On 08/01/2015 07:12 AM, Junio C Hamano wrote:
> On Fri, Jul 31, 2015 at 8:59 PM, Michael Haggerty
> wrote:
>>
>> It seems to me that adding a new top-level "worktree-refs" directory is
>> pretty traumatic. Lots of people and tools will have made the assumption
>> that all "normal" references live
On Thu, Jul 30, 2015 at 2:57 AM, Matthieu Moy
wrote:
> Karthik Nayak writes:
>
>> What I was thinking of was something like this :
>>
>> struct strbuf format = STRBUF_INIT;
>> char c = ' ';
>> if (current)
>> c = '*';
>> strbuf_addf(&format, "%c", c, other format options...);
>> show_ref_
On Thu, Jul 30, 2015 at 2:51 AM, Matthieu Moy
wrote:
> Junio C Hamano writes:
>
>> Junio C Hamano writes:
>>
Couldn't think of a better replacer, any suggestions would be welcome :)
>>>
>>> See below.
>>> ...
>>> One way to do all of the above is ...
>>
>> Note that is just "one way", not t
On Wed, Jul 29, 2015 at 11:30 PM, Junio C Hamano wrote:
> Karthik Nayak writes:
>
>>> A handful of "huh?" on the design.
>>>
>>> - The atom says "if *exists*" and explanation says "has a value".
>>>How are they related? Does an atom whose value is an empty
>>>string has a value? Or is
On Thu, Jul 30, 2015 at 11:26 PM, Eric Sunshine wrote:
>
> In an earlier review, Matthieu pointed out that this test failed to
> ensure that the 'padright' value did not leak into the next atom. In a
> subsequent version, you fixed the test to check that condition, but
> now you've somewhat lost i
On Fri, 2015-07-31 at 22:12 -0700, Junio C Hamano wrote:
> On Fri, Jul 31, 2015 at 8:59 PM, Michael Haggerty
> wrote:
> >
> > It seems to me that adding a new top-level "worktree-refs" directory is
> > pretty traumatic. Lots of people and tools will have made the assumption
> > that all "normal"
On Sat, 2015-08-01 at 06:04 +0200, Christian Couder wrote:
>
> Le 1 août 2015 09:01, "David Turner" a
> écrit :
> >
> > This is RFC because I'm not sure why show-ref only works on refs/
> (and
> > whether it should learn to look in worktree-refs/). I'm also not
> sure
> > whether there are other
On Fri, Jul 31, 2015 at 8:59 PM, Michael Haggerty wrote:
>
> It seems to me that adding a new top-level "worktree-refs" directory is
> pretty traumatic. Lots of people and tools will have made the assumption
> that all "normal" references live under "refs/".
> ...
> It's all a bit frightening, fra
On 08/01/2015 01:56 AM, David Turner wrote:
> This is RFC because I'm not sure why show-ref only works on refs/ (and
> whether it should learn to look in worktree-refs/). I'm also not sure
> whether there are other changes I should make to refs.c to handle
> per-worktree refs; I basically did the
Stefan Beller writes:
> -module_list()
> +module_list_shell()
> {
> eval "set $(git rev-parse --sq --prefix "$wt_prefix" -- "$@")"
> (
> @@ -187,6 +187,29 @@ module_list()
> '
> }
>
> +module_list()
> +{
> + # call both the old and new code
> + module_list_shell $@ >
Stefan Beller writes:
> +static const char * const git_submodule_helper_usage[] = {
> + N_("git submodule--helper --module_list [...]"),
Yuck. Please do not force --multi_word_opt upon us, which is simply
too ugly to live around here. --module-list is perhaps OK, but
because submodule--hel
On Sat, Aug 1, 2015 at 12:04 AM, Junio C Hamano wrote:
> Paul Tan writes:
>
>> I think I will introduce a format_patch() function that takes a single
>> commit-ish so that we can use tag names to name the patches:
>>
>> # Given a single commit $commit, formats the following patches with
>> # git-
We check the return value of read_ref in 19 out of 21 cases.
This adds checks to the missing cases.
Signed-off-by: Stefan Beller
---
transport-helper.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/transport-helper.c b/transport-helper.c
index 5d99a6b..68e498e 10064
We need a place to stick refs for bisects in progress that is not
shared between worktrees. So we use the worktree-refs/ hierarchy
instead of the refs/.
To do this, load loose refs from "worktree-refs/" as well as from "refs/".
