Some CMap mappings are copyrighted by Adobe, if I remember correctly. In PDFBox which is from Apache Software Foundation, you need to download the missing files from Adobe and compile the code manually so you can use it.
Regards, Ted Chien -- Sent from my HTC EVO 4G LinkedIn: http://linkedin.com/in/htchien ----- Reply message ----- 寄件者： "Yuan Chao" <yuanc...@gmail.com> 日期： 周四, 11 月 18 日, 2010 年 8:03 上午 主旨： [hojia] How to read Chinese PDFs anymore with xpdf? 收件者： <ho...@tossug.org> 副本： <der...@glyphandcog.com>, <591...@bugs.debian.org> On Wed, Nov 17, 2010 at 10:55 AM, <jida...@jidanni.org> wrote: > Can anybody read the Chinese these days with xpdf anymore? > wget -O x.pdf > http://www1.hl.gov.tw/bus/upload/%AA%E1%BD%AC%AB%C8%B9B%A5%FA%C2%D7%BDu%AE%C9%B6%A1%AA%ED.pdf > xpdf x.pdf > See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=591094 > Should I dump xpdf and instead use what? I can read it in Ubuntu lucid. But other PDF listed in the bugreport shows glyphs with broken radicals. It's obvious due to the byte-code-interpreter (BCI) settings of FreeType. Most of the usage of BCI in western fonts are for grid fitting (hinting), but some chinese fonts use it to assemble glyphs from radicals. (the well-know mingliu and kaiu from windows) As the related patents are expired, BCI is enabled by default in FT 2.4. But some distros still choose to turn it off by default or use some special rules to get around with it. Also, you'll need to re-compile xpdf against BCI enabled FreeType. ps. anyone know why Ubuntu only packs limited CMap nowadays? Others goes to poppler-data? -- Best regards, Yuan Chao _______________________________________________ ho...@tossug.org http://tossug.org/mailman/listinfo/hojia
