git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: git@vger.kernel.org
Cc: Seyi Kuforiji <kuforiji98@gmail.com>,
	 ps@pks.im, phillip.wood@dunelm.org.uk
Subject: Re: [PATCH v3 06/10] t/unit-tests: convert reftable table test to use clar
Date: Tue, 03 Jun 2025 12:12:35 -0700	[thread overview]
Message-ID: <xmqqbjr4iqqk.fsf@gitster.g> (raw)
In-Reply-To: <xmqq7c1uszu6.fsf@gitster.g> (Junio C. Hamano's message of "Mon, 02 Jun 2025 12:33:37 -0700")

Junio C Hamano <gitster@pobox.com> writes:

>> diff --git a/Makefile b/Makefile
>> index 3cccc73073..0227fdb3e1 100644
>> --- a/Makefile
>> +++ b/Makefile
>> ...
>> @@ -3972,4 +3972,4 @@ $(LIBGIT_HIDDEN_EXPORT): $(LIBGIT_PARTIAL_EXPORT)
>>  	$(OBJCOPY) --localize-hidden $^ $@
>>  
>>  contrib/libgit-sys/libgitpub.a: $(LIBGIT_HIDDEN_EXPORT)
>> -	$(AR) $(ARFLAGS) $@ $^
>> +	$(AR) $(ARFLAGS) $@ $^
>> \ No newline at end of file
>
> Can anybody tell, without running "make <something>", what this
> accidental and unintended change breaks?  It may be rather
> surprising ;-)

After dropping the last newline from the top-level Makefile, 

    $ make check-docs

ends like this:

    $ make check-docs
    make -C Documentation lint-docs
    make[1]: Entering directory '/home/gitster/git.git/Documentation'
        GEN lint-docs-manpages
    removed but documented: git
    removed but documented: git-add
    ...
    removed but documented: git-write-tree
    removed but documented: scalar
    removed but listed: git-add
    removed but listed: git-am
    ...
    removed but listed: git-write-tree
    removed but listed: scalar
    make[1]: *** [Makefile:504: lint-docs-manpages] Error 1
    make[1]: Leaving directory '/home/gitster/git.git/Documentation'
    make: *** [Makefile:3829: check-docs] Error 2

This is because of a cute Makefile trick used in lint-manpages
script is not written robustly enough, but that is not much less
interesting than the lesson we learn from this episode.


A more robust and traditional way to use the same Makefile trick is
to apply this patch to Documentation/lint-manpages.sh by the way.

 Documentation/lint-manpages.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git c/Documentation/lint-manpages.sh w/Documentation/lint-manpages.sh
index a0ea572382..65758a3159 100755
--- c/Documentation/lint-manpages.sh
+++ w/Documentation/lint-manpages.sh
@@ -2,11 +2,11 @@
 
 extract_variable () {
 	(
-		cat ../Makefile
 		cat <<EOF
 print_variable:
 	@\$(foreach b,\$($1),echo XXX \$(b:\$X=) YYY;)
 EOF
+		cat ../Makefile
 	) |
 	make -C .. -f - print_variable 2>/dev/null |
 	sed -n -e 's/.*XXX \(.*\) YYY.*/\1/p'

  parent reply	other threads:[~2025-06-03 19:12 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-02 12:25 [PATCH v3 00/10] t/unit-tests: convert unit-tests to use clar Seyi Kuforiji
2025-06-02 12:25 ` [PATCH v3 01/10] t/unit-tests: implement clar specific reftable test helper functions Seyi Kuforiji
2025-06-02 13:23   ` Kristoffer Haugsbakk
2025-06-02 12:25 ` [PATCH v3 02/10] t/unit-tests: convert reftable basics test to use clar test framework Seyi Kuforiji
2025-06-03 13:51   ` Patrick Steinhardt
2025-06-03 14:03     ` Seyi Kuforiji
2025-06-03 14:14       ` Patrick Steinhardt
2025-06-02 12:25 ` [PATCH v3 03/10] t/unit-tests: convert reftable block test to use clar Seyi Kuforiji
2025-06-03 14:18   ` Patrick Steinhardt
2025-06-02 12:25 ` [PATCH v3 04/10] t/unit-tests: convert reftable merged " Seyi Kuforiji
2025-06-02 12:25 ` [PATCH v3 05/10] t/unit-tests: convert reftable pq " Seyi Kuforiji
2025-06-02 12:25 ` [PATCH v3 06/10] t/unit-tests: convert reftable table " Seyi Kuforiji
2025-06-02 19:33   ` Junio C Hamano
2025-06-03 11:46     ` Seyi Kuforiji
2025-06-03 19:12     ` Junio C Hamano [this message]
2025-06-02 12:25 ` [PATCH v3 07/10] t/unit-tests: convert reftable readwrite " Seyi Kuforiji
2025-06-02 12:25 ` [PATCH v3 08/10] t/unit-tests: convert reftable record " Seyi Kuforiji
2025-06-02 12:25 ` [PATCH v3 09/10] t/unit-tests: convert reftable stack " Seyi Kuforiji
2025-06-02 12:25 ` [PATCH v3 10/10] t/unit-tests: finalize migration of reftable-related tests Seyi Kuforiji
2025-06-03 13:51   ` Patrick Steinhardt
2025-06-05 12:56     ` Seyi Kuforiji
2025-06-05 13:24       ` Patrick Steinhardt
2025-06-05 16:56         ` Junio C Hamano
2025-06-03 18:41 ` [PATCH v3 00/10] t/unit-tests: convert unit-tests to use clar Junio C Hamano
2025-06-04 10:09   ` Seyi Kuforiji
2025-06-04 10:09   ` Seyi Kuforiji

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=xmqqbjr4iqqk.fsf@gitster.g \
    --to=gitster@pobox.com \
    --cc=git@vger.kernel.org \
    --cc=kuforiji98@gmail.com \
    --cc=phillip.wood@dunelm.org.uk \
    --cc=ps@pks.im \
    /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).