MobileRead Forums - View Single Post

roger64 · 01-09-2018, 03:37 AM

Hi

Last October I uploaded to MR the trials concerning a subset generator named pyftsubset, which is part of the fonttools project. The file is named trials.zip (4.24 MB)

As you know, I was - and still am - unable to provide pyftsubset with the list of used italic (bold, bolditalic...) characters to work with and had to make do with the full list of characters provided by the Report tool of your nice Editor.

I would be keen to make further trials using such dedicated lists of characters. If everything goes well (...), I would like to contribute to build a subset plugin for Sigil using the pyftsubset tool.

As you made available the Calibre source code and you told me that the Calibre subset generator makes use of such dedicated lists, I thought that maybe you would accept to finger the relevant lines of code dealing with the creation of such lists of characters and to give the name of the libraries you did use for this purpose.

I have been told that a BS4 parser is not sufficient because it cannot handle the intricacies of CSS contextual rules. So, any comment regarding the creation of such lists would of course be very much appreciated.

01-09-2018, 03:37 AM	#1
roger64 Wizard Posts: 2,608 Karma: 3000161 Join Date: Jan 2009 Device: Kindle PW3 (wifi)	To create a list of italic characters Hi Last October I uploaded to MR the trials concerning a subset generator named pyftsubset, which is part of the fonttools project. The file is named trials.zip (4.24 MB) As you know, I was - and still am - unable to provide pyftsubset with the list of used italic (bold, bolditalic...) characters to work with and had to make do with the full list of characters provided by the Report tool of your nice Editor. I would be keen to make further trials using such dedicated lists of characters. If everything goes well (...), I would like to contribute to build a subset plugin for Sigil using the pyftsubset tool. As you made available the Calibre source code and you told me that the Calibre subset generator makes use of such dedicated lists, I thought that maybe you would accept to finger the relevant lines of code dealing with the creation of such lists of characters and to give the name of the libraries you did use for this purpose. I have been told that a BS4 parser is not sufficient because it cannot handle the intricacies of CSS contextual rules. So, any comment regarding the creation of such lists would of course be very much appreciated.