Jason St. John jstj...@purdue.edu writes:
On Wed, Mar 19, 2014 at 8:09 PM, 乙酸鋰 ch3co...@gmail.com wrote:
Hi,
Where to find git 1.9.1 tarball?
It is not uploaded to google code.
--
You can download a tarball for 1.9.1 from GitHub:
https://github.com/git/git/archive/v1.9.1.tar.gz
Jason
Hello, Michael, Matthieu and peff,
My name is Yao and I am interested in Git Configuration API Improvements listed
in idea page in Git. I came up some ideas and really want to discuss them with
you.
First is about when to start reading configuration file to cache. My idea is
the time user
On 3/19/2014 11:46 AM, Junio C Hamano wrote:
Ilya Bobyr ilya.bo...@gmail.com writes:
I can not find this particular patch in the latest What's cooking email.
Is there something I can do?
IIRC, I think I was waiting for the version with a new Usage text
section to the documentation you alluded
Built-in commands can specify names for option arguments when usage text
is generated for a command. sh based commands should be able to do the
same.
Option argument name hint is any text that comes after [*=?!] after the
argument name up to the first whitespace. Underscores are replaced with
On 03/20/2014 08:23 AM, Yao Zhao wrote:
Third one is about when to write back to file, I am really confused
about it. I think one way could be when user leave git repository
using cd to go back. But I am not sure if git could detect user
calls cd to leave repository.
I don't understand. The
Hi,
Yao Zhao zhaox...@umn.edu writes:
First is about when to start reading configuration file to cache. My
idea is the time user starts call command that need configuration
information (need to read configuration file).
I'd actually load the configuration lazily, when Git first requires a
Hi again guys,
I forgot to add the signed-of line to the tiny patch I sent earlier for GSOC.
Any ideas about the changes?
Thanks...
Signed-off-by: George Papanikolaou g3orge@gmail.com
---
builtin/apply.c | 25 +
1 file changed, 9 insertions(+), 16 deletions(-)
diff
Hello and welcome. See the comments inline.
On 03/20/2014 02:32 AM, George Papanikolaou wrote:
Hi fellows,
I'm planning on applying on GSOC 2014...
I tried my luck with that kinda weird microproject about inefficiencies,
and I think I've discovered some.
(also on a totally different
Hi, I have completed one of the microprojects -14th one: Change
fetch-pack.c:filter_refs() to use starts_with() instead of memcmp(). The only
line in the function filter_refs() containing memcmp() is changed with
starts_with(). I plan to apply for GSoC 2014. Any feedback is appreciated.
Last time I tried this was more than two years ago [1]. It's time for
another try and see if the community has any interest in it.
The command is straight forward, it's a ls-like version for listing
things in git. It respects $LS_COLORS and does column output like GNU
ls. ls shows cached entries
This could could help highlight files in ls-files or status output, or
even diff --name-only (but that's questionable).
This code is from coreutils.git commit
7326d1f1a67edf21947ae98194f98c38b6e9e527 file src/ls.c. This is the
last GPL-2 commit before coreutils turns to GPL-3.
The code is
Signed-off-by: Nguyễn Thái Ngọc Duy pclo...@gmail.com
---
ls_colors.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/ls_colors.c b/ls_colors.c
index 6385446..d492ab3 100644
--- a/ls_colors.c
+++ b/ls_colors.c
@@ -401,6 +401,11 @@ static void put_indicator(const struct bin_str *ind)
git-compat-util.h does not seem to carry S_IXUGO. Anyway as far as Git
is concerned, we only care one executable bit. Hard code it.
Signed-off-by: Nguyễn Thái Ngọc Duy pclo...@gmail.com
---
ls_colors.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/ls_colors.c
After coreutils moved to GPL-3 a couple more color descriptors were
added. parse_ls_color() will abort if it finds any of these so just
add them recognized (but never actually use them).
