodule add ../libs/libB

expecting that to point to libB.  But if we adjust the path then the
user has to do:

git submodule add ../../libs/libB

However, it is also perfectly reasonable to have no remote configured
and the library next to the repository itself.  In which case we do want
to specify the additional "../" so that shell completion works in the
natural way.

The only way I can see to resolve the ambiguity is to die when we hit
this particular case.  This should be acceptable because people
shouldn't be adding new submodules anywhere near as often as they
perform other submodule operations and it doesn't affect absolute URLs.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/15] Towards a more awesome git-branch

On Tue, Jun 4, 2013 at 7:35 PM, Ramkumar Ramachandra  wrote:
> Hi,
>
> Duy and I have been working on this topic for some time now.  Here's a
> review candidate.

I'm still hung up one the detached HEAD thing. It's a bit quirky to
put in for-each-ref, but for-each-ref can't truly replace branch
--list until it can display detached HEAD. But I think we can finish
this part and get it in first. Should be useful for some people
already.

> Duy is currently writing code to factor out -v[v] and adding --sort,
> --count to git-branch, but I'm having a lot of difficulty working with
> a series of this size.  I'm leaning towards getting this merged before
> tackling branch (who wants to review a 40-part series?).  Especially
> after [14/15] and [15/15], git for-each-ref should be usable in its
> own right.

Agreed (a bit of concern about 14 and 15/15, though). Signed-off-by:
me for the whole series.
--
Duy
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v2 0/3] Towards a useable git-branch

Duy Nguyen wrote:
> Nobody should ever parse these output
> with scripts. The color can be generated from color.branch.*.

How do we implement color.branch.(current|local|remote|plain)?  In the
current code, we take a crude approach by hand-constructing argc, argv
strings and passing it to cmd_for_each_ref().  There are no
conditionals in the format syntax (and introducing one is probably not
a good idea either): so, I'm guessing these configuration variables
have to be read by for-each-ref?

> A bigger
> problem is show_detached(), --[no-]merged, --with and --contains. We
> need to move some of those code over to for-each-ref.

I saw that you fixed this.

> Another problem
> is the new "branch -v" seems to less responsive than old "branch -v"
> because (I think) of sorting, even if we don't need it.

Does your track-responsiveness patch fix this?

> I checked out
> your branch, made some more updates and pushed out to my
> for-each-ref-pretty again. Changes are:

*pants* Sorry, I'm finding it hard to keep up.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[no subject]

2013-06-04 Thread Мурад




Отправлено с iPhone

Re: [PATCH 10/15] for-each-ref: introduce format specifier %>() and %<()

On Tue, Jun 4, 2013 at 7:35 PM, Ramkumar Ramachandra  wrote:
> +static unsigned int get_atom_width(struct format_one_atom_context *ctx,
> +  const char *start, const char *end)
> +{
> +   struct strbuf sb = STRBUF_INIT;
> +   int i, atom = parse_atom(start, end);
> +   unsigned int len = 0, sb_len;
> +   for (i = 0; i < ctx->maxcount; i++) {
> +   print_value(&sb, ctx->refs[i], atom, ctx->quote_style);
> +   sb_len = utf8_strnwidth(sb.buf, sb.len, 1);
> +   if (sb_len > len)
> +   len = sb_len;
> +   strbuf_reset(&sb);
> +   }
> +   strbuf_release(&sb);
> +   return len;
> +}
> +

I mentioned it before and I do it again. This is not optimal. We
should cache the result of get_atom_width() after the first
calculation. I haven't done it yet because I'm still not sure if it's
worth supporting %<(*)%non-atom at this stage. Caching for atoms only
is much easier because atom is indexed. But I guess it's ok in this
shape unless you run this over hundreds of refs.
--
Duy
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 15/15] for-each-ref: use get_pretty_userformat in --pretty

On Tue, Jun 4, 2013 at 7:35 PM, Ramkumar Ramachandra  wrote:
> Use get_pretty_userformat() to interpret the --pretty string.  This
> means that leading you can now reference a format specified in a pretty.*
> configuration variable as an argument to 'git for-each-ref --pretty='.
> There are two caveats:
>
> 1. A leading "format:" or "tformat:" is automatically stripped and
>ignored.  Separator semantics are not configurable (yet).
>
> 2. No built-in formats are available.  The ones specified in
>pretty-formats (oneline, short etc) don't make sense when displaying
>refs anyway.
>

I spoke too early about the concern with this patch. At first it
looked like you replace default formats with config keys. But nope,
this is brilliant.

