Re: Return value of FT_Get_Char_Index

[David Turner]

Hello Tom,

> 
> As I previously stated in a previous email note, a Type 1 font might very well
> have a *valid* glyph index of 0, even though the notes for FT_Get_Char_Index
> state that 0 means not defined.
> 
> I propose that the return type of FT_Get_Char_Index get changed to FT_Int or
> FT_Long and a return value of -1 means not defined.  Or, perhaps in the case 
> of
> Type 1 fonts (CFF fonts won't have this problem glyph because index 0 is 
> always
> for /.notdef), we add 1 to the return value, knowing that we should subtract 
> one
> later.  Or place the first glyph defined in the CharStrings dictionary into 
> slot
> 1 instead of slot 0.  Just throwing out ideas...
> 

I'd argue that I prefer to keep the current convention of using 0 as
the ".notdef" character for all formats, even at the cost of "translating"
an internal charmap for certain formats (not only Type1, but also windows
bitmap fonts, pcf, bdf, etc...)

well, just an opinion too.. :-)

Regards,

- David