Attention is currently required from: Alexander Goncharov, Anastasia Klimchuk, Anton Samsonov, Thomas Heijligen.

View Change

4 comments:

• Patchset:

  • Patch Set #1:

    Appreciate your explanation. I don't think there's any point in describing it in the commit message. […]

    Done

  • Patch Set #1:

    I would avoid introducing new pre-processor conditionals, because we are planning to go in the oppos […]

    I totally agree with all your and Alexander's propositions, however have no knowledge of how to test for header presence without preprocessor conditionals — whether it is C23's `__has_include()` check or autoconf's `HAVE_..._H` macro. I went for the latter and tried to adapt not only the Makefile (which I only care about currently), but meson.build as well (which I am not familiar with). While meson also succeeds on both my test systems with different software versions, it fails here on Jenkins, and I have no idea why.

• File include/cli_classic.h:

  • Patch Set #1, Line 21:

    #if !defined(__has_include) && (__STDC_VERSION__ < 202300L)
    #include <getopt.h>
    #define HAVE_GETOPT_H 1
    

    If we cannot identify whether this header file exists on the user's system, then let's assume it doe […]

    Done

  • Patch Set #1, Line 18:

    #if HAVE_GETOPT_H
    #include <getopt.h>
    #elif !defined(HAVE_GETOPT_H)
    #if !defined(__has_include) && (__STDC_VERSION__ < 202300L)
    #include <getopt.h>
    #define HAVE_GETOPT_H 1
    #elif __has_include(<getopt.h>)
    #include <getopt.h>
    #define HAVE_GETOPT_H 1
    #else
    #define HAVE_GETOPT_H 0
    #endif /* __has_include() */
    #endif /* HAVE_GETOPT_H */
    
    #if !HAVE_GETOPT_H
    

    Sorry it took me some time to get to this patch. […]

    Done

To view, visit change 77089. To unsubscribe, or for help writing mail filters, visit settings.