> Signed-off-by: Ramkumar Ramachandra 
> ---
>  Documentation/git-for-each-ref.txt |  3 +++
>  builtin/for-each-ref.c | 16 +---
>  2 files changed, 12 insertions(+), 7 deletions(-)
>
> diff --git a/Documentation/git-for-each-ref.txt 
> b/Documentation/git-for-each-ref.txt
> index 05ff7ba..7f3cba5 100644
> --- a/Documentation/git-for-each-ref.txt
> +++ b/Documentation/git-for-each-ref.txt
> @@ -60,6 +60,9 @@ calculated.
>  +
>  Caveats:
>
> +0. No built-in formats from PRETTY FORMATS (like oneline, short) are
> +   available.
> +
>  1. Many of the placeholders in "PRETTY FORMATS" are designed to work
> specifically on commit objects: when non-commit objects are
> supplied, those placeholders won't work.
> diff --git a/builtin/for-each-ref.c b/builtin/for-each-ref.c
> index edaa6b2..c00ab05 100644
> --- a/builtin/for-each-ref.c
> +++ b/builtin/for-each-ref.c
> @@ -1146,7 +1146,7 @@ int cmd_for_each_ref(int argc, const char **argv, const 
> char *prefix)
> int num_refs;
> const char *default_format = "%(objectname) 
> %(objecttype)\t%(refname)";
> const char *format = default_format;
> -   const char *pretty = NULL;
> +   const char *pretty_raw = NULL, *pretty_userformat = NULL;
> struct ref_sort *sort = NULL, **sort_tail = &sort;
> int maxcount = 0, quote_style = 0;
> struct refinfo **refs;
> @@ -1165,13 +1165,15 @@ int cmd_for_each_ref(int argc, const char **argv, 
> const char *prefix)
> OPT_GROUP(""),
> OPT_INTEGER( 0 , "count", &maxcount, N_("show only  
> matched refs")),
> OPT_STRING(  0 , "format", &format, N_("format"), N_("format 
> to use for the output")),
> -   OPT_STRING(  0 , "pretty", &pretty, N_("format"), 
> N_("alternative format to use for the output")),
> +   OPT_STRING(  0 , "pretty", &pretty_raw, N_("format"), 
> N_("alternative format to use for the output")),
> OPT_CALLBACK(0 , "sort", sort_tail, N_("key"),
> N_("field name to sort on"), &opt_parse_sort),
> OPT_END(),
> };
>
> parse_options(argc, argv, prefix, opts, for_each_ref_usage, 0);
> +   if (pretty_raw)
> +   pretty_userformat = get_pretty_userformat(pretty_raw);
> if (maxcount < 0) {
> error("invalid --count argument: `%d'", maxcount);
> usage_with_options(for_each_ref_usage, opts);
> @@ -1180,10 +1182,10 @@ int cmd_for_each_ref(int argc, const char **argv, 
> const char *prefix)
> error("more than one quoting style?");
> usage_with_options(for_each_ref_usage, opts);
> }
> -   if (format != default_format && pretty)
> +   if (format != default_format && pretty_userformat)
> die("--format and --pretty cannot be used together");
> -   if ((pretty && verify_format(pretty, 1)) ||
> -   (!pretty && verify_format(format, 0)))
> +   if ((pretty_userformat && verify_format(pretty_userformat, 1)) ||
> +   (!pretty_userformat && verify_format(format, 0)))
> usage_with_options(for_each_ref_usage, opts);
>
> if (!sort)
> @@ -1204,8 +1206,8 @@ int cmd_for_each_ref(int argc, const char **argv, const 
> char *prefix)
> if (!maxcount || num_refs < maxcount)
> maxcount = num_refs;
>
> -   if (pretty)
> -   show_pretty_refs(refs, maxcount, pretty, quote_style);
> +   if (pretty_userformat)
> +   show_pretty_refs(refs, maxcount, pretty_userformat, 
> quote_style);
> else
> show_refs(refs, maxcount, format, quote_style);
> return 0;
> --
> 1.8.3.GIT
>



--
Duy
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v2 0/3] Towards a useable git-branch

On Tue, Jun 4, 2013 at 7:52 PM, Ramkumar Ramachandra  wrote:
> Duy Nguyen wrote:
>> Nobody should ever parse these output
>> with scripts. The color can be generated from color.branch.*.
>
> How do we implement color.branch.(current|local|remote|plain)?  In the
> current code, we take a crude approach by hand-constructing argc, argv
> strings and passing it to cmd_for_each_ref().  There are no
> conditionals in the format syntax (and introducing one is probably not
> a good idea either): so, I'm guessing these configuration variables
> have to be read by for-each-ref?

