bug#37473: 27.0.50; antialias setting is not preserved by, inheriting

[Eli Zaretskii]

> Cc: handa@gnu.org, larsi@gnus.org, 37473@debbugs.gnu.org, da_vid@orange.fr
> Date: Mon, 20 Jun 2022 16:37:32 +0300
> From: Eli Zaretskii <eliz@gnu.org>
> 
> > From: Po Lu <luangruo@yahoo.com>
> > Cc: handa@gnu.org,  da_vid@orange.fr,  37473@debbugs.gnu.org,  
> > larsi@gnus.org
> > Date: Mon, 20 Jun 2022 21:20:09 +0800
> > 
> > Eli Zaretskii <eliz@gnu.org> writes:
> > 
> > > ??? The original bug report talked about antialiasing, but it didn't
> > > say that other attributes in the :extra property don't exhibit the
> > > same issue.
> > 
> > They might, but apparently :antialias is the only one to have caused
> > enough of a problem for someone to open a bug report.
> > 
> > I guess copying the entire extras alist would work too, if that's what
> > you would prefer.
> 
> Yes, I think that's what we should do, at least as long as the
> solution is based on this technique.

Actually, why not go to the code we had in realize_gui_face before
commit bf0d3f7?  The problem described in bug#1793, which led to that
commit, only happens if one uses a specific (misc-fixed) font family,
not for the usual default fonts used by Emacs, and I'm not sure what's
described there is a bug at all.  At least for the purposes of
text-scale-adjust, it makes no sense to ignore the
foundry/family/adstyle of the original font, because we _want_ the
same (or very similar) font, just of a different size.

And likely in other use cases: if the :font attribute of a face
specifies some font properties, we want to keep them all, not
arbitrarily to ignore some of them.  And definitely catering to an
obscure use case such as the one cited in bug#17973 is NOT a good
reason to make such radical changes in face-realization behavior.

So I think we should just back out that part of commit bf0d3f7, and if
we decide that this causes bug#17973 in too many situations we care
about, I'd rather find a kludge for that specific case than do that
for every face realization.

Specifically, I propose the change for the master branch.  Any
objections?

diff --git a/src/xfaces.c b/src/xfaces.c
index f70fe87..10bde40 100644
--- a/src/xfaces.c
+++ b/src/xfaces.c
@@ -5952,28 +5952,8 @@ realize_gui_face (struct face_cache *cache, Lisp_Object 
attrs[LFACE_VECTOR_SIZE]
            emacs_abort ();
        }
       if (! FONT_OBJECT_P (attrs[LFACE_FONT_INDEX]))
-       {
-         /* We want attrs to allow overriding most elements in the
-            spec (IOW, to start out as an empty font spec), but
-            preserve the antialiasing attribute.  (bug#17973,
-            bug#37473).  */
-         temp_spec = Ffont_spec (0, NULL);
-         temp_extra = AREF (attrs[LFACE_FONT_INDEX],
-                            FONT_EXTRA_INDEX);
-         /* If `:antialias' wasn't specified, keep it unspecified
-            instead of changing it to nil.  */
-
-         if (CONSP (temp_extra))
-           antialias = Fassq (QCantialias, temp_extra);
-         else
-           antialias = Qnil;
-
-         if (FONTP (attrs[LFACE_FONT_INDEX]) && !NILP (antialias))
-           Ffont_put (temp_spec, QCantialias, Fcdr (antialias));
-
-         attrs[LFACE_FONT_INDEX]
-           = font_load_for_lface (f, attrs, temp_spec);
-       }
+       attrs[LFACE_FONT_INDEX]
+         = font_load_for_lface (f, attrs, attrs[LFACE_FONT_INDEX]);
       if (FONT_OBJECT_P (attrs[LFACE_FONT_INDEX]))
        {
          face->font = XFONT_OBJECT (attrs[LFACE_FONT_INDEX]);