The is_per_worktree_ref function and associated docs learn that
workt
Using the new worktree-refs/ refs, make bisection per-worktree.
git show-ref presently only handles refs under refs/, so we change
git bisect to use git rev-parse instead.
Signed-off-by: David Turner
---
Documentation/git-bisect.txt | 4 ++--
Documentation/rev-list-options.txt | 14 +
This is RFC because I'm not sure why show-ref only works on refs/ (and
whether it should learn to look in worktree-refs/). I'm also not sure
whether there are other changes I should make to refs.c to handle
per-worktree refs; I basically did the simplest thing I could think of
to start with.
--
T
When sending an e-mail, the client and server must
agree on an authentication mechanism. Some servers
(due to misconfiguration or a bug) denies valid
credentials for certain mechanisms. In this patch,
a new option --smtp-auth and configuration entry
smtpauth are introduced.
If smtp_auth is defined
On Sun, Jul 19, 2015 at 9:10 AM, Paul Tan wrote:
> Since 3e5057a (git am --abort, 2008-07-16), git-am supported the --abort
> option that will rewind HEAD back to the original commit. Re-implement
> this through am_abort().
>
> Since 7b3b7e3 (am --abort: keep unrelated commits since the last failu
I am sorry for being late to the review, I looked into coverity today as Duy
bugged me to fix the memory allocation stuff[1]
[1] $gmane/275046
On Thu, Jul 30, 2015 at 11:06 PM, David Turner wrote:
> +
> + if (old_sha1) {
> + unsigned char actual_old_sha1[20];
> +
From: Jacob Keller
This series adds a default merge strategy option for git-notes, so that
the user does not have to type "-s" every time. It is overridden by the
-s option.
I also added some tests to ensure that the "--abort" "--commit" and "-s"
options must be independent. In addition, I found
From: Jacob Keller
Teach documentation about the cat_sort_uniq rewriteMode that got added
at the same time as the equivalent merge strategy.
Signed-off-by: Jacob Keller
Cc: Johan Herland
Cc: Michael Haggerty
Cc: Eric Sunshine
---
Documentation/config.txt| 4 ++--
Documentation/git-notes
From: Jacob Keller
Teach git-notes about a new configuration option "notes.merge" for
selecting the default notes merge strategy. Document the option in
config.txt and git-notes.txt
Add tests for the configuration option. Ensure that command line
--strategy option overrides the configured value.
Most of the submodule operations work on a set of submodules.
Calculating and using this set is usually done via:
module_list "$@" | {
while read mode sha1 stage sm_path
do
# the actual operation
done
}
Currently the function `module_
Signed-off-by: Stefan Beller
---
Notes:
The output of the differential testing is below,
which points out 2 bugs:
* resolving relative paths seems is broken, so we
would need to have the equivalent of
eval "set $(git rev-parse --sq --prefix "$wt_prefix" -- "$@")"
(
Hi,
so eventually we want to have the whole Android project inside a git repository,
which includes > 500 submodules for the different sub systems. To make that
feasable
we want to improve git-submodule for a huge set of submodules such as
parallelizing
`git submodule update`. As a first step I
On Fri, Jul 31, 2015 at 6:11 PM, David Turner wrote:
> Add a new function, find_shared_symref, which contains the heart of
> die_if_checked_out, but works for all symrefs. Refactor
Slightly more explanatory:
..., but works for any symref, not just HEAD. Refactor
More below.
> die_if_check
Add a new function, find_shared_symref, which contains the heart of
die_if_checked_out, but works for all symrefs. Refactor
die_if_checked_out to use the same infrastructure as
find_shared_symref.
Soon, we will use find_shared_symref to protect notes merges in
worktrees.
Signed-off-by: David Tur
Before creating NOTES_MERGE_REF, check NOTES_MERGE_REF using
find_shared_symref and die if we find one. This prevents simultaneous
merges to the same notes branch from different worktrees.
Signed-off-by: David Turner
---
builtin/notes.c | 5 +++
t/t3320-notes-merge-worktrees.s
On Fri, 2015-07-31 at 17:36 -0400, Eric Sunshine wrote:
> On Fri, Jul 31, 2015 at 5:15 PM, David Turner
> wrote:
> > On Fri, 2015-07-31 at 15:35 -0400, Eric Sunshine wrote:
> >> On Fri, Jul 31, 2015 at 3:01 PM, David Turner
> >> wrote:
> >> > Add a new function, die_if_shared_symref, which work
David Turner writes:
> I further propose to add some patches (to go on top of the pseudorefs
> series):
>
> (a) include worktree-refs/ in per_worktree_refs and include
> worktree-refs/ in per_worktree_refs in common_list, and
Do you mean .git/worktree-refs/, just next to .git/refs? Then I
think
> I think I have identified an obscure bug. I have a reproducible test case.
