Add a .gitattributes file to force LF line-endings in net/http/*.pins files.

Add a .gitattributes file to force LF line-endings in .pins files.

This change forces the pins files to use LF line-endings and adds a check to
detect CRLF line-endings that outputs a more helpful error message.

BUG=729553
Review-Url: https://codereview.chromium.org/2941543002
Cr-Commit-Position: refs/heads/master@{#485165}
Committed: https://chromium.googlesource.com/chromium/src/+/89d553d1c6f6d386c52fc3ade38eb4fe2508ab7f

[martijnc, 2017-06-13 16:19:00]

The CQ bit was checked by martijn@martijnc.be to run a CQ dry run

[commit-bot: I haz the power, 2017-06-13 16:19:27]

Dry run: CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-06-13 16:31:47]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-06-13 16:31:48]

Dry run: This issue passed the CQ dry run.

[martijnc, 2017-06-13 16:45:55]

Description was changed from

==========
Add .gitattributes file to force LF line-endings in .pins files.

BUG=729553,732746
==========

to

==========
Add a .gitattributes file to force LF line-endings in .pins files.

This change forces the pins files to use LF line-endings and adds a check to
detect CRLF line-endings that outputs a more helpful error message.

BUG=729553,732746
==========

[martijnc, 2017-06-13 16:45:55]

martijn@martijnc.be changed reviewers:
+ bratell@opera.com, rsleevi@chromium.org

[martijnc, 2017-06-13 16:46:12]

The CQ bit was checked by martijn@martijnc.be to run a CQ dry run

[commit-bot: I haz the power, 2017-06-13 16:46:33]

Dry run: CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[martijnc, 2017-06-13 16:57:33]

Description was changed from

==========
Add a .gitattributes file to force LF line-endings in .pins files.

This change forces the pins files to use LF line-endings and adds a check to
detect CRLF line-endings that outputs a more helpful error message.

BUG=729553,732746
==========

to

==========
Add a .gitattributes file to force LF line-endings in .pins files.

This change forces the pins files to use LF line-endings and adds a check to
detect CRLF line-endings that outputs a more helpful error message.

BUG=729553
==========

[commit-bot: I haz the power, 2017-06-13 17:01:00]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-06-13 17:01:01]

Dry run: Try jobs failed on following builders:
  cast_shell_linux on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_linu...)
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_chromeos_ozone_rel_ng on master.tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_compile_dbg_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_headless_rel on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_tsan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[martijnc, 2017-06-13 18:58:51]

rsleevi: can you review the net changes?
bratell: FYI

The .gitattributes change should ensure the .pins files in net/http/ are never
converted to CRLF line-endings. I used the same syntax as the .gitattributes
file in src/ for .py files, according to the git documentation this should have
the desired effect [1].

This CL also adds a check for CRLF line-endings and fails the generator with a
more informative error message when they are found.

[1] https://git-scm.com/docs/gitattributes#gitattributes-Settostringvaluelf

[Ryan Sleevi, 2017-06-14 10:13:35]

rsleevi@chromium.org changed reviewers:
+ maruel@chromium.org

[Ryan Sleevi, 2017-06-14 10:13:35]

I don't feel qualified to opine on whether this is right thing to do, but my
feeling is that it is not. I'm going to tag in Marc-Antoine here, who would know
more about the infrastructure and what 'guarantees' (or not) are made.

My understanding is that we only officially support heremetic depot-tools git.

[Daniel Bratell, 2017-06-14 11:27:05]

On 2017/06/14 10:13:35, Ryan Sleevi wrote:
> I don't feel qualified to opine on whether this is right thing to do, but my
> feeling is that it is not. I'm going to tag in Marc-Antoine here, who would
know
> more about the infrastructure and what 'guarantees' (or not) are made.
> 
> My understanding is that we only officially support heremetic depot-tools git.

I've heard that statement before but I think it's a bit of wishful thinking. We
(Opera) ran for a long time with other gits because the bundled git was too old
and slow just as an example, and it's hard to guarantee that paths stay
consistent at a machine which may have multiple git installations.

So if it's trivial to make the code robust, then it would be sad to not do that.
IMHO.

