Re: Hollow cursor under images

[Eli Zaretskii]

> From: Evgeny Zajcev <address@hidden>
> Date: Mon, 27 Jan 2020 15:54:56 +0300
> Cc: Eli Zaretskii <address@hidden>, emacs-devel <address@hidden>
> 
> This patch implements (box . WIDTH) cursor type, allowing user to define his 
> own notion for "Large" image

Thanks.  This needs a NEWS entry and also an update for the ELisp
manual.  See also a few comments below.

> From 6406d198e9f3f5b2d93987ed0fd4fe22bd590624 Mon Sep 17 00:00:00 2001
> From: Zajcev Evgeny <address@hidden>
> Date: Mon, 27 Jan 2020 15:49:46 +0300
> Subject: [PATCH] * Support for (box . WIDTH) `cursor-type'
> 
> Before this commit, block cursor becomes hollow under "large" masked
> images.  Notion for "large" was hardcoded to be image larger then
> 32x32 in any dimension.
> 
> This patch allows user to define his own notion for "large" image,
> taking full control of block cursor look under masked images.
> 
> With cursor-type equal to `box' cursor will always be block under
> masked images.  This differs from former behavior for box cursor.
> To get former behavior, set `cursor-type' to (box . 32)

Please include in the log message a ChangeLog-style list of files and
functions where you are making changes.

> +  (box . WIDTH)   display a filled box cursor, but make it
> +                  hollow if cursor is under masked image larger then
> +                  WIDTH in either dimention.
                                     ^^^^^^^^^
"dimension"

Also, please make it clear that WIDTH is measured in pixels.

> +Except for (box . WIDTH) case.

This is not a complete sentence, please clarify it.

> +  if (CONSP (arg)
> +      && EQ (XCAR (arg), Qbox)
> +      && RANGED_FIXNUMP (0, XCDR (arg), INT_MAX))
> +    {
> +      *width = XFIXNUM (XCDR (arg));

This calls XFIXNUM no less than 3 times.  I wonder if we could tweak
the code to do that only once.

> +      return FILLED_BOX_CURSOR;
> +    }
> +
>    if (EQ (arg, Qhollow))
>      return HOLLOW_BOX_CURSOR;

If the condition above is not fulfilled, you then make the code do 5
more gratuitous comparisons, before it falls back on
HOLLOW_BOX_CURSOR.  It is much better to return HOLLOW_BOX_CURSOR
right away if the conditions for FILLED_BOX_CURSOR weren't satisfied,
it makes the code which implements this feature much easier to read
and understand.