bug#14233: 24.3; Don't constrain frame size to character multiples

[martin rudalics]

>> One can check whether pixel-edges are equal.
>
> Yes, I think that's a better solution for that.

But we also should try to not break code outside our code base.

>> Also the frame's internal border is drawn in between toolbar and the
>> frame's root window
>
> At least on MS-Windows, I see no border.  Or perhaps I don't know what
> to look for.

With emacs -Q try

(set-frame-parameter (selected-frame) 'internal-border-width 12)

>>  > As for display margins, they do display text, don't they?
>>
>> IIUC they could display anything fringes can display
>
> No.  Fringes can only display bitmaps.  The cannot display text or
> images that we support in the text area or on margins.

Yes.  But I meant that display margins can display images and bitmaps as
well.  Or am I missing something?

> I'm trying to think why the fringes were included.  It is possible
> that the reasons were practical rather then anything else.  Like the
> desire to have the text area contiguous.

With the default they fail precisely in this regard.

>> BTW, I've never been able to understand the manuals and doc-strings in
>> this regard.  Consider the doc-string of `set-frame-width': "Specify
>> that the frame FRAME has COLS columns."  Or its manual entry: "This
>> function sets the width of FRAME, measured in characters."
>
> What's wrong with those?

That they never explain what the "width of a frame" is.  The best
explanation I could find is

http://www.xemacs.org/Documentation/21.5/html/internals_29.html#SEC275

(in particular section 29.7 "The Displayable Area") but it still lacks
an explanation of frame fringe widths (maybe because XEmacs doesn't have
them).

martin