> I am trying to come up with enough of a test case that can be used for
> finding and fixing the issue.
BTW, I have tested this on 1.7.6.6 and 2.5.0. Same results on either version.
--
To unsubscribe from this list: sen
On Fri, 2015-07-31 at 11:46 -0700, Junio C Hamano wrote:
> David Turner writes:
>
> > Before creating NOTES_MERGE_REF, check NOTES_MERGE_REF using
> > die_if_shared_symref. This prevents simultaneous merges to the same
> > notes branch from different worktrees.
> >
> > Signed-off-by: David Turne
On Fri, Jul 31, 2015 at 5:15 PM, David Turner wrote:
> On Fri, 2015-07-31 at 15:35 -0400, Eric Sunshine wrote:
>> On Fri, Jul 31, 2015 at 3:01 PM, David Turner
>> wrote:
>> > Add a new function, die_if_shared_symref, which works like
>> > die_if_checked_out, but for all references. Refactor
>>
On Thu, Jul 30, 2015 at 11:58:05AM +0200, Matthieu Moy wrote:
> "brian m. carlson" writes:
>
> > On Mon, Jul 27, 2015 at 02:23:04PM -0700, Junio C Hamano wrote:
> >> * bc/object-id (2015-06-17) 10 commits
> > Is there anything I can do to make this series less painful (e.g. a
> > reroll or such)?
I think I have identified an obscure bug. I have a reproducible test case.
I am trying to come up with enough of a test case that can be used for
finding and fixing the issue.
I have a set of patches. They are diffs. Some have whitespace issues.
(Which I believe triggers the bug.)
The patches go
On Fri, 2015-07-31 at 15:35 -0400, Eric Sunshine wrote:
> On Fri, Jul 31, 2015 at 3:01 PM, David Turner
> wrote:
> > Add a new function, die_if_shared_symref, which works like
> > die_if_checked_out, but for all references. Refactor
> > die_if_checked_out to work in terms of die_if_shared_symref
On Fri, Jul 31, 2015 at 3:21 PM, Jacob Keller wrote:
> Teach git-notes about a new configuration option "notes.merge" for
> selecting the default notes merge strategy. Document the option in
> config.txt and git-notes.txt
>
> Add tests for the configuration option. Ensure that command line
> --str
On Fri, Jul 31, 2015 at 12:16:46 -0700, Junio C Hamano wrote:
> Or even "git remote get url [$there]", "git remote get push-url [$there]".
Looking at `git remote`'s existing subcommands, consistency there would
be something like:
git remote get-url $there
git remote get-url --push $there
Eric Sunshine writes:
> On Fri, Jul 31, 2015 at 3:01 PM, David Turner
> wrote:
>> Add a new function, die_if_shared_symref, which works like
>> die_if_checked_out, but for all references. Refactor
>> die_if_checked_out to work in terms of die_if_shared_symref.
>>
>> Soon, we will use die_if_sh
CC'ing yourself is ok, but please do not forget to ask reviews from
people who have touched the area of the code you are touching (they
are hopefully the ones who are more knowledgeable and can help you).
Old timers like I know Johan is the go-to person on things around
notes, but new people can a
On Fri, Jul 31, 2015 at 3:01 PM, David Turner wrote:
> Add a new function, die_if_shared_symref, which works like
> die_if_checked_out, but for all references. Refactor
> die_if_checked_out to work in terms of die_if_shared_symref.
>
> Soon, we will use die_if_shared_symref to protect notes merge
From: Jacob Keller
Teach documentation about the cat_sort_uniq rewriteMode that got added
at the same time as the equivalent merge strategy.
Signed-off-by: Jacob Keller
---
Documentation/config.txt| 4 ++--
Documentation/git-notes.txt | 3 ++-
2 files changed, 4 insertions(+), 3 deletions(
From: Jacob Keller
Teach git-notes about a new configuration option "notes.merge" for
selecting the default notes merge strategy. Document the option in
config.txt and git-notes.txt
Add tests for the configuration option. Ensure that command line
--strategy option overrides the configured value.
From: Jacob Keller
This small series is a precursor to some thoughts I have for enabling
easier notes collaboration. First, make it so that users can configure
the default notes merge strategy. Mostly useful if you always want to
use cat_sort_uniq.
