Re: [RFC PATCH 0/4] Fix occasional test failures in http tests

2018-06-14 Thread Jeff King
On Thu, Jun 14, 2018 at 02:31:03PM +0200, SZEDER Gábor wrote:

> 't5561-http-backend.sh' is prone to occasional failures; luckily it's
> not 'git-http-backend's fault, but the test script is a bit racy.  I
> won't go into the details here, patch 4/4's commit message discusses
> it at length.  4/4 also fixes this issue, but I'm not particularly
> happy with that fix, the note attached to that patch explains why,
> along with possible alternatives, hence the RFC.
> 
> Even if we settle for a different fix, I think the first two patches
> are worthy cleanups on their own right.

Yes, the first two look like improvements on their own (and I left more
detailed comments on the final one). Thanks for tackling this!

-Peff


Re: [RFC PATCH 0/4] Fix occasional test failures in http tests

2018-06-14 Thread Junio C Hamano
SZEDER Gábor  writes:

> 't5561-http-backend.sh' is prone to occasional failures; luckily it's
> not 'git-http-backend's fault, but the test script is a bit racy.  I
> won't go into the details here, patch 4/4's commit message discusses
> it at length.  4/4 also fixes this issue, but I'm not particularly
> happy with that fix, the note attached to that patch explains why,
> along with possible alternatives, hence the RFC.

Interesting.

> Even if we settle for a different fix, I think the first two patches
> are worthy cleanups on their own right.

Yeah, the first two are unquestionally good.  

The sed expression of the second one could use sq around (instead of
dq) to lose hard-to-read backslash quoting from there, but it
probably is OK to keep the second one a pure "refactoring" patch,
and leave it to a separate "further cleanup" patch to make such a
change.

> SZEDER Gábor (4):
>   t5541: avoid empty line when truncating access log
>   t/lib-httpd: add the strip_access_log() helper function
>   t/lib-httpd: add minor and patch version to $HTTPD_VERSION
>   t/lib-httpd: sort log based on timestamp to avoid occasional failure
>
>  t/lib-httpd.sh  | 24 ++--
>  t/lib-httpd/apache.conf |  5 +
>  t/t5541-http-push-smart.sh  | 17 +++--
>  t/t5551-http-fetch-smart.sh |  7 +--
>  t/t5561-http-backend.sh |  7 +--
>  5 files changed, 32 insertions(+), 28 deletions(-)


[RFC PATCH 0/4] Fix occasional test failures in http tests

2018-06-14 Thread SZEDER Gábor
't5561-http-backend.sh' is prone to occasional failures; luckily it's
not 'git-http-backend's fault, but the test script is a bit racy.  I
won't go into the details here, patch 4/4's commit message discusses
it at length.  4/4 also fixes this issue, but I'm not particularly
happy with that fix, the note attached to that patch explains why,
along with possible alternatives, hence the RFC.

Even if we settle for a different fix, I think the first two patches
are worthy cleanups on their own right.

SZEDER Gábor (4):
  t5541: avoid empty line when truncating access log
  t/lib-httpd: add the strip_access_log() helper function
  t/lib-httpd: add minor and patch version to $HTTPD_VERSION
  t/lib-httpd: sort log based on timestamp to avoid occasional failure

 t/lib-httpd.sh  | 24 ++--
 t/lib-httpd/apache.conf |  5 +
 t/t5541-http-push-smart.sh  | 17 +++--
 t/t5551-http-fetch-smart.sh |  7 +--
 t/t5561-http-backend.sh |  7 +--
 5 files changed, 32 insertions(+), 28 deletions(-)

-- 
2.18.0.rc0.207.ga6211da864