You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The standard 14 PDF fonts have an uncertain future and the recommendation is to embed fonts, even for those 14 standard fonts.
For the future of the numbering feature, we need to enable embedding fonts as part of the PDFWriter or PDF object.
There are a number of issues to think about:
Where would the fonts be embedded?
The natural place would be the PDFWriter object... but... embedding fonts in the PDFWriter object might cause duplication of fonts when importing multiple pages. Hence, the embedding of fonts should be carefully coded, so that the add_referenced method of the PDF object could recognize any duplication.
On this matter, thought should be given to the question, weather the "add_referenced" method have a better duplication recognition algorithm then the basic one currently in place. Speed performance should overrule space conservation.
Metrics calculations
The current metrics calculations will be broken with any external fonts.
Adding fonts support will require adding a new metrics calculation for "dimensions_of" method currently defined in the metrics_dictionary.rb file.
Unicode support
Some languages are written from right to left or up to down (or down to up, but that's more history and is quire of a non issue at the current stage).
Limitations of Font support
The font support should, by definition, be limited. Libraries like Prawn provide PDF Authoring, whereas the CombinePDF Library is looking for providing a pdftk alternative.
Hence, it is more important to have pristine support for limited number of fonts (maybe only TrueType fonts, maybe only OpenType fonts or maybe a closed library of specific unicode royalty-free fonts) rather then an extensive font library.
It seems to me that having a library of hard-coded fonts might be the best solution for allowing the use of Unicode in the page numbering and text box features.
The text was updated successfully, but these errors were encountered:
Hard coders can now import Type0 fonts from existing PDF pages for use in their new PDF files or new PDF pages (using the CombinePDF.create_page and using the CombinePDF::Fonts.register_font_from_pdf_object
It is still inner workings for the CombinePDF model, so I guess it's an irrelevant improvement.
The standard 14 PDF fonts have an uncertain future and the recommendation is to embed fonts, even for those 14 standard fonts.
For the future of the numbering feature, we need to enable embedding fonts as part of the PDFWriter or PDF object.
There are a number of issues to think about:
The natural place would be the PDFWriter object... but... embedding fonts in the PDFWriter object might cause duplication of fonts when importing multiple pages. Hence, the embedding of fonts should be carefully coded, so that the add_referenced method of the PDF object could recognize any duplication.
On this matter, thought should be given to the question, weather the "add_referenced" method have a better duplication recognition algorithm then the basic one currently in place. Speed performance should overrule space conservation.
The current metrics calculations will be broken with any external fonts.
Adding fonts support will require adding a new metrics calculation for "dimensions_of" method currently defined in the metrics_dictionary.rb file.
Some languages are written from right to left or up to down (or down to up, but that's more history and is quire of a non issue at the current stage).
The font support should, by definition, be limited. Libraries like Prawn provide PDF Authoring, whereas the CombinePDF Library is looking for providing a pdftk alternative.
Hence, it is more important to have pristine support for limited number of fonts (maybe only TrueType fonts, maybe only OpenType fonts or maybe a closed library of specific unicode royalty-free fonts) rather then an extensive font library.
It seems to me that having a library of hard-coded fonts might be the best solution for allowing the use of Unicode in the page numbering and text box features.
The text was updated successfully, but these errors were encountered: