>>>>> On Fri, 28 Apr 2000 12:28:33 -0400 (EDT), "Alexander N. Kabaev" <[EMAIL PROTECTED]>
>said:
AK> I am no expert either, but as far as I remember, there is special field in
AK> XFontStruct to indicate font direction. I guess, even with left-to-right font
AK> there may be characters with negative width. At least, that's what Fteetype
AK> based TTF rasterizer thinks in XFree86 4.0.
Yes, the doc says that the direction member of an XFontStruct is just a hint.
To be sure it isn't a bug, it would be interesting to know which characters
have negative widths.
AK> I have no idea if the formula is correct, but I think it should be right thing
AK> to use _absolute_ values for {max|min}_bounds.width values.
It surely depends on how the average width is used. For example, in a font
where all the widths are negative, one might expect to have a negative average
width. If Lesstif cannot handle right-to-left drawing, it might be better to
round negative widths up to 0.
__Martin
_____________________________________________________________________
This message has been checked for all known viruses by the
MessageLabs Virus Control Centre. For further information visit-
http://www.messagelabs.com/stats.asp
