Re: [PATCH] doc: set actual revdate for manpages

From: Junio C Hamano <gitster@pobox.com>
Date: Apr 14, 2023, 15:27 UTC
Message-ID: <xmqq5y9yv601.fsf@gitster.g>
In-Reply-To: <20230414070449.GA540206@coredump.intra.peff.net>

Jeff King <peff@peff.net> writes:

Show 5 quoted lines

> ...  The reason is that asciidoc puts the <date> tags in the
> header, and the custom header removed by 8806120de6c suppresses
> asciidoc's default header entirely (so a workaround would be to include
> the <date> tags in our custom header, but I don't see any reason not to
> just build on top of 8806120de6c, as you did here).

I agree building on what is in-flight instead of making duplicate effort makes sense.

Show 20 quoted lines

>> diff --git a/Documentation/Makefile b/Documentation/Makefile
>> index 3133ea3182..b629176d7d 100644
>> --- a/Documentation/Makefile
>> +++ b/Documentation/Makefile
>> @@ -144,13 +144,16 @@ man5dir = $(mandir)/man5
>>  man7dir = $(mandir)/man7
>>  # DESTDIR =
>>  
>> +GIT_DATE := $(shell git show --quiet --pretty='%as')
>
> What will/should this do in a distribution tarball, where we won't have
> a Git repository at all? I think we'll just end up with a blank date in
> the xml file, though it looks like docbook turns that into today's date
> in the result.
>
> That's not _too_ bad, but feels a bit inconsistent (and it uses the
> format you're trying to get rid of!).
>
> It would be nicer to populate the date variable in that case, like we do
> for GIT_VERSION. I think that could look something like this:

I did wonder the same "what about tarball" thing while reading this last night and then stopped, and it is good to see somebody else brought it up ;-). I like the "version" approach, as that is already an established way to deal with the same "the builder somehow wants to force the identifying string to a certain value in the build product".

Show 24 quoted lines

> diff --git a/Documentation/Makefile b/Documentation/Makefile
> index 2ccc3a9bc9..307634a94f 100644
> --- a/Documentation/Makefile
> +++ b/Documentation/Makefile
> @@ -144,8 +144,6 @@ man5dir = $(mandir)/man5
>  man7dir = $(mandir)/man7
>  # DESTDIR =
>  
> -GIT_DATE := $(shell git show --quiet --pretty='%as')
> -
>  ASCIIDOC = asciidoc
>  ASCIIDOC_EXTRA =
>  ASCIIDOC_HTML = xhtml11
> diff --git a/GIT-VERSION-GEN b/GIT-VERSION-GEN
> index 9a1111af9b..14903bd261 100755
> --- a/GIT-VERSION-GEN
> +++ b/GIT-VERSION-GEN
> @@ -10,7 +10,8 @@ LF='
>  # then try git-describe, then default.
>  if test -f version
>  then
> -	VN=$(cat version) || VN="$DEF_VER"
> +	VN=$(cut -d" " -f1 version) || VN="$DEF_VER"
> +	DN=$(cut -d" " -f2 version) || DN=""

Are we sure historical GIT_VERSION strings never had a SP in it? I would be very surprised if we did, but the correctness of the approach depends on that assumption.

Thanks.