Reference commits (in coreutils.git)
0df338f (ls --color: do not colorize files with multiple hard links by
Signed-off-by: Nguyễn Thái Ngọc Duy pclo...@gmail.com
---
builtin/ls-files.c | 38 --
1 file changed, 36 insertions(+), 2 deletions(-)
diff --git a/builtin/ls-files.c b/builtin/ls-files.c
index 47c3880..463280e 100644
--- a/builtin/ls-files.c
+++
Default pathspec behavior is recursive which includes too many files
for effective column output. But if you can do
git ls-files --column ':(glob)*'
to limit to one level only. It's not exactly the same as GNU ls
(e.g. directories are never shown) but much closer.
Signed-off-by: Nguyễn Thái
The use case in mind is --max-depth=0 to stop recursion. With this we can do
git config --global alias.ls 'ls-files --column --color --max-depth=0'
and have git ls with an output very similar to GNU ls. Another
interesting one is
git config --global alias.lso 'ls-files --column --color
Hello,
The porcelain format of git status is described as not based on user
configuration.
But with --branch, behind/ahead are translated following the user's locale.
Is it normal that scripts need to take care of that?
Thanks.
--
To unsubscribe from this list: send the line unsubscribe git in
Hi,
Thanks for the feedback,
On Thu, Mar 20, 2014 at 11:58 AM, Michael Haggerty mhag...@alum.mit.edu wrote:
Why is this an improvement? Do you expect this function to be called
often for empty lines (as opposed, for example, to lines consisting
solely of whitespace characters)?
Yes, you
Nguyễn Thái Ngọc Duy pclo...@gmail.com writes:
git-compat-util.h does not seem to carry S_IXUGO. Anyway as far as Git
is concerned, we only care one executable bit. Hard code it.
Why not use S_IXUSR instead of a hardcoded value? (already used in
path.c, so shouldn't be a problem wrt
On 03/19/2014 04:21 PM, Eric Sunshine wrote:
Thanks for the resubmission. Comments below...
On Tue, Mar 18, 2014 at 10:33 AM, Aleksey Mokhovikov
moxobu...@gmail.com wrote:
Subject: [PATCH][GSOC] Selection of the verbose message is replaced with
generated message in install_branch_config()
Nguyễn Thái Ngọc Duy pclo...@gmail.com writes:
+int cmd_ls(int argc, const char **argv, const char *cmd_prefix)
+{
+ struct argv_array av = ARGV_ARRAY_INIT;
+ argv_array_pushl(av, ls-files,
+ --color, --column, --max-depth=0,
+
Sylvestre Ledru sylvestre at mozilla.com writes:
As attachment, the backtrace. I removed about 87250 calls to the
name_rev function. I guess that is a potential source of problem.
FYI, ulimit -s 10 (increase the stack size) fixes the issue.
Sylvestre
--
To unsubscribe from this list:
git status --branch --porcelain displays the status of the branch
(ahead, behind, gone), and used gettext to translate the string.
Use hardcoded strings when --porcelain is used, but keep the gettext
translation for git status --short which is essentially the same, but
meant to be read by a
On Thu, Mar 20, 2014 at 6:46 PM, Matthieu Moy
matthieu@grenoble-inp.fr wrote:
Nguyễn Thái Ngọc Duy pclo...@gmail.com writes:
git-compat-util.h does not seem to carry S_IXUGO. Anyway as far as Git
is concerned, we only care one executable bit. Hard code it.
Why not use S_IXUSR instead of
On Wed, Mar 19, 2014 at 12:16 PM, Chris Angelico ros...@gmail.com wrote:
Two parts to the question, then. Firstly, is it acceptable to use 'git
config' for a hook like this? And secondly, either: Is there a naming
convention to follow? or, what alternative would you recommend?
1. I would say
User config file location now complies with the XDG base directory specification
Signed-off-by: Astril Hayato astrilhay...@gmail.com
---
Documentation/gitk.txt | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/Documentation/gitk.txt b/Documentation/gitk.txt
index
Hi,
Sorry for this late reply, I was busy for past few days.
On Fri, Mar 14, 2014 at 12:34 PM, Jeff King p...@peff.net wrote:
On Wed, Mar 12, 2014 at 04:19:23PM +0800, Yuxuan Shui wrote:
I'm Yuxuan Shui, a undergraduate student from China. I'm applying for
GSoC 2014, and here is my proposal:
Am 19.03.2014 01:46, schrieb sza...@chromium.org:
This adds a Windows implementation of pread. Note that it is NOT
safe to intersperse calls to read() and pread() on a file
descriptor.