Maybe. I don't really like the idea that for-each-ref reads _branch_
config. We could introduce the same set of keys but in
color.for-each-ref namespace. %C(auto) will take care of the logic and
choose the correct color key. When we replace branch's listing code
with for-each-ref, I think we could somehow override for-each-ref keys
with branch ones in-core. Too hacky?

>> A bigger
>> problem is show_detached(), --[no-]merged, --with and --contains. We
>> need to move some of those code over to for-each-ref.
>
> I saw that you fixed this.

Nope. --[no-]merged and --contains seem easy. show_detached is still
WIP, mostly because detached HEAD may or may not show when you provide
a pattern to git-branch (e.g. git branch --list 'foo*') and because
HEAD is always the first item, regardless of sorting order.
get_head_description also seems more porcelain-ish than a plumbing
feature..

>> Another problem
>> is the new "branch -v" seems to less responsive than old "branch -v"
>> because (I think) of sorting, even if we don't need it.
>
> Does your track-responsiveness patch fix this?

Yes.

>> I checked out
>> your branch, made some more updates and pushed out to my
>> for-each-ref-pretty again. Changes are:
>
> *pants* Sorry, I'm finding it hard to keep up.

Sorry that branch was in a better shape the day I sent my previous
email. Then I kind of used it as a playground with --[no-]merged,
--contains and stuff :-P
--
Duy
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Can `git blame` show the date that each line was merged?

2013-06-04 Thread Matt McClure

Can `git blame` show the date that each line was merged to the current
branch rather than the date it was committed?

Aside: in some trial and error I notice this oddity:

$ git blame --merges
usage: git blame [options] [rev-opts] [rev] [--] file

[rev-opts] are documented in git-rev-list(1)
...

$ git help rev-list | grep -F -e --merges
[ --merges ]
   --merges
   --min-parents=2 is the same as --merges.
--max-parents=0 gives all


-- 
Matt McClure
http://matthewlmcclure.com
http://www.mapmyfitness.com/profile/matthewlmcclure
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

git-daemon: needs /root/.config/git/config?

2013-06-04 Thread Ian Kumlien

Hi again, 

Due to the earlier problem I upgraded git on all machines 
and eneded up with a ubunut machine running in to problems.

I started getting errors like:
"fatal: protocol error: bad line length character: fata"

Which after some head scratching caused me to tell xinetd to directly
launch git-daemon, eventually it worked fine, but i did get this error
message:

Jun  4 16:12:05 xyz git-daemon[10246]: unable to access
'/root/.config/git/config': Permission denied

It's not the first time i've seen it but i've been able to ignore it
before. This is running as a local user (as in not root) and this user
shouldn't have access to /root. But i eventually had to do chown o+x
/root to workaround this error.

Now, this must be wrong somehow? Or does --user work in inetd mode now?

So... comments, ideas?
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 3/7] unpack-trees: factor out dup_entry

2013-06-04 Thread Peter Krefting


René Scharfe:


While we're add it,


 "add" → "at"

--
\\// Peter - http://www.softwolves.pp.se/
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

* glibc detected * git: double free or corruption (fasttop): 0x0000000001fab820 ***

2013-06-04 Thread Jason Gross

I get "*** glibc detected *** git: double free or corruption
(fasttop): 0x01fab820 ***" reliably on the following set of
commands.  I'm working on a remote machine where I don't have
administrative privileges, so I can't update from git 1.7.2.5 to a
newer version.  I figured I'd report it, even though I haven't checked
to see if it still happens in a newer version of git.

