What makes you think that the ToUnicode table for that font is bad? It may not be what you expect, but that doesn't make it bad. For all you know, that is the information in the original font…
Leonard From: 王璐 <[email protected]<mailto:[email protected]>> To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Cc: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [poppler] About parseCharName in GfxFont.cc I tried to send the files through attachment, but got rejected from the mailling list The pdf can be found at http://dl.dropbox.com/u/75853179/med-9.pdf Please check the 'LEKSJON' on the top left corner, without ToUnicode map you should get the correct characters. btw, if you try to extract fonts using fontforge, it won't apply ToUnicode for non-ttf fonts. - Lu On Fri, Aug 24, 2012 at 9:33 AM, 王璐 <[email protected]<mailto:[email protected]>> wrote: I've attached a problematic pdf, notice the 'LEKSJON' in the top left corner, if you copy the text out, you'll get LeKSjoN So in the ToUnicode map for that font, both 'E' and 'e' are mapped to 'e' I've extracted the font as 'f2.cff' attached. The font itself is ok. I've also attached a file showing the font->getToUnicode(), the format for each line is GlyphID Unicode [Unicode...] # CharCode You can see problem at lines of 0x45 and 0x65. Thanks - Lu Wang On Fri, Aug 24, 2012 at 9:21 AM, suzuki toshiya <[email protected]<mailto:[email protected]>> wrote: 王璐 wrote: Usually this is done by ToUnicode map, but I've many bad mapping for Type 1 font, where Type 1 font itself provides good mappings. Could you give some concrete examples? Regards, mpsuzuki
_______________________________________________ poppler mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/poppler