This is a bad idea. You're basically fixing the multi-threaded issue twice,
while at the same time breaking
We offer all purpose loan at 3% interest rate. Contact Us for more details by
Email:santanderfinancegr...@gmail.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
Hello,
My name is Guanglin Xu. I am a second-year master student at Sun
Yat-sen University in China, majoring in Software Engineering. I am
also a perspective PhD student matriculated in the US. I'm planning on
doing summer projects which I can work remotely. The GSoC 2014 program
of Git project
On Thu, Mar 20, 2014 at 6:54 AM, Karsten Blees karsten.bl...@gmail.com wrote:
Am 19.03.2014 01:46, schrieb sza...@chromium.org:
This adds a Windows implementation of pread. Note that it is NOT
safe to intersperse calls to read() and pread() on a file
descriptor.
This is a bad idea. You're
Hello,
I have completed my microproject under the guidance of Eric, After
going through the code and
previous mailing lists. I have drafted my Proposal. Still going
through the code
as of now and figuring things out. Would be great to have your
suggestions on my proposal, so that i can improve it
Jonathan Nieder jrnie...@gmail.com writes:
Junio C Hamano wrote:
...
I am a bit reluctant to name the helper sane_echo to declare echo
that interprets backslashes in the string is insane, though. For
these print a single line uses, we are only interested in using a
subset of the features
On Thu, Mar 20, 2014 at 11:53 PM, Kevin i...@ikke.info wrote:
On Wed, Mar 19, 2014 at 12:16 PM, Chris Angelico ros...@gmail.com wrote:
Two parts to the question, then. Firstly, is it acceptable to use 'git
config' for a hook like this? And secondly, either: Is there a naming
convention to
Chris Angelico ros...@gmail.com writes:
file. It doesn't really care about the full history, and wants to be
reasonably fast (as the user is waiting for it). It's just a
convenience, so correctness isn't a huge issue. The easiest way to
keep it moving through quickly is to limit the search:
On 03/11/2014 10:41 PM, Brad King wrote:
On Tue, Mar 11, 2014 at 4:06 PM, Junio C Hamano gits...@pobox.com wrote:
I may be misremembering things, but your first sentence quoted above
was exactly my reaction while reviewing the original change, and I
might have even raised that as an issue
Duy Nguyen pclo...@gmail.com writes:
On Thu, Mar 20, 2014 at 5:11 AM, Junio C Hamano gits...@pobox.com wrote:
...
I know that the 512MiB default for the bitFileThreashold (aka
forget about delta compression) came out of thin air. It was just
1GB is always too huge for anybody, so let's cut
On Fri, Mar 21, 2014 at 3:53 AM, Junio C Hamano gits...@pobox.com wrote:
Chris Angelico ros...@gmail.com writes:
file. It doesn't really care about the full history, and wants to be
reasonably fast (as the user is waiting for it). It's just a
convenience, so correctness isn't a huge issue.
Dragos Foianu dragos.foi...@gmail.com writes:
parse_dirstat_params() goes through a chain of if statements using
strcmp to parse parameters. When the parameter is a digit, the
value must go through all comparisons before the function realises
it is a digit. Optimise this logic by only going
Hi,
I have gone through commit.c, builtin/commit.c and api-config.txt but one
thing I cannot find is which functions handle writing config file to disk
after adding a new variable,value pair(for example git config my.option
true) . It is also marked TODO on the api-config.txt file. Can somebody
Matthieu Moy matthieu@grenoble-inp.fr writes:
Why?
(In general, explaining why you chose something is more important than
explaining what you chose)
Good educational comment. Thanks.
A tree (AST, Abstract syntax tree) can be interesting if you have some
source-to-source
Duy Nguyen pclo...@gmail.com writes:
On Thu, Mar 20, 2014 at 6:46 PM, Matthieu Moy
matthieu@grenoble-inp.fr wrote:
Nguyễn Thái Ngọc Duy pclo...@gmail.com writes:
git-compat-util.h does not seem to carry S_IXUGO. Anyway as far as Git
is concerned, we only care one executable bit. Hard
Matthieu Moy matthieu@imag.fr writes:
git status --branch --porcelain displays the status of the branch
(ahead, behind, gone), and used gettext to translate the string.
