Re: master d0a66f3e0e: Display a warning for some uses of nil in face attributes.

Re: master d0a66f3e0e: Display a warning for some uses of nil in face attributes.

[Stefan Kangas]

Gregory Heytings <gregory@heytings.org> writes:

> branch: master
> commit d0a66f3e0e668d8c12c54436740c62f8e238a664
> Author: Gregory Heytings <gregory@heytings.org>
> Commit: Gregory Heytings <gregory@heytings.org>
>
>     Display a warning for some uses of nil in face attributes.
>
>     * src/xfaces.c (HANDLE_INVALID_NIL_VALUE): New macro, which displays
>     a warning for invalid uses of nil as a face attribute value.
>     (Finternal_set_lisp_face_attribute): Use the macro for the attributes
>     :foreground, :distant-foreground and :background.
> ---
>  src/xfaces.c | 21 ++++++++++++---------
>  1 file changed, 12 insertions(+), 9 deletions(-)
>
> diff --git a/src/xfaces.c b/src/xfaces.c
> index 70d5cbeb4c..5e3a47d7f8 100644
> --- a/src/xfaces.c
> +++ b/src/xfaces.c
> @@ -3052,6 +3052,15 @@ The value is TO.  */)
>  }
>
>
> +#define HANDLE_INVALID_NIL_VALUE(A,F)					\
> +  if (NILP (value))							\
> +    {									\
> +      add_to_log ("Warning: setting attribute `%s' of face `%s': nil "	\
> +		  "value is invalid, use `unspecified' instead.", A, F); \
> +      /* Compatibility with 20.x.  */					\
> +      value = Qunspecified;						\
> +    }
> +

Thanks, it makes sense to warn users using Emacs 20 face specifications.

But should the newly introduced warning be called out in NEWS?

Also, why not make this into a function?  Is it performance critical
enough to warrant using a macro?

>  DEFUN ("internal-set-lisp-face-attribute", Finternal_set_lisp_face_attribute,
>         Sinternal_set_lisp_face_attribute, 3, 4, 0,
>         doc: /* Set attribute ATTR of FACE to VALUE.
> @@ -3390,9 +3399,7 @@ FRAME 0 means change the face on all frames, and change the default
>      }
>    else if (EQ (attr, QCforeground))
>      {
> -      /* Compatibility with 20.x.  */
> -      if (NILP (value))
> -	value = Qunspecified;
> +      HANDLE_INVALID_NIL_VALUE (QCforeground, face);
>        if (!UNSPECIFIEDP (value)
>  	  && !IGNORE_DEFFACE_P (value)
>  	  && !RESET_P (value))
> @@ -3409,9 +3416,7 @@ FRAME 0 means change the face on all frames, and change the default
>      }
>    else if (EQ (attr, QCdistant_foreground))
>      {
> -      /* Compatibility with 20.x.  */
> -      if (NILP (value))
> -	value = Qunspecified;
> +      HANDLE_INVALID_NIL_VALUE (QCdistant_foreground, face);
>        if (!UNSPECIFIEDP (value)
>  	  && !IGNORE_DEFFACE_P (value)
>  	  && !RESET_P (value))
> @@ -3428,9 +3433,7 @@ FRAME 0 means change the face on all frames, and change the default
>      }
>    else if (EQ (attr, QCbackground))
>      {
> -      /* Compatibility with 20.x.  */
> -      if (NILP (value))
> -	value = Qunspecified;
> +      HANDLE_INVALID_NIL_VALUE (QCbackground, face);
>        if (!UNSPECIFIEDP (value)
>  	  && !IGNORE_DEFFACE_P (value)
>  	  && !RESET_P (value))

Re: master d0a66f3e0e: Display a warning for some uses of nil in face attributes.

[Gregory Heytings]

>
> But should the newly introduced warning be called out in NEWS?
>

I don't think so, the nil value there was supported for compatibility with 
Emacs 20 but was not documented (and the message seems clear enough).

>
> Also, why not make this into a function?  Is it performance critical 
> enough to warrant using a macro?
>

It's not a macro for performance reasons, it's a macro because it didn't 
seem worth to me to make it a function.

Re: master d0a66f3e0e: Display a warning for some uses of nil in face attributes.

[Stefan Kangas]

Gregory Heytings <gregory@heytings.org> writes:

> It's not a macro for performance reasons, it's a macro because it didn't
> seem worth to me to make it a function.

I don't follow your reasoning here.  I would make it into a function
unless there is some specific reason to make it into a macro.

Re: master d0a66f3e0e: Display a warning for some uses of nil in face attributes.

[Eli Zaretskii]

> From: Stefan Kangas <stefankangas@gmail.com>
> Date: Sun, 27 Nov 2022 10:20:05 -0800
> Cc: emacs-devel@gnu.org
> 
> Gregory Heytings <gregory@heytings.org> writes:
> 
> > It's not a macro for performance reasons, it's a macro because it didn't
> > seem worth to me to make it a function.
> 
> I don't follow your reasoning here.  I would make it into a function
> unless there is some specific reason to make it into a macro.

It's the other way around: a function comes with the call overhead, so it
cannot be the default.