Path: utzoo!utgpu!watmath!clyde!att!ucbvax!bloom-beacon!EXPO.LCS.MIT.EDU!rws
From: rws@EXPO.LCS.MIT.EDU (Bob Scheifler)
Newsgroups: comp.windows.x
Subject: Re: Fonts
Message-ID: <8812301258.AA08155@EXPO.LCS.MIT.EDU>
Date: 30 Dec 88 12:58:29 GMT
References: <14034@srcsip.UUCP>
Sender: daemon@bloom-beacon.MIT.EDU
Organization: The Internet
Lines: 14


    In Release 2, the per_char value of a XFontStruct was null when
    the font was fixed in size, but now in Release 3 it points to
    some real per_char information (at least for the 9x15 font), even
    though all characters have the same size, WHY?

Because R2 was broken, and R3 has is righter.  9x15 has constant
character-width, but not constant side bearings, ascent, or descent.
"Terminal" fonts like 9x15 were lying about their "ink" metrics in R2,
but are reporting real metrics in R3.

Also, note that even if a font has constant metrics, the protocol
does not *require* that the per-char stuff be returned empty;
it's an "if", not an "if and only if".