Use hardcoded strings when --porcelain is used, but keep the gettext
translation for git status --short which is
Astril Hayato astrilhay...@gmail.com writes:
User config file location now complies with the XDG base directory
specification
Signed-off-by: Astril Hayato astrilhay...@gmail.com
---
Documentation/gitk.txt | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git
Hiroyuki Sano sh19910...@gmail.com writes:
fsck_tree() has two different ways to set a flag,
which are the followings:
1. Using a if-statement that guards assignment.
2. Using a bitwise-or assignment operator.
Currently, many with the former way,
and one with the latter way.
In
Hi all,
This is a first draft of my Proposal for GSoC, I'd love feedback about
what I might be missing and any other files I should read regarding
this, so far I have read most of tag.c, branch.c,
builtin/for-each-ref.c, parse-options.c. once again I hope I can get
the same amount of helpful
Ilya Bobyr ilya.bo...@gmail.com writes:
Built-in commands can specify names for option arguments when usage text
is generated for a command. sh based commands should be able to do the
same.
Option argument name hint is any text that comes after [*=?!] after the
argument name up to the
Junio C Hamano gits...@pobox.com writes:
In other words, it is brief and to the point merely _allows_ these
statements to be expressed in the latter form; it does not say
anything about which is better between the former and the latter.
In any case, that is a minor point. I'll queue the
Ordinarily, we would say VAR=VAL command to execute a tested command with
environment variable(s) set only for that command. This however does not work
if 'command' is a shell function (most notably 'test_must_fail'); the result
of the assignment is retained and affects later commands.
To avoid
Nguyễn Thái Ngọc Duy pclo...@gmail.com writes:
Option explanation is in rev-list-options.txt. The interaction with -z
is left undecided.
Signed-off-by: Nguyễn Thái Ngọc Duy pclo...@gmail.com
---
Thanks.
* Revert back to the old option name --show-linear-break
* Get rid of
Nguyễn Thái Ngọc Duy pclouds at gmail.com writes:
This could could help highlight files in ls-files or status output, or
even diff --name-only (but that's questionable).
This code is from coreutils.git commit
7326d1f1a67edf21947ae98194f98c38b6e9e527 file src/ls.c. This is the
last GPL-2
Removing the bloat of checking for both '\r' and '\n' with the prettier
iswspace() function which checks for other characters as well. (read: \f \t \v)
---
This is one more try to clean up this fuzzy_matchlines() function as part of a
microproject for GSOC. The rest more clarrified microprojects
Sorry it looks like I missed some basic documentation:
http://git-scm.com/docs/git-rebase
This issue (combining preserve merges with interactive option) is
described in the BUGS section.
Is there a way I can go about this without losing my merge commits?
On Thu, Mar 20, 2014 at 2:37 PM, Robert
Avoid the conflict between read_directory() from diff-no-index.c and
read_directory() from dir.h
Signed-off-by: Andrei Dinu mandrei.d...@gmail.com
---
I plan on applying to GSOC 2014
Submit again on the list for an older bug that I solved, to show you that
I received your feedback and I
On Thu, Mar 20, 2014 at 4:44 AM, Ilya Bobyr ilya.bo...@gmail.com wrote:
Built-in commands can specify names for option arguments when usage text
is generated for a command. sh based commands should be able to do the
same.
Option argument name hint is any text that comes after [*=?!] after
Currently git blame has a big problem finding copies and moves when you
split up a big file into smaller ones. One example in the git repository
is 2cf565c, which split the documentation into smaller files.
In 582aa00 XDF_NEED_MINIMAL was removed as the default for performance
reasons, mainly for
Michael Andreen h...@ruin.nu writes:
There hasn't been any arguments against this patch. Just updated the message
with a note about --no-minimal.
There hasn't been any argument for this patch, either.
