[PATCH 01/10] refs: add "for_each_bisect_ref" function

Signed-off-by: Christian Couder <redacted>
---
 refs.c |    5 +++++
 refs.h |    1 +
 2 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/refs.c b/refs.c
index 8d3c502..2b21148 100644
--- a/refs.c
+++ b/refs.c

@@ -662,6 +662,11 @@ int for_each_remote_ref(each_ref_fn fn, void *cb_data)
 	return do_for_each_ref("refs/remotes/", fn, 13, 0, cb_data);
 }
 
+int for_each_bisect_ref(each_ref_fn fn, void *cb_data)
+{
+	return do_for_each_ref("refs/bisect/", fn, 12, 0, cb_data);
+}
+
 int for_each_rawref(each_ref_fn fn, void *cb_data)
 {
 	return do_for_each_ref("refs/", fn, 0,

diff --git a/refs.h b/refs.h
index 29bdcec..e5d6e80 100644
--- a/refs.h
+++ b/refs.h

@@ -23,6 +23,7 @@ extern int for_each_ref(each_ref_fn, void *);
 extern int for_each_tag_ref(each_ref_fn, void *);
 extern int for_each_branch_ref(each_ref_fn, void *);
 extern int for_each_remote_ref(each_ref_fn, void *);
+extern int for_each_bisect_ref(each_ref_fn, void *);
 
 /* can be used to learn about broken ref and symref */
 extern int for_each_rawref(each_ref_fn, void *);

-- 
1.6.2.1.317.g3d804

Heya

On Thu, Mar 26, 2009 at 05:55, Christian Couder [off-list ref] wrote:

Signed-off-by: Christian Couder <redacted>

A 10 patches series with no cover letter? And no description of the
individual patches either! C'mon Christian, you know better than that
;).

-- 
Cheers,

Sverre Rabbelier

Hi Sverre,

Le jeudi 26 mars 2009, Sverre Rabbelier a écrit :

Heya

On Thu, Mar 26, 2009 at 05:55, Christian Couder [off-list ref] 

wrote:

quoted

Signed-off-by: Christian Couder <redacted>

A 10 patches series with no cover letter? 

I am not a big fan of cover letters. Usually I prefer adding comments in the 
patches.

And no description of the 
individual patches either! 

There is a commit message in each patch. And many of the patches are very 
small.

C'mon Christian, you know better than that 
;).

If some commit messages are not clear enough, please tell me and I will try 
to improve them ;)

Regards,
Christian.

Heya,

On Thu, Mar 26, 2009 at 08:48, Christian Couder [off-list ref] wrote:

I am not a big fan of cover letters. Usually I prefer adding comments in the
patches.

The downside of that is that it makes it harder to quickly scan the
series; now I have to go through each patch (which involves trying
finding which patch is next, as my MUA is retarded and doesn't
understand proper threading).

quoted

And no description of the
individual patches either!

There is a commit message in each patch. And many of the patches are very
small.

Hehe, my bad; the first one didn't have a commit message, which is the
one I looked at first.

If some commit messages are not clear enough, please tell me and I will try
to improve them ;)

The rest of the series is nicely readable, I guess I shouldn't send
whine mails before reading the entire series next time :).

-- 
Cheers,

Sverre Rabbelier

Christian Couder venit, vidit, dixit 26.03.2009 08:48:

Hi Sverre,

Le jeudi 26 mars 2009, Sverre Rabbelier a écrit :

quoted

Heya

On Thu, Mar 26, 2009 at 05:55, Christian Couder [off-list ref] 

wrote:

quoted

quoted

Signed-off-by: Christian Couder <redacted>

A 10 patches series with no cover letter? 

I am not a big fan of cover letters. Usually I prefer adding comments in the 
patches.

I'm sorry I have to say that, but your individual preferences don't
matter. Many of us would do things differently, each in their own way,
but people adjust to the list's preferences. It's a matter of attitude.
So, please...

Cheers,
Michael

quoted

And no description of the 
individual patches either! 

There is a commit message in each patch. And many of the patches are very 
small.

quoted

C'mon Christian, you know better than that 
;).

If some commit messages are not clear enough, please tell me and I will try 
to improve them ;)

Regards,
Christian.

Hi,

On Thu, 26 Mar 2009, Michael J Gruber wrote:

Christian Couder venit, vidit, dixit 26.03.2009 08:48:

quoted

Le jeudi 26 mars 2009, Sverre Rabbelier a écrit :

quoted

A 10 patches series with no cover letter?

I am not a big fan of cover letters. Usually I prefer adding comments 
in the patches.

I'm sorry I have to say that, but your individual preferences don't 
matter. Many of us would do things differently, each in their own way, 
but people adjust to the list's preferences. It's a matter of attitude. 
So, please...

