From: Junio C Hamano <gitster@pobox.com>
To: Jonathan Nieder <jrnieder@gmail.com>
Cc: Johannes Sixt <j.sixt@viscovery.net>,
git@vger.kernel.org, Greg Brockman <gdb@MIT.EDU>,
Ilari Liusvaara <ilari.liusvaara@elisanet.fi>,
Elijah Newren <newren@gmail.com>
Subject: Re: jn/apply-filename-with-sp (Re: What's cooking in git.git (Aug 2010, #02; Wed, 11))
Date: Thu, 12 Aug 2010 16:17:51 -0700 [thread overview]
Message-ID: <7v1va3pgio.fsf@alter.siamese.dyndns.org> (raw)
In-Reply-To: <20100812224044.GK2029@burratino> (Jonathan Nieder's message of "Thu\, 12 Aug 2010 17\:40\:44 -0500")
Jonathan Nieder <jrnieder@gmail.com> writes:
> Johannes Sixt wrote:
>> Am 8/12/2010 1:35, schrieb Junio C Hamano:
>
>>> * jn/apply-filename-with-sp (2010-07-23) 4 commits
>>> - apply: Handle traditional patches with space in filename
>>> - t4135 (apply): use expand instead of pr for portability
>>> - tests: Test how well "git apply" copes with weird filenames
>>> - apply: Split quoted filename handling into new function
>>>
>>> Looked Ok. Will merge to 'next'.
>>
>> The test cases that work with files with tabs must have a prerequisite;
>> see t3600-rm.sh.
>
> Would this work?
>
> -- 8< --
> Subject: t4135 (apply): filenames with tabs are not usable on NTFS
>
> Move the code setting up the FUNNYNAMES prerequisite from
> v1.3.0-rc1~67 (workaround fat/ntfs deficiences for t3600-rm.sh,
> 2006-03-03) to test-lib and use it.
Hmm. FUNNYNAMES is used by only these two but we have nearly 500
individual test scripts in the test suite, and that was the original reason
why we don't have it in test-lib.sh.
I wonder if we can/should lazily check this kind of thing upon the first
use of symbol during a single test session?
* Create an empty $TEST_DIRECTORY/test-prereq file part of pre-clean
target of t/Makefile. This file needs to be removed as part of clean
target;
* t/test-lib.sh "."-sources the test-prereq file before doing anything
else;
* Instead of using "satisfied" variable as a SP separated list of
prerequisite tokens, use PREREQ_FOO shell variable (whose value can be
true, false) to represent FOO prerequisite we have already checked;
* When checking for prerequite FOO, test_have_prereq checks if PREREQ_FOO
variable is set. If not, then it calls a shell function of the same
name (i.e. PREREQ_FOO) defined in test-lib.sh (the function can exit
with zero or non-zero just like usual tests to signal if the
prerequisite is met), sets PREREQ_FOO to an appropriate value, and then
runs "echo PREREQ_FOO=$PREREQ_FOO >>$TEST_DIRECTORY/test-prereq".
Then $PREREQ_FOO can be examined to see if prerequisite FOO is
satisfied.
The file based caching of the above wouldn't work well under "make -j";
without it, we can still have a prerequisite checked only once per a test
script run, which is not worse than the current situation (and it would
even save cycles if a test does not check for a particular prereq,
e.g. SYMLINKS, that is always checked in test-lib).
