Interesting. This fell off my radar. Maybe it would have been better to
file an official ticket. Emails can get lost.

David Christensen <dpchr...@holgerdanske.com> schrieb am Sa., 15. Jan.
2022, 20:55:

> module-authors:
>
> I submitted a patch for Filter::Cpp about three months ago to Reini
> Urban and Paul Marquess, but never heard back from either of them.
>
>
> I seem to recall that I later tested the changes on macOS and FreeBSD,
> and they failed.
>
>
> Suggestions?
>
>
> David
>
>
>
> -------- Forwarded Message --------
> Subject: Filter-1.60
> Date: Wed, 20 Oct 2021 21:47:14 -0700
> From: David Christensen <dpchr...@holgerdanske.com>
> To: rur...@cpan.org
> CC: p...@cpan.org
>
> Reini:
>
> Thank you for your work on Filter.  :-)
>
>
> Attached please find a tarball (Filter-1.60-dpchrist-20211016.tar.gz)
> with some changes I made to Filter::cpp to help it work better with Perl
> code:
>
> 1.  Added a pre-filter to escape octothorps ('#') not involved with C
> preprocessor control-lines.
>
> 2.  Added a post-filter to unescape octothorps.
>
> 3.  Added an options hash %opt with default values that can be
> overridden via Filter::cpp import() arguments.
>
> 4.  Reworked the cpp and sh invocations to use %opt.
>
> 5.  Added a debugging function, _d(), that is controlled by the
> DEBUG_FILTER__CPP environment variable and writes debug messages to a
> file named after the running script: "$0.debug".
>
> 6.  Reworked cpp.t to dump variables on test failure, and to keep
> generated files on test failure or when debugging is enabled.
>
> 7.  Added generated files to Makefile.PL 'clean' target.
>
>
> At this point, all of the test suite except t/z_manifest.t (skipped)
> passes on Debian GNU/Linux 9, and Filter::cpp works correctly with a
> simple example module (MyLib-0.01-dpchrist-20211016.tar.gz, attached).
>
>
> I would like to contribute my changes to Filter, if you find them
> agreeable.  I am willing to learn and help do the work, but am
> unfamiliar with the process (CPAN, Git, and/or GitHub).
>
>
> Please advise.
>
>
> David
>
>

Reply via email to