Actually, a better way to ask for a cover letter would have been to 
convince Christian.  So I'll try that.

From the patch series' titles (especially when they are cropped due to the 

text window being too small to fit the indented thread), it is not all 
that obvious what you want to achieve with those 10 patches.

From recent discussions, I seem to remember that you wanted to have some 

cute way to mark commits as non-testable during a bisect, and I further 
seem to remember that Junio said that very method should be usable outside 
of bisect, too.

Unfortunately, that does not reveal to me, quickly, what is the current 
state of affairs, and what you changed since the last time.

In addition, I am very sorry that I cannot review your patches; day job is 
killing me right now.

Ciao,
Dscho

Heya,

On Thu, Mar 26, 2009 at 17:52, Johannes Schindelin
[off-list ref] > From the patch series' titles
(especially when they are cropped due to the

text window being too small to fit the indented thread), it is not all
that obvious what you want to achieve with those 10 patches.

<snip>

Unfortunately, that does not reveal to me, quickly, what is the current
state of affairs, and what you changed since the last time.

This is exactly what I meant to say, only worded much much better,
thanks Johannes! :)

-- 
Cheers,

Sverre Rabbelier

Le jeudi 26 mars 2009, Johannes Schindelin a écrit :

Hi,

On Thu, 26 Mar 2009, Michael J Gruber wrote:

quoted

Christian Couder venit, vidit, dixit 26.03.2009 08:48:

quoted

Le jeudi 26 mars 2009, Sverre Rabbelier a écrit :

quoted

A 10 patches series with no cover letter?

I am not a big fan of cover letters. Usually I prefer adding comments
in the patches.

I'm sorry I have to say that, but your individual preferences don't
matter. Many of us would do things differently, each in their own way,
but people adjust to the list's preferences. It's a matter of attitude.
So, please...

Actually, a better way to ask for a cover letter would have been to
convince Christian.  So I'll try that.

Thanks.

As you know, I have been sending patches since nearly 3 years ago to this 
list. And it's only since a few weeks ago that I am asked to send cover 
letters...

From the patch series' titles (especially when they are cropped due to
the text window being too small to fit the indented thread), it is not
all that obvious what you want to achieve with those 10 patches.

From recent discussions, I seem to remember that you wanted to have some
cute way to mark commits as non-testable during a bisect, and I further
seem to remember that Junio said that very method should be usable
outside of bisect, too.

Well, we want to move "git bisect skip" code from shell (in "git-bisect.sh") 
to C. So this patch series does that by creating a new "git bisect--helper" 
command in C that contains the new code and using that new command 
in "git-bisect.sh".

Unfortunately, that does not reveal to me, quickly, what is the current
state of affairs, and what you changed since the last time.

Yeah, I should have at least put something in the comment section of my 
first patch in this series.

And I try to improve, you know, I even tried to use "git send-email" again 
this morning to see if perhaps I could use it to send my patch series.

I did:

$ git send-email --compose --dry-run bh15/*
Can't call method "repo_path" on an undefined value 
at /home/christian/libexec/git-core//git-send-email line 160.

and then I gave up, because I don't like spending a lot of my free time to 
fight with tools I don't like.

If someone knows some other tools that can easily send a threaded patch 
series, I will try to see if I can use them...

Thanks in advance,
Christian.

On Fri, 27 Mar 2009, Christian Couder wrote:

If someone knows some other tools that can easily send a threaded patch
series, I will try to see if I can use them...

I long ago gave up on send-email, as it seemed to cumbersome for what I 
wanted, and my perl had got so rusty I really couldn't face trying to 
improve it.

So I wrote a replacement in Python (attached), which I have subsequently 
used for all patches I've sent.  It calls format-patch, passing through 
arguments (and you can use -- to let it pass options too).

(the only setting it reads from git config atm is mail-commit.to)

I find it much easier to use than send-email, but as usual YMMV ...

-- 
Julian

  ---
Have you seen the latest Japanese camera?  Apparently it is so fast it can
photograph an American with his mouth shut!

Hi,

On Fri, 27 Mar 2009, Christian Couder wrote:

Le jeudi 26 mars 2009, Johannes Schindelin a écrit :

quoted

On Thu, 26 Mar 2009, Michael J Gruber wrote:

quoted

Christian Couder venit, vidit, dixit 26.03.2009 08:48:

quoted

Le jeudi 26 mars 2009, Sverre Rabbelier a écrit :

quoted

A 10 patches series with no cover letter?

I am not a big fan of cover letters. Usually I prefer adding 
comments in the patches.

I'm sorry I have to say that, but your individual preferences don't 
matter. Many of us would do things differently, each in their own 
way, but people adjust to the list's preferences. It's a matter of 
attitude. So, please...

Actually, a better way to ask for a cover letter would have been to 
convince Christian.  So I'll try that.

