Re: [ft] Origin and Line spacing

[Werner LEMBERG]

Sorry for the late reply.

> For line spacing, i used "height" and it seems to be good with many
> fonts.

I assume the `height' field of the FT_FaceRec field (or
FT_Size_Metrics).  Yes, this is the standard choice.  Typographical
applications often use even bigger values.

> I look at the ftdiff for the corner origin and you use the
> "ascender" to define the top position. That works with some fonts,
> but not with others.  I tried to position ten fonts:
> For the top positionning:
> - 5 was good by using "ascender"
> - 5 was good by using "bbox.yMax"
> How can i know if i must use ascender or yMax?

You can't, I think.  David has apparently just uses a heuristical
value to avoid a gap at the top.  Typographical applications like
groff by default simply move down by `height', then write the top text
line.  However, such programs have plenty of whitespace above the top
line (this is, the top margin).  If you need tight typesetting (say,
within a small box), the only choice is to scan the real height of all
affected glyphs.

> For the left positioning:
> - All works fine with my magic formula : (FontSize * 16) / 72
> I dont know why my formula works, but my text is exactly positioned
> as under Windows C# .NET GDI+

I can't comment on this.

> I got a new problem with symbol fonts like "webdings, wingdings,
> etc.", the characters are not the same as under windows.
> 
> For example with Wingdings with the 'A' character:
> - Windows C# .NET GDI+ : A hand with 2 fingers.
> - Freetype : A moon with a star.

Look up the FreeType mailing list archive for problems with such
fonts.  Windows maps symbol fonts to PUA Unicode values, IIRC.


    Werner