Hi, Andreas, for extracting CMAP from the True Type font, I couldn't do it in PDFBox, and used a piece of code from JPedal LGPL source code. Is that okay for you to use that code in PDFBox? I've created the issue PDFBOX-1287 anyway, and attached the patch.
Mehdi, thanks, I guess that is my only remaining issue on rendering PDF to images. -Hamed On Tue, Apr 10, 2012 at 10:22 AM, Andreas Lehmkuehler <[email protected]>wrote: > Hi, > > Am 04.04.2012 09:03, schrieb Hamed Iravanchi: > > Hi, >> >> I managed to fix a few issues with PDF to Image convertion. >> Andreas, please reply. Let me know what can I do to fix these in your code >> too. >> > Please create an issue on JIRA [1] and attach the changes as diff to it. > Add an example pdf too. Maybe is is a good idea to subscribe the dev@list too > where most of the tech discussions take place. > > > What I've done so far: >> - Made all true type fonts use code points (instead of extracted text) to >> render the image >> - Mapped the code point to glyph code by reading the font's CMAP (because >> what I've mentioned in >> http://pdfbox-users.markmail.**org/message/bxfiab2der5dphlh?**page=1<http://pdfbox-users.markmail.org/message/bxfiab2der5dphlh?page=1> >> ) >> - Used glyph codes to draw text >> > Sounds, exactly like my plan. ;-) > > > This fixes ALL of my PDF files that contain true type fonts. >> One of my sample PDF files that have a CIDFontType0 still renders garbage, >> and I think it is because not creating a correct AWT font. >> I've reported the issue along with the PDF file in issue PDFBOX-1278. >> > The embedded font will be substituted if it isn't readable and in many > cases the encoding doesn't work any more, so that one gets garbage. > > > Waiting for your reply, >> -Hamed >> >> > BR > Andreas Lehmkühler > [1] > https://issues.apache.org/**jira/browse/PDFBOX<https://issues.apache.org/jira/browse/PDFBOX> >
