RE: Display of Mongolian in Arabic or Hebrew documents
Peter Constable <[email protected]>
| Newsgroups | gmane.text.unicode.devel |
|---|---|
| Message-ID | <DDB6DE6E9D27DD478AE6D1BBBB83579561E4CB310C@NA-EXMSG-C117.redmond.corp.microsoft.com> |
> From: [email protected] [mailto:[email protected]] > > In a text-drawing API implementation, text is drawn along a path; > > there is no notion or awareness of a page. All that matters is how > > glyphs are displayed in relation to that path. > BTW I take it here you are refering to Windows text-drawing API. I strongly suspect that in any graphics library or OS platform at some level in text-drawing stacks you will find it going through an API that draws text along a path -- possibly with some parameterization for modes that relate to directionality, but with no direct awareness of a page. For instance, I don't know much about Cocoa, but glancing through Apple's docs, it looks like drawGlyphsForGlyphRange:atPoint: is a key interface. It only takes a glyph range and a point on the drawing surface as parameters; everything else is controlled at a higher level, with the default path (I assume) being a horizontal line extending to the right from the specified point. If you want a class that will layout out a line horizontally with embedded Mongolian runs set vertically, that would require a bunch of processing above that layer. When you draw Mongolian text through that API without any path transforms applied, I expect it will draw the Mongolian string horizontally (like a vertical Mongolian string rotated on the page 90° CCW). (I'm sure any number of people here can correct me if I'm wrong.) Peter