Thanks.

As you know, I have been sending patches since nearly 3 years ago to 
this list. And it's only since a few weeks ago that I am asked to send 
cover letters...

Heh, I have the feeling that your patch series were much shorter, and did 
not have many revisions, until a few weeks ago ;-)

quoted

From the patch series' titles (especially when they are cropped due to 
the text window being too small to fit the indented thread), it is not 
all that obvious what you want to achieve with those 10 patches.

From recent discussions, I seem to remember that you wanted to have 
some cute way to mark commits as non-testable during a bisect, and I 
further seem to remember that Junio said that very method should be 
usable outside of bisect, too.

Well, we want to move "git bisect skip" code from shell (in 
"git-bisect.sh") to C. So this patch series does that by creating a new 
"git bisect--helper" command in C that contains the new code and using 
that new command in "git-bisect.sh".

Oh?  I _completely_ missed that.  And that's being one of the original 
Cc:ed persons...

quoted

Unfortunately, that does not reveal to me, quickly, what is the 
current state of affairs, and what you changed since the last time.

Yeah, I should have at least put something in the comment section of my 
first patch in this series.

No.  I would still have missed it.

The cover letter is outside of any patch, because it describes the purpose 
of the _whole_ patch series, not just one patch.

So, it would have been nice to get a heads-up that this is not your 
bisect-skip-a-whole-bunch-of-commits series, but a new animal.

This way, I decided I do not have time for something I do not need, and 
deleted it without having a look.

Ciao,
Dscho

Le vendredi 27 mars 2009, Johannes Schindelin a écrit :

Hi,

On Fri, 27 Mar 2009, Christian Couder wrote:

quoted

Le jeudi 26 mars 2009, Johannes Schindelin a écrit :

quoted

On Thu, 26 Mar 2009, Michael J Gruber wrote:

quoted

Christian Couder venit, vidit, dixit 26.03.2009 08:48:

quoted

Le jeudi 26 mars 2009, Sverre Rabbelier a écrit :

quoted

A 10 patches series with no cover letter?

I am not a big fan of cover letters. Usually I prefer adding
comments in the patches.

I'm sorry I have to say that, but your individual preferences don't
matter. Many of us would do things differently, each in their own
way, but people adjust to the list's preferences. It's a matter of
attitude. So, please...

Actually, a better way to ask for a cover letter would have been to
convince Christian.  So I'll try that.

Thanks.

As you know, I have been sending patches since nearly 3 years ago to
this list. And it's only since a few weeks ago that I am asked to send
cover letters...

Heh, I have the feeling that your patch series were much shorter, and did
not have many revisions, until a few weeks ago ;-)

Please try to look for a 9 patch long series that you reviewed around 
october 2007 with "dunno" or "skip" in the title ;-)

quoted

quoted

From the patch series' titles (especially when they are cropped due
to the text window being too small to fit the indented thread), it is
not all that obvious what you want to achieve with those 10 patches.

From recent discussions, I seem to remember that you wanted to have
some cute way to mark commits as non-testable during a bisect, and I
further seem to remember that Junio said that very method should be
usable outside of bisect, too.

Well, we want to move "git bisect skip" code from shell (in
"git-bisect.sh") to C. So this patch series does that by creating a new
"git bisect--helper" command in C that contains the new code and using
that new command in "git-bisect.sh".

Oh?  I _completely_ missed that.  And that's being one of the original
Cc:ed persons...

quoted

quoted

Unfortunately, that does not reveal to me, quickly, what is the
current state of affairs, and what you changed since the last time.

Yeah, I should have at least put something in the comment section of my
first patch in this series.

No.  I would still have missed it.

The cover letter is outside of any patch, because it describes the
purpose of the _whole_ patch series, not just one patch.

So, it would have been nice to get a heads-up that this is not your
bisect-skip-a-whole-bunch-of-commits series, but a new animal.

This way, I decided I do not have time for something I do not need, and
deleted it without having a look.

Well as I said in my previous email I am willing to improve. So perhaps next 
time.

Best regards,
Christian.

Le vendredi 27 mars 2009, Julian Phillips a écrit :

On Fri, 27 Mar 2009, Christian Couder wrote:

quoted

If someone knows some other tools that can easily send a threaded patch
series, I will try to see if I can use them...

I long ago gave up on send-email, as it seemed to cumbersome for what I
wanted, and my perl had got so rusty I really couldn't face trying to
improve it.

So I wrote a replacement in Python (attached), which I have subsequently
used for all patches I've sent.  It calls format-patch, passing through
arguments (and you can use -- to let it pass options too).

(the only setting it reads from git config atm is mail-commit.to)

I find it much easier to use than send-email, but as usual YMMV ...

Thanks I will try to have a look at it,
Christian.