--- Joseph Koshy <[EMAIL PROTECTED]> wrote: > > > >>>>> "jk" == "Joseph Koshy", "ks" == "Keyur Shroff" > > jk> o An implementation is not allowed to place glyphs at > jk> anything other than the /precise/ positions in the > request (i.e. no kerning). > > The first glyph of the glyph list specified in the > PolyText or > ImageText requests has to be drawn with its origin > coinciding with the > [x,y] coordinate given in the request. This is specified > in the > definition of these requests. > > The origin for the next glyph is defined to be: > > [x + character-width,y] > > where "character-width" is a per-character metric. This > information > is however tucked away in the definition QueryFont > protocol request > [approx page 51 of the protocol specification]. > > There is another constraint on text rendering imposed by > the core protocol. > > o PolyText glyph display is defined as a fill operation > using the > current graphics context, with the glyph bits forming a > mask. This > effectively precludes anti-aliased text. ImageText > rendering is > also defined in terms of 2 colors, also precluding > anti-aliased > display of text.
The above mentioned text in X11 specification is only guidelines. X protocol doesn't speak anything more specific to implementations. X11 also predates OpenType font technology. So naturally at that time the writer didn't think at all about glyph's relative origin. Since introduction of anti-aliased text in X11 is not violation of X protocol, how can one consider changing [x,y] position of a glyph as violation of it? - Keyur __________________________________________________ Do You Yahoo!? Yahoo! Greetings - Send FREE e-cards for every occasion! http://greetings.yahoo.com _______________________________________________ Fonts mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/fonts