I also found that the rewriteMode supports cat_
Ben Boeckel writes:
> On Fri, Jul 31, 2015 at 12:02:14 -0700, Junio C Hamano wrote:
>> Ben Boeckel writes:
>>
>> > With some sed, yes, but then so would `git remote show` just as useful
>> > too (and in that case, "why does --get-url exist either?" comes to
>> > mind).
>>
>> Either carelessnes
On Fri, Jul 31, 2015 at 12:02:14 -0700, Junio C Hamano wrote:
> Ben Boeckel writes:
>
> > With some sed, yes, but then so would `git remote show` just as useful
> > too (and in that case, "why does --get-url exist either?" comes to
> > mind).
>
> Either carelessness let it slip in, or it came be
Ben Boeckel writes:
> With some sed, yes, but then so would `git remote show` just as useful
> too (and in that case, "why does --get-url exist either?" comes to
> mind).
Either carelessness let it slip in, or it came before 'git remote show'.
--
To unsubscribe from this list: send the line "uns
Before creating NOTES_MERGE_REF, check NOTES_MERGE_REF using
die_if_shared_symref. This prevents simultaneous merges to the same
notes branch from different worktrees.
Signed-off-by: David Turner
---
builtin/notes.c | 2 ++
t/t3320-notes-merge-worktrees.sh | 71 +++
Add a new function, die_if_shared_symref, which works like
die_if_checked_out, but for all references. Refactor
die_if_checked_out to work in terms of die_if_shared_symref.
Soon, we will use die_if_shared_symref to protect notes merges in
worktrees.
Signed-off-by: David Turner
---
Oops, forgot
On Fri, Jul 31, 2015 at 11:40:12 -0700, Junio C Hamano wrote:
> Probably get-url makes (some) sense because ls-remote is a "fetch
> that does not actually fetch anything". But "get-push-url" to
> ls-remote makes _no_ sense whatsoever. ls-remote and fetch do not
> have to know or care about push-u
David Turner writes:
> Before creating NOTES_MERGE_REF, check NOTES_MERGE_REF using
> die_if_shared_symref. This prevents simultaneous merges to the same
> notes branch from different worktrees.
>
> Signed-off-by: David Turner
> ---
>
> This version addresses Eric Sunshine's critiques of v1. I
Ben Boeckel writes:
> Not sure if it would be better to make a new variable or to reuse the existing
> one. I'm reusing it currently because it makes it easier to ensure they
> are mutually exclusive.
>
> Please keep me CC'd to the list; I'm not subscribed.
>
> Thanks,
>
> --Ben
>
> Ben Boeckel (
Junio C Hamano writes:
> On Thu, Jul 30, 2015 at 11:31 AM, Joey Hess wrote:
>> I think this comes down to a lack of quoting where git-pull runs
>> git-fetch. Before eb2a8d9ed3fca2ba2f617b704992d483605f3bb6,
>> "$@" was passed through to git-fetch, but now there is a $upload_pack
>> which is pass
Before creating NOTES_MERGE_REF, check NOTES_MERGE_REF using
die_if_shared_symref. This prevents simultaneous merges to the same
notes branch from different worktrees.
Signed-off-by: David Turner
---
This version addresses Eric Sunshine's critiques of v1. It breaks out
the symref-checking func
With pushInsteadOf and triangle workflows, a flag to have git fully
expand the push URL is also useful since it can be very different from
the --get-url output.
Signed-off-by: Ben Boeckel
---
Documentation/git-ls-remote.txt | 7 ++-
builtin/ls-remote.c | 15 +--
2 fi
Not sure if it would be better to make a new variable or to reuse the existing
one. I'm reusing it currently because it makes it easier to ensure they
are mutually exclusive.
Please keep me CC'd to the list; I'm not subscribed.
Thanks,
--Ben
Ben Boeckel (1):
ls-remote: add --get-push-url opti
Remove a check that would disable the untracked cache for sparse
checkouts. Add tests that ensure that the untracked cache works with
sparse checkouts -- specifically considering the case that a file
foo/bar is checked out, but foo/.gitignore is not.
