MobileRead Forums - View Single Post - Converting PDF with code samples

yelloguy · 01-18-2011, 12:01 PM

I just downloaded the latest Calibre update released today. And I have tried converting some programming books in PDF format with this new version as well the previous version before this. The problem I have is that the sample code listings are mangled or re-flown during conversion to epub.

I even downloaded Sigil to see if it was something I could fix manually. But it seems like Calibre is outputting the code in a <p> tag with "calibre2" style applied just like the rest of the book. Meaning it is too late for Sigil to do anything.

I also noticed that the page headers are not being removed correctly so I have unnecessary page breaks (and page numbers and text). However, there are options in Calibre to specify a regex. So it's only a matter of (!) learning the right regex and putting it in there. But if there is no way to fix the code listing problem then I'd rather not waste my time on this.

I noticed that every listing in the PDF begins with a header with text "Listing 1.1". I am not sure if there is a way to specify the regex for inserting a <Pre> tag at these places.

Save for these two issues with PDF's, the tool works like a charm. So a grateful thanks to the author kovid. I used it to convert a .CHM with code listings and all to epub. So if it understood that, there's gotta be a way to make it understand this.

01-18-2011, 12:01 PM	#1
yelloguy Enthusiast Posts: 30 Karma: 4698 Join Date: Jan 2011 Device: none	Converting PDF with code samples - losing battle? I just downloaded the latest Calibre update released today. And I have tried converting some programming books in PDF format with this new version as well the previous version before this. The problem I have is that the sample code listings are mangled or re-flown during conversion to epub. I even downloaded Sigil to see if it was something I could fix manually. But it seems like Calibre is outputting the code in a <p> tag with "calibre2" style applied just like the rest of the book. Meaning it is too late for Sigil to do anything. I also noticed that the page headers are not being removed correctly so I have unnecessary page breaks (and page numbers and text). However, there are options in Calibre to specify a regex. So it's only a matter of (!) learning the right regex and putting it in there. But if there is no way to fix the code listing problem then I'd rather not waste my time on this. I noticed that every listing in the PDF begins with a header with text "Listing 1.1". I am not sure if there is a way to specify the regex for inserting a <Pre> tag at these places. Save for these two issues with PDF's, the tool works like a charm. So a grateful thanks to the author kovid. I used it to convert a .CHM with code listings and all to epub. So if it understood that, there's gotta be a way to make it understand this.