Here are patches based on revision 1686. Please try them.
There is a little change inside PdfFontCID class. Generation of Widths and ToUnicode tables is delayed until EmbedFont is called. I do not see a reason to do generate them twice. If this tables are created first in constructor, they will be huge as all chars must be included. Later, they will be reduced any way to used subset characters only. BTW, a do not like a dynamic cast's which are some times required, as there in no other way to get a real FontMetrics implementation from FontMetrics object. Did you think about extending that class with a two virtual methods which can be overridden in implementation? PfdFontMetricsFreetype* PdfFontMetrics::GetPdfFontMetricsFreetype(); PfdFontMetricsBase14* PdfFontMetrics::GetPdfFontMetricsBase14(); or PdfFontMetrics* PdfFontMetrics::GetPdfFontMetricsImplementation(PdfFontMetrics::Freetype | PdfFontMetrics::Base14) Another thing which is missing, is ability te get actual font face index selected when GetFontFromCollection is used. Later if we try to embed this font then the same logic should be applied again or wrong font might be included (regular vs italic or bold). But this problem is introduced before. 23.10.2015. u 23:22, Palmer Zent wrote: > Hi, > > Sorry for the delay. I’m having issues with your patch file. Can you > regenerate it using "svndiff” based on the latest HEAD? > > I’ll have time in the next few days to run a few tests and commit it. > > Thanks >
subset3.patch.gz
Description: GNU Zip compressed data
------------------------------------------------------------------------------
_______________________________________________ Podofo-users mailing list Podofo-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/podofo-users