> On Jun 10, 2017, at 4:26 PM, Peter Backes <r...@helen.plasma.xg8.de> wrote:
> 
> Hello,
> 
> popt ignores /.popt if /etc/popt.d is empty.
> 
> See https://bugzilla.redhat.com/show_bug.cgi?id=1051685 (you can find a 
> patch there).
> 

Applied: thanks for the pointer to rhbz#1051685!

> Oddly enough, the current CVS version reads /.popt, but only 
> because it has another bug which causes it to ignore /etc/popt.d 
> completely. Here is a patch for that bug:
> 
> Index: poptconfig.c
> ===================================================================
> RCS file: /v/rpm/cvs/popt/poptconfig.c,v
> retrieving revision 1.67
> diff -u -r1.67 poptconfig.c
> --- poptconfig.c      27 Jun 2016 03:00:40 -0000      1.67
> +++ poptconfig.c      10 Jun 2017 20:16:30 -0000
> @@ -509,11 +509,11 @@
> #if defined(HAVE_GLOB_H)
>     {
>     struct stat sb;
> -    if (!stat("SYSCONFDIR/popt.d", &sb) && S_ISDIR(sb.st_mode)) {
> +    if (!stat(SYSCONFDIR "/popt.d", &sb) && S_ISDIR(sb.st_mode)) {
>       const char ** av = NULL;
>       int ac = 0;
> 
> -     if ((rc = poptGlob(con, "SYSCONFDIR/popt.d/*", &ac, &av)) == 0) {
> +     if ((rc = poptGlob(con, SYSCONFDIR "/popt.d/*", &ac, &av)) == 0) {
>           int i;
>           for (i = 0; rc == 0 && i < ac; i++) {
>               const char * fn = av[i];
> 

Applied: thanks for the patch!

> When will a new version of popt be released? The most recent version, 
> 1.16, is now over seven years old! The current CVS version seems to be 
> under active development. Please put out a new version of popt.
> 

Yes popt-1.17 is 7+ years from popt-1.16 release.

Yes, I am still actively maintaining popt, going on nearly 20 years now. RPM is 
intimate with
POPT in very complicated ways, unlike most other applications that link POPT.

Meanwhile, please look at reality:

    1) you are the first to post a patch “upstream". The report rhbz#1051685 
report is dated
                2014-01-10 16:34:19 EST
        (and your patch to rhbz#1051685 was posted  2017-06-09 19:27 EDT)

     2) Both the original implementor/requestor of POPT globs as well as the 
Fedora maintainer
        are members @rpm5.org where POPT is maintained. Neither has chosen to 
“fix” the issue(s),
        likely because they have better things to do these days.

I’d suggest that I (as “upstream maintainer”) am not the bottleneck to a 
popt-1.17 release.

I can do a popt-1.17 release (as in preparing the tar ball and pushing to 
http://rpm5.ort/files/popt
within an hour. The due diligence of testing across multiple platforms and 
operating systems
is beyond my means/energy these days, and takes perhaps 2 weeks, drudgery 
setting up
VM’s to explicitly test a POPT release.

I simply cannot justify the work — either by POPT  “feature set” or “bug 
fixing” changes -- to spend the
necessary time to release popt-1.17 so far (and both of the patches I just 
applied are rather minor:
the core issue is one of design, and add glob(3) to popt was a bad mistake 
IMHO).

Anyways, I can certainly be convinced of a need to release popt-1.17, just 
there doesn’t seem to 
be any interest: you message is the first request since
        http://rpm5.org/community/popt-devel/0288.html
dated nearly 3y ago: Tue 26 Aug 2014

That’s not very much interest in a popt release IMHO.


Again, I can easily be convinced otherwise, and release popt-1.17, if there is
sufficient interest in helping to do the tests on multiple platforms and
operating systems (to which I no longer have routine access, and the VM setup 
drudgery
is far lower priority than my other RPM development interests atm).

hth

73 de Jeff
            
> Best wishes
> Peter
> -- 
> Peter Backes, r...@helen.plasma.xg8.de
> ______________________________________________________________________
> POPT Library                                           http://rpm5.org
> Developer Communication List                       popt-devel@rpm5.org

Reply via email to