git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Duy Nguyen <pclouds@gmail.com>
To: Junio C Hamano <gitster@pobox.com>
Cc: Git Mailing List <git@vger.kernel.org>
Subject: Re: [PATCH] Move test-* to t/helper/ subdirectory
Date: Sun, 1 May 2016 13:00:09 +0700	[thread overview]
Message-ID: <20160501060009.GA30748@lanh> (raw)
In-Reply-To: <20160501002852.GA3963@lanh>

On Sun, May 01, 2016 at 07:28:52AM +0700, Duy Nguyen wrote:
> On Wed, Apr 27, 2016 at 09:15:41AM -0700, Junio C Hamano wrote:
> > Duy Nguyen <pclouds@gmail.com> writes:
> > 
> > > This patch forces bin-wrappers regeneration every time a test program
> > > is updated. A bit wasteful, but I don't see a better option (which is
> > > also why I limit this to test programs only).
> > 
> > In other words, when we update the location where the programs that
> > would be eventually installed are created, we'd see the same
> > problem.
> > 
> > I actually wonder if it is a better overall structure to move
> > t/helper/test-foo back to test-foo, while keeping the source file
> > that contains main() for test-foo at t/helper/test-foo.c.  Then we
> > do not have to have many copies that are slightly different in
> > bin-wrappers, but they can all be
> > 
> > 	exec "${GIT_EXEC_PATH}/$0" "$@"
> > 
> > instead of "bin-wrappers/git-bar" being
> > 
> > 	exec "${GIT_EXEC_PATH}/git-bar" "$@"
> > 
> > and "bin-wrappers/test-foo" being
> > 
> > 	exec "${GIT_EXEC_PATH}/t/helper/test-foo" "$@"
> > 
> 
> It's not a perfect solution (rebuild bin-wrappers when the real binary
> moves) but I think it's the best option so far.

I may have rushed to judgement. wrap-for-bin.sh has always been the
dependency for bin-wrappers/*. If we force that file to change, then
bin-wrappers/* will be recreated when switching branches. So how about
this?

-- 8< --
Subject: [PATCH] wrap-for-bin.sh: handle t/helper/ paths internally

Commit e6e7530 (test helpers: move test-* to t/helper/ subdirectory -
2016-04-13) moves test-* to t/helper. However because bin-wrappers/*
only depend on wrap-for-bin.sh, when switching between a branch that has
this commit and one that does not, bin-wrappers/* may not be regenerated
and point to the old/outdated test programs. Fix it by force updating
wrap-for-bin.sh so that it will be regenerated anyway.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
 Makefile        |  2 +-
 wrap-for-bin.sh | 12 ++++++++++--
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/Makefile b/Makefile
index dd178ee..38dcdf7 100644
--- a/Makefile
+++ b/Makefile
@@ -2204,7 +2204,7 @@ bin-wrappers/%: wrap-for-bin.sh
 	@mkdir -p bin-wrappers
 	$(QUIET_GEN)sed -e '1s|#!.*/sh|#!$(SHELL_PATH_SQ)|' \
 	     -e 's|@@BUILD_DIR@@|$(shell pwd)|' \
-	     -e 's|@@PROG@@|$(patsubst test-%,t/helper/test-%,$(@F))|' < $< > $@ && \
+	     -e 's|@@PROG@@|$(@F)|' < $< > $@ && \
 	chmod +x $@
 
 # GNU make supports exporting all variables by "export" without parameters.
diff --git a/wrap-for-bin.sh b/wrap-for-bin.sh
index db0ec6a..bb334ed 100644
--- a/wrap-for-bin.sh
+++ b/wrap-for-bin.sh
@@ -18,11 +18,19 @@ GITPERLLIB='@@BUILD_DIR@@/perl/blib/lib'"${GITPERLLIB:+:$GITPERLLIB}"
 GIT_TEXTDOMAINDIR='@@BUILD_DIR@@/po/build/locale'
 PATH='@@BUILD_DIR@@/bin-wrappers:'"$PATH"
 export GIT_EXEC_PATH GITPERLLIB PATH GIT_TEXTDOMAINDIR
+case "@@PATH@@" in
+    test-*)
+	PROG="${GIT_EXEC_PATH}/t/helper/@@PROG@@"
+	;;
+    *)
+	PROG="${GIT_EXEC_PATH}/@@PROG@@"
+	;;
+esac
 
 if test -n "$GIT_TEST_GDB"
 then
 	unset GIT_TEST_GDB
-	exec gdb --args "${GIT_EXEC_PATH}/@@PROG@@" "$@"
+	exec gdb --args "$PROG" "$@"
 else
-	exec "${GIT_EXEC_PATH}/@@PROG@@" "$@"
+	exec "$PROG" "$@"
 fi
-- 
2.8.0.rc0.210.gd302cd2
-- 8< --

  reply	other threads:[~2016-05-01  6:00 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-13 13:22 [PATCH] Move test-* to t/helper/ subdirectory Nguyễn Thái Ngọc Duy
2016-04-15 15:09 ` Junio C Hamano
2016-04-15 17:06   ` Junio C Hamano
2016-04-16  0:08     ` Duy Nguyen
2016-04-26 22:07 ` Junio C Hamano
2016-04-27  0:52   ` Duy Nguyen
2016-04-27 10:18     ` Duy Nguyen
2016-04-27 16:15       ` Junio C Hamano
2016-05-01  0:28         ` Duy Nguyen
2016-05-01  6:00           ` Duy Nguyen [this message]
2016-05-02 17:34             ` Junio C Hamano
2016-05-03  0:15               ` Duy Nguyen
2016-05-08  9:51                 ` Duy Nguyen
2016-05-09 16:03                   ` Junio C Hamano
2016-05-10 11:58                     ` Duy Nguyen
2016-05-10 20:23                       ` Junio C Hamano
  -- strict thread matches above, loose matches on Subject: below --
2011-02-09 22:15 [PATCH/RFC] Move test-*.c to test/ subdirectory Junio C Hamano
2011-02-10  2:14 ` [PATCH] Move test-* to t/helper/ subdirectory Nguyễn Thái Ngọc Duy

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=20160501060009.GA30748@lanh \
    --to=pclouds@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.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).