Signed-off-by: David Turner
---
dir.c
Am 31.07.2015 um 12:16 schrieb SZEDER Gábor:
>
> Quoting John Keeping :
>
>> On Thu, Jul 30, 2015 at 01:24:03PM +0200, SZEDER Gábor wrote:
>>>
>>> Quoting Thomas Braun :
>>>
Signed-off-by: Thomas Braun
---
> John Keeping hat am 13. Juli 2015 um 15:11
>>> geschrieben:
> git-reb
On Fri, Jul 31, 2015 at 9:11 AM, Stefan Beller wrote:
> On Fri, Jul 31, 2015 at 4:24 AM, Duy Nguyen wrote:
>> Jeff, I suppose you are the admin of git on scan.coverity, or knows
>> him/her, perhaps we can add a model for xmalloc to suppress these
>> "null pointer deferences" reports? We are sure
Stefan Beller writes:
> Signed-off-by: Stefan Beller
> ---
>
> So I was trying to understand how to implement "git add .gitmodules" as
> I intend to rewrite git submodules in C.
A request and a suggestion.
- Please keep Jens and Heiko in the loop and pick their brains ;-)
- It may not be t
On Fri, Jul 31, 2015 at 4:24 AM, Duy Nguyen wrote:
> Jeff, I suppose you are the admin of git on scan.coverity, or knows
> him/her, perhaps we can add a model for xmalloc to suppress these
> "null pointer deferences" reports? We are sure xmalloc() never returns
> NULL. Qemu did it [1] and it looks
Paul Tan writes:
> I think I will introduce a format_patch() function that takes a single
> commit-ish so that we can use tag names to name the patches:
>
> # Given a single commit $commit, formats the following patches with
> # git-format-patch:
> #
> # 1. $commit.eml: an email patch with a Mess
On 07/31/2015 12:12 AM, Junio C Hamano wrote:
> Thanks, queued.
>
> By the way, do you plan to revisit two rather large-ish stalled
> topics of yours queued on 'pu' any time soon?
I hope to revive the "tempfile" and "numparse" topics earlyish in this
release cycle if at all possible.
Michael
--
Currently, git commit --dry-run does not run any hooks, not even pre-commit:
% mkdir test
% cd test
% git init
Initialized empty Git repository in /home/eda/test/.git/
% ln -s /bin/false .git/hooks/pre-commit
% touch a
% git add a
% git commit --dry-run -m. a >/dev/null && echo yes
yes
% git commi
On Fri, Jul 31, 2015 at 12:13 PM, David Turner wrote:
> I should mention that other than that, skip-worktree + untracked cache
> seems to work fine.
Please go ahead and make a patch to allow it. I'll spend some more
time looking at this code. But I think it'll be fine.
--
Duy
--
To unsubscribe f
Jeff, I suppose you are the admin of git on scan.coverity, or knows
him/her, perhaps we can add a model for xmalloc to suppress these
"null pointer deferences" reports? We are sure xmalloc() never returns
NULL. Qemu did it [1] and it looks simple.. I think something like
this would do
void *xmallo
On Fri, Jul 31, 2015 at 4:40 AM, Matthieu Moy wrote:
> The previous code broke for example
>
> git pull --upload-pack 'echo --foo'
>
> Reported-by: Joey Hess
> Fix-suggested-by: Junio C Hamano
> Signed-off-by: Matthieu Moy
Thanks for cleaning up my mess! ><
Regards,
Paul
--
To unsubscribe f
On Wed, Jul 29, 2015 at 1:09 AM, Junio C Hamano wrote:
> Paul Tan writes:
>
>> diff --git a/t/t4153-am-resume-override-opts.sh
>> b/t/t4153-am-resume-override-opts.sh
>> new file mode 100755
>> index 000..c49457c
>> --- /dev/null
>> +++ b/t/t4153-am-resume-override-opts.sh
>> @@ -0,0 +1,144
Hi,
When discussing establishing git over Smart HTTP on pages 140 and 141:
[quote]
Finally you’ll want to make writes be authenticated somehow, possibly
with an Auth block like this:
AuthType Basic
140GitWeb
AuthName "Git Access"
AuthUserFile /opt/git/.htpasswd
Require valid-user
That will re
Quoting John Keeping :
On Thu, Jul 30, 2015 at 01:24:03PM +0200, SZEDER Gábor wrote:
Quoting Thomas Braun :
Signed-off-by: Thomas Braun
---
John Keeping hat am 13. Juli 2015 um 15:11
geschrieben:
git-rebase.sh contains:
if test "$action" = "edit-todo" && test "$type" != "inter
On Thu, Jul 30, 2015 at 05:58:04PM -0700, Junio C Hamano wrote:
> Eric Sunshine writes:
>
> > On Wed, Jul 29, 2015 at 11:51 AM, Patrick Steinhardt wrote:
> >> Test behavior of `git clone` when working with an empty path
> >> component. This may be the case when cloning a file system's root
> >>
71 matches
Mail list logo
