On 2023-06-02 15:03, Paul Eggert wrote:
How about changing this to use __builtin_constant_p instead of
__OPTIMIZE__? I.e., use the comma expression if __builtin_constant_p
(status), and use the statement expression otherwise.
I gave that a shot by installing the attached.From cfd41b3ac199a4f95
On 6/2/23 11:10, Bruno Haible wrote:
+/* Avoid evaluating STATUS twice, if this is possible without making the
+ "warning: this statement may fall through" reappear. */
+# if __OPTIMIZE__
+# define __gl_error_call(function, status, ...) \
That means semantics are different when optimizing;
On 6/2/23 11:47, Bruno Haible wrote:
I'm therefore considering (c). Although it will mean that xasprintf
will depend on vasnprintf, even on glibc systems.
Sounds like the best way to go, unfortunately.
propername_lite acts like propername_utf8 but needs less
infrastructure, e.g., it does not worry about memory allocation.
* MODULES.html.sh (func_all_modules): Mention it.
* lib/propername.h (proper_name_lite): New decl.
* lib/propername-lite.c, modules/propername-lite: New files.
* pygnulib/GLEmit
In GNU gettext and other packages, some source files are being compiled twice.
It's not useful to see the warnings twice, only once. Therefore a way to
inhibit all warnings in a particular compiler invocation is useful.
This patch adds this ability to the 'warnings' module. (Creating a separate
mo
The gcc -Wmissing-prototype / -Wmissing-declaration option is not something
I would universally turn on, because sometimes functions are not defined for
explicit invocation, but only for backward compatibility.
But sometimes it can highlight maintainability problems. In particular, this
warning se
The gcc -Wshadow warning option highlights possible maintainability problems.
This one seemed worth fixing:
libtextstyle/lib/term-style-control.c:740:17: warning: declaration of ‘sig’
shadows a parameter
2023-06-02 Bruno Haible
term-style-control: Silence -Wshadow warning.
*
The gcc -Wunused-macros warning option triggers many warnings, too many to
regularly look at. But occasionally, it highlights code that can be cleaned up.
Such as this one:
gettext-tools/gnulib-lib/pipe-filter-aux.c:40: warning: macro "_" is not used
2023-06-02 Bruno Haible
pipe-filt
Another warning from gcc13's -Wanalyzer-possible-null-argument, seen in
GNU gettext, is:
gettext-tools/gnulib-lib/javacomp.c:296:20: warning: use of possibly-NULL
'javac' where non-null expected [CWE-690]
The problem here is that we use xasprintf with a format string such as
"%s -source %s"
to
The gcc 13 -Wanalyzer-possible-null-argument warning option reported a couple
of bugs in GNU gettext. No false positives on this one. I can therefore
highly recommend it.
One of the warnings is:
gettext-tools/gnulib-lib/striconveha.c:339:7: warning: use of possibly-NULL
'to_codeset_suffixed' whe
Paul Eggert wrote on 2023-05-30:
> > +# define error(status, ...) \
> > + ((error)(0, __VA_ARGS__), (status) ? exit (status) : (void)0)
> ...
> Also, it's better to not evaluate 'status' twice. Not that I think
> 'status' should have side effects or even that it does have side effects
> in a
11 matches
Mail list logo
