Re: Native compilation on Windows, was Re: Bootstrap Compilation Speed

[H. Dieter Wilhelm]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> I'm pretty sure the tests aren't shipped with Emacs, while the etc
>> folder is.  The header comments say this is expected to be available
>> for end-users to see if features may be missing under Windows.
>
> FWIW, I'll just (re)mention that I think `w32-feature.el` should be made
> to work on non-w32 systems and then moved to `lisp`.

>From a cursory glance I can't see code which might be restricted to
Windows or "w32" systems, except the naming.  Just ran the tests of
w32-features.el on:

GNU Emacs 28.0.91 (build 1, x86_64-pc-linux-gnu, GTK+ Version 3.22.30,
cairo version 1.15.10) of 2022-01-14

And the result is as expected for my (default) configuration:

  Selector: t
  Passed:  13
  Failed:  3 (3 unexpected)
  Skipped: 0
  Total:   16/16

  Started at:   2022-02-06 18:05:36+0100
  Finished.
  Finished at:  2022-02-06 18:05:36+0100

  ...F.F...F......

  F feature-harfbuzz
      (ert-test-failed
       ((should
         (eq 'harfbuzz
              (car
               (frame-parameter nil ...))))
        :form
        (eq harfbuzz ftcrhb)
        :value nil))

  F feature-json
      (ert-test-failed
       ((should
         (fboundp 'json-serialize))
        :form
        (fboundp json-serialize)
        :value nil))

  F feature-optimization
      (ert-test-failed
       ((should
         (string-match-p "CFLAGS=-O2" system-configuration-options))
        :form
        (string-match-p "CFLAGS=-O2" "")
        :value nil))

Please tell me what else do you expect from w32-feature.el, maybe
extensive build information for any platform? :-)

   Dieter
-- 
Best wishes
H. Dieter Wilhelm
Zwingenberg, Germany