jgross@dr-wily:/tmp$ uname -a
Linux dr-wily 2.6.32-5-xen-amd64 #1 SMP Mon Feb 25 02:51:39 UTC 2013
x86_64 GNU/Linux
jgross@dr-wily:/tmp$ git --version
git version 1.7.2.5
jgross@dr-wily:/tmp$ git clone git://github.com/JasonGross/barnowl.git
Cloning into barnowl...
remote: Counting objects: 16064, done.
remote: Compressing objects: 100% (8172/8172), done.
remote: Total 16064 (delta 8517), reused 14679 (delta 7321)
Receiving objects: 100% (16064/16064), 8.69 MiB | 3.44 MiB/s, done.
Resolving deltas: 100% (8517/8517), done.
jgross@dr-wily:/tmp$ cd barnowl
jgross@dr-wily:/tmp/barnowl$ git checkout git-bug
Branch git-bug set up to track remote branch git-bug from origin.
Switched to a new branch 'git-bug'
jgross@dr-wily:/tmp/barnowl$ git shortlog | grep : | grep -v '' | less
*** glibc detected *** git: double free or corruption (fasttop):
0x023fa800 ***
=== Backtrace: =
/lib/libc.so.6(+0x71e16)[0x7fe7ed258e16]
/lib/libc.so.6(cfree+0x6c)[0x7fe7ed25db8c]
git[0x48b125]
git[0x4b7e08]
git[0x4577c2]
git[0x458151]
git[0x404671]
git[0x40484d]
/lib/libc.so.6(__libc_start_main+0xfd)[0x7fe7ed205c8d]
git[0x404009]
=== Memory map: 
0040-0050e000 r-xp  ca:01 541613
  /usr/bin/git
0070d000-00714000 rw-p 0010d000 ca:01 541613
  /usr/bin/git
00714000-0076 rw-p  00:00 0
023fa000-025b5000 rw-p  00:00 0  [heap]
7fe7e800-7fe7e8021000 rw-p  00:00 0
7fe7e8021000-7fe7ec00 ---p  00:00 0
7fe7ec721000-7fe7ec737000 r-xp  ca:01 655372
  /lib/libgcc_s.so.1
7fe7ec737000-7fe7ec936000 ---p 00016000 ca:01 655372
  /lib/libgcc_s.so.1
7fe7ec936000-7fe7ec937000 rw-p 00015000 ca:01 655372
  /lib/libgcc_s.so.1
7fe7ec937000-7fe7ed1e7000 r--p  ca:03 136990
  
/tmp/barnowl/.git/objects/pack/pack-67364344f941243318353ed78170fd5d35869d1f.pack
7fe7ed1e7000-7fe7ed34 r-xp  ca:01 659512
  /lib/libc-2.11.3.so
7fe7ed34-7fe7ed53f000 ---p 00159000 ca:01 659512
  /lib/libc-2.11.3.so
7fe7ed53f000-7fe7ed543000 r--p 00158000 ca:01 659512
  /lib/libc-2.11.3.so
7fe7ed543000-7fe7ed544000 rw-p 0015c000 ca:01 659512
  /lib/libc-2.11.3.so
7fe7ed544000-7fe7ed549000 rw-p  00:00 0
7fe7ed549000-7fe7ed56 r-xp  ca:01 659507
  /lib/libpthread-2.11.3.so
7fe7ed56-7fe7ed75f000 ---p 00017000 ca:01 659507
  /lib/libpthread-2.11.3.so
7fe7ed75f000-7fe7ed76 r--p 00016000 ca:01 659507
  /lib/libpthread-2.11.3.so
7fe7ed76-7fe7ed761000 rw-p 00017000 ca:01 659507
  /lib/libpthread-2.11.3.so
7fe7ed761000-7fe7ed765000 rw-p  00:00 0
7fe7ed765000-7fe7ed77c000 r-xp  ca:01 527676
  /usr/lib/libz.so.1.2.3.4
7fe7ed77c000-7fe7ed97b000 ---p 00017000 ca:01 527676
  /usr/lib/libz.so.1.2.3.4
7fe7ed97b000-7fe7ed97c000 rw-p 00016000 ca:01 527676
  /usr/lib/libz.so.1.2.3.4
7fe7ed97c000-7fe7ed99a000 r-xp  ca:01 659508
  /lib/ld-2.11.3.so
7fe7edaef000-7fe7edb5e000 r--p  ca:03 136991
  
/tmp/barnowl/.git/objects/pack/pack-67364344f941243318353ed78170fd5d35869d1f.idx
7fe7edb5e000-7fe7edb61000 rw-p  00:00 0
7fe7edb75000-7fe7edb99000 rw-p  00:00 0
7fe7edb99000-7fe7edb9a000 r--p 0001d000 ca:01 659508
  /lib/ld-2.11.3.so
7fe7edb9a000-7fe7edb9b000 rw-p 0001e000 ca:01 659508
  /lib/ld-2.11.3.so
7fe7edb9b000-7fe7edb9c000 rw-p  00:00 0
7fffd3766000-7fffd377b000 rw-p  00:00 0  [stack]
7fffd37ff000-7fffd380 r-xp  00:00 0  [vdso]
ff60-ff601000 r-xp  00:00 0
  [vsyscall]


-Jason
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Can `git blame` show the date that each line was merged?