next prev parent reply other threads:[~2010-08-12 23:18 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-11 23:35 What's cooking in git.git (Aug 2010, #02; Wed, 11) Junio C Hamano
2010-08-12 1:41 ` Jonathan Nieder
2010-08-12 2:33 ` Ævar Arnfjörð Bjarmason
2010-08-12 3:15 ` jn/commit-no-change-wo-status (Re: What's cooking in git.git (Aug 2010, #02; Wed, 11)) Jonathan Nieder
2010-08-12 5:47 ` What's cooking in git.git (Aug 2010, #02; Wed, 11) Elijah Newren
2010-08-12 15:49 ` Junio C Hamano
2010-08-12 21:12 ` Elijah Newren
2010-08-12 9:23 ` Johannes Sixt
2010-08-12 9:37 ` Greg Brockman
2010-08-12 10:11 ` Ævar Arnfjörð Bjarmason
2010-08-12 22:08 ` Junio C Hamano
2010-08-12 22:13 ` Greg Brockman
2010-08-12 22:19 ` Junio C Hamano
2010-08-12 10:20 ` Ævar Arnfjörð Bjarmason
2010-08-12 11:35 ` Erik Faye-Lund
2010-08-12 16:50 ` Ævar Arnfjörð Bjarmason
2010-08-12 17:34 ` Chris Packham
2010-08-12 18:35 ` Ævar Arnfjörð Bjarmason
2010-08-12 22:19 ` windows smoke tester (was Re: What's cooking in git.git (Aug 2010, #02; Wed, 11)) Chris Packham
2010-08-12 22:29 ` Ævar Arnfjörð Bjarmason
2010-08-12 22:58 ` Chris Packham
2010-08-13 1:01 ` Ævar Arnfjörð Bjarmason
2010-08-14 0:42 ` Chris Packham
2010-08-14 0:46 ` Ævar Arnfjörð Bjarmason
2010-08-15 0:54 ` Tay Ray Chuan
2010-08-15 1:08 ` Ævar Arnfjörð Bjarmason
2010-08-15 17:39 ` Tay Ray Chuan
2010-08-12 10:21 ` What's cooking in git.git (Aug 2010, #02; Wed, 11) Ilari Liusvaara
2010-08-12 10:31 ` Johannes Sixt
2010-08-12 15:25 ` Ilari Liusvaara
2010-08-12 12:43 ` Elijah Newren
2010-08-12 22:21 ` Junio C Hamano
2010-08-12 21:58 ` Junio C Hamano
2010-08-12 22:40 ` jn/apply-filename-with-sp (Re: What's cooking in git.git (Aug 2010, #02; Wed, 11)) Jonathan Nieder
2010-08-12 22:46 ` Ævar Arnfjörð Bjarmason
2010-08-12 23:17 ` Junio C Hamano [this message]
2010-08-13 0:59 ` Ævar Arnfjörð Bjarmason
2010-08-13 21:44 ` Johannes Sixt
2010-08-14 2:27 ` Jonathan Nieder
2010-08-14 18:37 ` Johannes Sixt
2010-08-15 0:05 ` Jonathan Nieder
2010-08-19 1:45 ` [PATCH v2 0/3] apply: handle traditional patches with space in filename Jonathan Nieder
2010-08-19 1:46 ` [PATCH 1/3] apply: split quoted filename handling into new function Jonathan Nieder
2010-08-19 1:48 ` [PATCH 2/3] tests: exercise "git apply" with weird filenames Jonathan Nieder
2010-08-19 1:50 ` [PATCH 3/3] apply: handle traditional patches with space in filename Jonathan Nieder
2010-08-19 19:56 ` [PATCH v2 0/3] " Johannes Sixt
2010-08-20 6:26 ` Jonathan Nieder
2010-08-13 0:08 ` jn/svn-fe Jonathan Nieder
2010-08-13 10:18 ` jn/svn-fe Jakub Narebski
2010-08-13 21:33 ` jn/svn-fe Johannes Sixt
2010-08-13 23:47 ` [PATCH v2 jn/svn-fe 0/5] vcs-svn: Port to Windows Jonathan Nieder
2010-08-13 23:59 ` [PATCH 1/5] compat: add strtok_r() Jonathan Nieder
2010-08-14 0:01 ` [PATCH 2/5] vcs-svn: Rename dirent pool to build on Windows Jonathan Nieder
2010-08-14 0:03 ` [PATCH 3/5] vcs-svn: Avoid %z in format string Jonathan Nieder
2010-08-14 0:04 ` [PATCH 4/5] t9010 (svn-fe): use Unix-style path in URI Jonathan Nieder
2010-08-14 0:06 ` [PATCH 5/5] t9010 (svn-fe): avoid symlinks in test Jonathan Nieder
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=7v1va3pgio.fsf@alter.siamese.dyndns.org \
--to=gitster@pobox.com \
--cc=gdb@MIT.EDU \
--cc=git@vger.kernel.org \
--cc=ilari.liusvaara@elisanet.fi \
--cc=j.sixt@viscovery.net \
--cc=jrnieder@gmail.com \
--cc=newren@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).