Mike Maxwell wrote: > Hussein Shafie wrote: >> I didn't realize that. I'm stunned to learn that Acrobat Reader does >> not support the whole Unicode Basic Multilingual Plane > > I'm not sure I understand the problem here. Once we got Java (and > therefore XMLmind) to use an appropriate font for the Bengali region of > Unicode (as documented in emails on this list earlier this year), we did > not have much trouble getting PDFs with *embedded* Unicode Bengali (and > IPA region) fonts. And since we embedded the fonts in the PDFs, Acrobat > Reader (or other PDF reader programs) does not have a problem.
PDF supports 14 *Latin-1* fonts in standard, for example Helvetica. Let's say I want my PDF document to contain Cyrillic text displayed using a SansSerif fonts: nothing fancy at all by nowadays. [1] I cannot use Helvetica. I need to find a SansSerif font supporting Cyrillic. Let's call it FooFont. [2] The PDF must contain references to FooFont and not to Helvetica. [3] You need to embed FooFont in the PDF or instruct the consumers of your PDF to install font packs containing FooFont in Acrobat Reader. I call this a problem. > > [1] You must configure FOP or XEP to use Central Europe Fonts (not > > Courier, Helvetica, and Times). This step is *pretty* *hard*. > > We chose to go through the DocBook-to-WordXML-to-PDF path, rather than > the DocBook-to-FO-to-PDF path; maybe that's the difference. Yes, thanks to the ``PDF creator'' you use. (At one > point, we used to DocBook-to-RTF route, and that also handled the > Bengali fonts just fine; but the WordXML route was much better for other > reasons.) I estimate that configuring FOP or XEP to embed Central Europe Fonts would take me at least half a day of work (and I'm not a rookie). I consider that this is not a problem with FOP or XEP, but a shortcoming of Acrobat Reader and PDF which make it much too hard working with non-Latin-1 documents.