It is not like we are still in year 2007; timing result in a small
project like Git itself
Дилян Палаузов dilyan.palau...@aegee.org writes:
diff --git a/Makefile b/Makefile
index dddaf4f..dce4694 100644
--- a/Makefile
+++ b/Makefile
@@ -59,9 +59,9 @@ all::
# FreeBSD can use either, but MinGW and some others need to use
# libcharset.h's locale_charset() instead.
#
-# Define
On Thursday, March 20, 2014 01:45:21 PM Junio C Hamano wrote:
There hasn't been any argument for this patch, either.
It is not like we are still in year 2007; timing result in a small
project like Git itself is not a good enough argument to change a
well established default at this late in
Junio C Hamano gits...@pobox.com writes:
Duy Nguyen pclo...@gmail.com writes:
The only
downside I see is large blobs will be packed undeltified, which could
increase pack size if you have lots of them.
I think that is something that can be tweaked, unless the user tells
us otherwise via
Am 20.03.2014 17:08, schrieb Stefan Zager:
On Thu, Mar 20, 2014 at 6:54 AM, Karsten Blees karsten.bl...@gmail.com
wrote:
Am 19.03.2014 01:46, schrieb sza...@chromium.org:
This adds a Windows implementation of pread. Note that it is NOT
safe to intersperse calls to read() and pread() on a
On Thu, Mar 20, 2014 at 2:35 PM, Karsten Blees karsten.bl...@gmail.com wrote:
Am 20.03.2014 17:08, schrieb Stefan Zager:
Going forward, there is still a lot of performance that gets left on
the table when you rule out threaded file access. There are not so
many calls to read, mmap, and pread
On 03/20/2014 10:09 PM, Junio C Hamano wrote:
* ap/remote-hg-skip-null-bookmarks (2014-03-19) 1 commit
- remote-hg: do not fail on invalid bookmarks
Will merge to 'next'.
Hmm, am I the only one who has 11 failures in test-hg-hg-git.sh,
like this:
(Tested under Debian 7, commit
Signed-off-by: Justin Lebar jle...@google.com
---
Documentation/CodingGuidelines | 4 ++--
Documentation/config.txt| 6 +++---
Documentation/diff-generate-patch.txt | 2 +-
Documentation/diff-options.txt | 2
Signed-off-by: Justin Lebar jle...@gmail.com
---
Makefile| 2 +-
builtin/apply.c | 2 +-
builtin/checkout.c | 2 +-
builtin/log.c | 2 +-
builtin/pack-objects.c
Signed-off-by: Justin Lebar jle...@google.com
---
contrib/examples/git-commit.sh | 2 +-
contrib/svn-fe/svn-fe.txt | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/contrib/examples/git-commit.sh b/contrib/examples/git-commit.sh
index 23ffb02..4aab1a6 100755
---
Junio C Hamano gits...@pobox.com writes:
Quite a few topics are still outside 'pu' and I suspect some of the
larger ones deserve deeper reviews to help moving them to 'next'.
In principle, I'd prefer to keep any large topic that touch core
part of the system cooking in 'next' for at least a
David Kastrup d...@gnu.org writes:
Junio C Hamano gits...@pobox.com writes:
Duy Nguyen pclo...@gmail.com writes:
The only
downside I see is large blobs will be packed undeltified, which could
increase pack size if you have lots of them.
I think that is something that can be tweaked,
Torsten Bögershausen tbo...@web.de writes:
On 03/20/2014 10:09 PM, Junio C Hamano wrote:
* ap/remote-hg-skip-null-bookmarks (2014-03-19) 1 commit
- remote-hg: do not fail on invalid bookmarks
Will merge to 'next'.
Hmm, am I the only one who has 11 failures in test-hg-hg-git.sh,
Still?
