Re: cb/pedantic-build-for-developers, POSIX-but-not-C99 and -Wno-pedantic-ms-format

["René Scharfe" <l.s.r@web.de>]

Am 21.09.21 um 12:10 schrieb Carlo Arenas:
>>     -       strbuf_addf(&fmt, "%%s%%s%%s-%%0%1$lud.%%0%1$lud-%%s-%%s-%%s",
>>     -                   (unsigned long)tmp.len);
>>     +       strbuf_addf(&fmt, "%%s%%s%%s-%%0%lud.%%0%lud-%%s-%%s-%%s",
>>     +                   (unsigned long)tmp.len, (unsigned long)tmp.len);
>
> The C standard version seems simpler to me, so I would say this was a
> good outcome overall.

I get hypnotized by all those percent signs. %-O

>> I may be wrong, but I believe that construct is widely portable, we
>> don't use it in the main source, but in the po/ files (so anything that
>> uses git + gettext tests for this already):
>>
>>     git grep -P '%\d+\$' -- po

I wasn't aware of that extension.  Seems useful for translations, when
reordering elements is required or convenient to match the sentence
structure of a different language.

The lack of warnings about the use of that extension in po/ means
translated formats are not checked by the compiler at all, right?
According to the gettext documentation [2] msgfmt takes care of that
when building the translation files, though.

>
> It is not part of ANY C standard, so it will just print gibberish in
> systems that don't carry this extension.
>
> FWIW is not supported in Windows, which probably explains why they
> don't carry the po files.

Right, Microsoft doesn't mention it in its printf format documentation
[1].  The gettext documentation [3] claims that such platforms get
replacement functions automatically via libintl.h.

René


[1] https://docs.microsoft.com/en-us/cpp/c-runtime-library/format-specification-syntax-printf-and-wprintf-functions?view=msvc-160
[2] https://www.gnu.org/software/gettext/manual/gettext.html#c_002dformat-Flag
[3| https://www.gnu.org/software/gettext/manual/gettext.html#c_002dformat