On Tue, Jun 04, 2013 at 09:39:45AM -0400, Matt McClure wrote:

> Can `git blame` show the date that each line was merged to the current
> branch rather than the date it was committed?

Not exactly. Git does not record when a commit entered a particular
branch (or what the "ours" branch was called during a merge).  If you
follow a topic branch workflow in which an integrator merges each branch
into master, then following the first parent of each merge will show the
commits directly on master.

You can see this in action in git.git, which follows such a workflow, by
doing "git log --first-parent", which shows only the commits created
directly on master (mostly merges of topics, with a few trivial fixups
interspersed).

Similarly, you should be able to do "git blame --first-parent foo.c" to
pass blame only along the first-parent lines. Unfortunately, while
"blame" uses the regular revision code to parse its options, it does its
own traversal and does not respect each option. However, the patch to
teach it about --first-parent is pretty trivial:

diff --git a/builtin/blame.c b/builtin/blame.c
index 57a487e..0fb67af 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -1199,6 +1199,8 @@ static int num_scapegoats(struct rev_info *revs, struct 
commit *commit)
 {
int cnt;
struct commit_list *l = first_scapegoat(revs, commit);
+   if (revs->first_parent_only)
+   return l ? 1 : 0;
for (cnt = 0; l; l = l->next)
cnt++;
return cnt;

(though I suspect it would interact oddly with the "--reverse" option,
and we would want to either declare them mutually exclusive or figure
out some sane semantics).

> Aside: in some trial and error I notice this oddity:
> 
> $ git blame --merges
> usage: git blame [options] [rev-opts] [rev] [--] file
> 
> [rev-opts] are documented in git-rev-list(1)
> ...

Your problem is not the presence of "--merges" here, but that you forgot
the necessary "file" argument. Try "git blame --merges foo.c".

However, this suffers from the same problem as --first-parent, in that
it is accepted but not respected. Doing so would not be impossible, but
it is a little more than the two-liner above.

-Peff
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: git-daemon: needs /root/.config/git/config?

On Tue, Jun 04, 2013 at 04:13:14PM +0200, Ian Kumlien wrote:

> Due to the earlier problem I upgraded git on all machines 
> and eneded up with a ubunut machine running in to problems.
> 
> I started getting errors like:
> "fatal: protocol error: bad line length character: fata"
> 
> Which after some head scratching caused me to tell xinetd to directly
> launch git-daemon, eventually it worked fine, but i did get this error
> message:

Looks like your stderr was being redirected to your stdout; this
particular error aside, that is likely to cause weird protocol problems
for any error that git outputs.

> Jun  4 16:12:05 xyz git-daemon[10246]: unable to access
> '/root/.config/git/config': Permission denied
> 
> It's not the first time i've seen it but i've been able to ignore it
> before. This is running as a local user (as in not root) and this user
> shouldn't have access to /root. But i eventually had to do chown o+x
> /root to workaround this error.

The problem is that you have presumably dropped privileges in the daemon
instance, but your $HOME environment variable still points to /root. Git
cannot read all of its config files (nor even find out if they exist),
so it bails rather than continue.

Older versions of git silently ignored errors reading config files, but
it was tightened in v1.8.1.1, as there can be quite serious implications
to failing to read expected config (e.g., imagine transfer.fsckobjects,
or receive.deny* is ignored).

However, since changing user id and leaving $HOME is so common, there is
a patch under consideration to loosen the check only for the case of
EACCES on files in $HOME. That commit is 4698c8f (config: allow
inaccessible configuration under $HOME, 2013-04-12); it's not yet in any
released version of git, though.

In the meantime, the suggested workaround is to set $HOME for the
git-daemon user, rather than loosening /root.

-Peff
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] dir.c: fix ignore processing within not-ignored directories

2013-06-04 Thread Karsten Blees

Am 02.06.2013 21:25, schrieb Junio C Hamano:
> Duy Nguyen  writes:
> 
>>> +   then
>>> +   false
>>> +   fi
>>> +'
>>
>> Nit pick, maybe this instead?
>>
>> test_must_fail grep "^one/a.1" output
> 
> Neither.
> 
>   ! grep "^one/a.1" output
> 

Nice. I actually tried "!" but without the space - which didn't work so I took 
the other t3001 tests as example...
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

:6/4/2013 5:25:42 PM:

2013-06-04 Thread Matthew Mendell

http://thielois.free.fr/kk/miwrmacrfkhwwidtajnzttpnim.bnopj
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: * glibc detected * git: double free or corruption (fasttop): 0x0000000001fab820 ***