On Thu, Mar 20, 2014 at 6:16 PM, Justin Lebar jle...@google.com wrote:
Signed-off-by: Justin Lebar jle...@gmail.com
---
Makefile| 2 +-
builtin/apply.c | 2 +-
builtin/checkout.c | 2 +-
builtin/log.c
I got annoyed by git's awkward use of nor in man pages and in git add -p, so
I went ahead and audited all uses of nor in the tree. One might be able to
argue that some of the uses I've changed are technically acceptable, but that's
a pretty low bar to set for ourselves. I aimed to make
Signed-off-by: Justin Lebar jle...@gmail.com
---
builtin/clean.c | 6 +++---
builtin/commit.c | 2 +-
git-add--interactive.perl | 4 ++--
perl/Git/SVN.pm | 4 ++--
sha1_file.c | 2 +-
t/t1001-read-tree-m-2way.sh
From: Jeff King p...@peff.net
On Thu, Feb 20, 2014 at 08:35:33AM +0100, Ephrim Khong wrote:
Hi, git log seems to omit merge commits that delete a file
if --follow or
--diff-filter=D is given. Below is a testcase. I'm not sure if it is
desired
behaviour for --diff-filter=D, but it's probably
On Wed, Mar 19, 2014 at 10:28:46AM -0700, Junio C Hamano wrote:
[git config --file versus GIT_CONFIG=]
Thanks. Then I think it makes sense to do such a conversion but it
probably should be done on top of this patch (we could do it before
this patch), not as a part of this patch.
Here's a
Once upon a time, the setting of GIT_CONFIG in the
environment could affect how tests ran. Commit 9c3796f (Fix
setting config variables with an alternative GIT_CONFIG,
2006-06-20) unconditionally set GIT_CONFIG in the Makefile
when running tests to give us a known starting point.
This is
Thanks for the quick reply.
When I send a new patch, should I fold these changes into the original
commit, or should I send them as a separate commit?
diff --git a/builtin/apply.c b/builtin/apply.c
index b0d0986..6013e19 100644
--- a/builtin/apply.c
+++ b/builtin/apply.c
@@ -4061,7 +4061,7
This is already handled by the mass GIT_* unsetting added by
95a1d12 (tests: scrub environment of GIT_* variables,
2011-03-15).
Signed-off-by: Jeff King p...@peff.net
---
t/test-lib.sh | 1 -
1 file changed, 1 deletion(-)
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 1531c24..625f06e 100644
Several test scripts manually unset GIT_CONFIG and other
GIT_* variables. These are generally taken care of for us by
test-lib.sh already.
Unsetting these is not only useless, but can be confusing to
a reader, who may wonder why some tests in a script unset
them and others do not (t0001 is
This lets us get rid of an extra env invocation in the
middle, and is slightly more readable.
Signed-off-by: Jeff King p...@peff.net
---
The case that started this all...
This is also the only reason this series needs to go on top of David's
patch.
t/t1300-repo-config.sh | 8
1 file
Doing:
GIT_CONFIG=foo git config ...
is equivalent to:
git config --file=foo ...
The latter is easier to read and slightly less error-prone,
because of issues with one-shot variables and shell
functions (e.g., you cannot use the former with
test_must_fail).
Note that we explicitly leave
t0001 predates the test_path_is_* helpers, and uses test
-f and test -d directly. Using the helpers provides
better debugging output, and are a little more robust.
As opposed to ! test -d, test_path_is_missing will
actually makes sure the path does not exist at all.
Signed-off-by: Jeff King
In the final test of t0001, we have a repo whose .git is a
symlink to a directory here, and we use
--separate-git-dir to migrate that to a .git file pointing
to a different directory. We check that the data is migrated
to the new directory and that .git looks like a git-file.
We also check that
We hand-set several config options using :
git config -f $HOME/.gitconfig ...
Instead, we can use test_config_global. Not only is this
more readable, but it cleans up for us so that subsequent
tests aren't polluted by our settings.
Signed-off-by: Jeff King p...@peff.net
---
t/t0001-init.sh |
On 3/20/2014 11:38 AM, Junio C Hamano wrote:
Ilya Bobyr ilya.bo...@gmail.com writes:
Built-in commands can specify names for option arguments when usage text
is generated for a command. sh based commands should be able to do the
same.
Option argument name hint is any text that comes after
We've hand-rolled several if statements looking for
failures. We can use test_must_fail here, which is shorter
and more robust.
Note that we modify the commands slightly (to use git init
foo rather than cd foo git init) to avoid dealing with
a subshell, but this should not affect the outcome.
Many tests use subshells, but don't actually change the
shell environment. They were probably cargo-culted from
earlier tests which did need subshells. Drop the useless
ones.
Signed-off-by: Jeff King p...@peff.net
---
These ones should produce no behavior change at all; they're purely
mechanical
Many tests do something like:
(
mkdir foo
cd foo
git init
)
You can do the same these days with git init foo, which
makes the tests shorter and simpler to read.