[M-A Ruel, 2017-06-14 17:49:34]

On 2017/06/14 11:27:05, Daniel Bratell wrote:
> On 2017/06/14 10:13:35, Ryan Sleevi wrote:
> > I don't feel qualified to opine on whether this is right thing to do, but my
> > feeling is that it is not. I'm going to tag in Marc-Antoine here, who would
> know
> > more about the infrastructure and what 'guarantees' (or not) are made.
> > 
> > My understanding is that we only officially support heremetic depot-tools
git.
> 
> I've heard that statement before but I think it's a bit of wishful thinking.
We
> (Opera) ran for a long time with other gits because the bundled git was too
old
> and slow just as an example, and it's hard to guarantee that paths stay
> consistent at a machine which may have multiple git installations.
> 
> So if it's trivial to make the code robust, then it would be sad to not do
that.
> IMHO.

https://chromium.googlesource.com/chromium/src/+/master/.gitattributes would be
the right place.

[M-A Ruel, 2017-06-14 17:49:34]

On 2017/06/14 11:27:05, Daniel Bratell wrote:
> On 2017/06/14 10:13:35, Ryan Sleevi wrote:
> > I don't feel qualified to opine on whether this is right thing to do, but my
> > feeling is that it is not. I'm going to tag in Marc-Antoine here, who would
> know
> > more about the infrastructure and what 'guarantees' (or not) are made.
> > 
> > My understanding is that we only officially support heremetic depot-tools
git.
> 
> I've heard that statement before but I think it's a bit of wishful thinking.
We
> (Opera) ran for a long time with other gits because the bundled git was too
old
> and slow just as an example, and it's hard to guarantee that paths stay
> consistent at a machine which may have multiple git installations.
> 
> So if it's trivial to make the code robust, then it would be sad to not do
that.
> IMHO.

https://chromium.googlesource.com/chromium/src/+/master/.gitattributes would be
the right place.

[martijnc, 2017-06-14 18:50:16]

Patchset #3 (id:40001) has been deleted

[martijnc, 2017-06-14 18:53:22]

Patchset #3 (id:60001) has been deleted

[martijnc, 2017-06-14 18:55:04]

On 2017/06/14 at 17:49:34, maruel wrote:
> https://chromium.googlesource.com/chromium/src/+/master/.gitattributes would
be the right place.

Moved the rule to src/.gitattributes in PS#3.

[Daniel Bratell, 2017-06-28 16:06:48]

On 2017/06/14 18:55:04, martijnc wrote:
> On 2017/06/14 at 17:49:34, maruel wrote:
> > https://chromium.googlesource.com/chromium/src/+/master/.gitattributes would
> be the right place.
> 
> Moved the rule to src/.gitattributes in PS#3.

Looks good to me (non-owner). Need rsleevi and/or maruel to appprove?

[Ryan Sleevi, 2017-06-28 16:09:02]

On 2017/06/28 16:06:48, Daniel Bratell wrote:
> Looks good to me (non-owner). Need rsleevi and/or maruel to appprove?

Entirely deferring to Marc-Antoine here, as the impact/expectations of
infrastructure are his ken.

[martijnc, 2017-06-30 16:26:03]

maruel: can you take a look or comment on whether this change is acceptable?
Thanks!

[M-A Ruel, 2017-07-09 01:16:34]

lgtm

[Ryan Sleevi, 2017-07-09 20:30:43]

lgtm

[martijnc, 2017-07-09 20:37:27]

The CQ bit was checked by martijn@martijnc.be

[commit-bot: I haz the power, 2017-07-09 20:37:40]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-07-09 21:44:34]

CQ is committing da patch.

Bot data: {"patchset_id": 80001, "attempt_start_ts": 1499632647942940,
"parent_rev": "c8ddaa9787cdff766f819c402d2dcd6de796fa90", "commit_rev":
"89d553d1c6f6d386c52fc3ade38eb4fe2508ab7f"}

[commit-bot: I haz the power, 2017-07-09 21:44:47]

Description was changed from

==========
Add a .gitattributes file to force LF line-endings in .pins files.

This change forces the pins files to use LF line-endings and adds a check to
detect CRLF line-endings that outputs a more helpful error message.

BUG=729553
==========

to

==========
Add a .gitattributes file to force LF line-endings in .pins files.

This change forces the pins files to use LF line-endings and adds a check to
detect CRLF line-endings that outputs a more helpful error message.

BUG=729553

Review-Url: https://codereview.chromium.org/2941543002
Cr-Commit-Position: refs/heads/master@{#485165}
Committed:
https://chromium.googlesource.com/chromium/src/+/89d553d1c6f6d386c52fc3ade38e...
==========

[commit-bot: I haz the power, 2017-07-09 21:44:48]

Committed patchset #3 (id:80001) as
https://chromium.googlesource.com/chromium/src/+/89d553d1c6f6d386c52fc3ade38e...