Signed-off-by: Jeff King p...@peff.net
---
Unlike the last patch, this one _could_ have an affect. I
Junio and Eric,
Thank you for the reviews and helpful advice.
I should have read more past commit messages before send patch.
Regards,
2014-03-21 3:20 GMT+09:00 Junio C Hamano gits...@pobox.com:
Hiroyuki Sano sh19910...@gmail.com writes:
fsck_tree() has two different ways to set a flag,
On Wed, Mar 19, 2014 at 10:34:29PM +, Dragos Foianu wrote:
The name_rev function recursively calls itself which is why the backtrace is
so big. Unfortunately, for repos with long histories it can lead to Stack
Overflows. This is pretty much what happened in your case.
I tested it on my
On Fri, Mar 21, 2014 at 03:51:16AM +1100, Chris Angelico wrote:
1. I would say yes. git config is made to be extended and doesn't
require a config item to be known.
2. Namespacing the config items like you did is a good thing to do so
it won't interfere with other options.
Excellent!
On Wed, Mar 19, 2014 at 01:38:32PM +0100, David Kastrup wrote:
The default of 16MiB causes serious thrashing for large delta chains
combined with large files.
Does it make much sense to bump this without also bumping
MAX_DELTA_CACHE in sha1_file.c? In my measurements of linux.git, bumping
the
On Thu, Mar 20, 2014 at 3:52 PM, Andrei Dinu mandrei.d...@gmail.com wrote:
Avoid the conflict between read_directory() from diff-no-index.c and
read_directory() from dir.h
This message is lacking a bit of information. Since diff-no-index.c
does not presently include dir.h, the reader is left to
On Wed, Mar 19, 2014 at 9:44 AM, Othman Darraz darraz...@gmail.com wrote:
Subject: fsck.c:fsck_commit() use skip_prefix() and starts_with()
Add a space after the colon. Add a colon after fsck_commit().
make buffer const char* because the content of the memory referenced by this
variable
On Wed, Mar 19, 2014 at 7:12 PM, Dragos Foianu dragos.foi...@gmail.com wrote:
Eric Sunshine sunshine at sunshineco.com writes:
On Tue, Mar 18, 2014 at 6:31 PM, Eric Sunshine sunshine at
sunshineco.com wrote:
One other observation: You have a one-off error in your out-of-bounds
check. It
On Thu, Mar 20, 2014 at 8:40 PM, Eric Sunshine sunsh...@sunshineco.com wrote:
On Wed, Mar 19, 2014 at 7:12 PM, Dragos Foianu dragos.foi...@gmail.com
wrote:
Eric Sunshine sunshine at sunshineco.com writes:
On Tue, Mar 18, 2014 at 6:31 PM, Eric Sunshine sunshine at
sunshineco.com wrote:
One
On Wed, Mar 19, 2014 at 03:39:42PM -0700, Junio C Hamano wrote:
Jeff King p...@peff.net writes:
On Fri, Mar 07, 2014 at 08:08:37AM +0100, Christian Couder wrote:
Be it graft or replace, I do not think we want to invite people to
use these mechansims too lightly to locally rewrite
On Fri, Mar 21, 2014 at 2:15 AM, Junio C Hamano gits...@pobox.com wrote:
* Get rid of saved_linear, use another flag in struct object instead
I cannot offhand say if I like this change or not. A flag bit is a
scarce and limited resource; commit slabs felt more suited for
implementation of
On Fri, Mar 21, 2014 at 4:56 AM, Stefan Zager sza...@chromium.org wrote:
Considering all that, Duy's solution of opening separate file descriptors
per thread seems to be the best pattern for future multi-threaded work.
Does that mean you would endorse the (N threads) * (M pack files)
On Thu, Mar 20, 2014 at 11:08 PM, Stefan Zager sza...@chromium.org wrote:
Duy, would you like to re-post your patch without the new pread
implementation?
I will but let me try out the sliding window idea first. My quick
tests on git.git show me we may only need 21k mmap instead of 177k
pread.
1 - 100 of 114 matches
Mail list logo