Stripping Fonts

[Rob Lister]

Hi all. I apologize if this is covered elsewhere, but [standard excuses for not doing a thorough search here].

When converting an e-book to epub format for use on a Nook Touch, I can't figure out how to strip out the existing font definitions. I have several compilations that, after conversion, show up on the reader in what appears to be Times New Roman bold (exceedingly ugly, whatever it is) and overrides the Nook's font type selection.

There doesn't seem to be a Calibre conversion tool to remove these definitions. Anyway, I know I can do it manually by editing the CSS file directly, but I just can't believe there isn't already a conversion tool for that <-- that's why I think I must be missing something fundamental.

[theducks]

Originally Posted by Rob Lister

Hi all. I apologize if this is covered elsewhere, but [standard excuses for not doing a thorough search here].

When converting an e-book to epub format for use on a Nook Touch, I can't figure out how to strip out the existing font definitions. I have several compilations that, after conversion, show up on the reader in what appears to be Times New Roman bold (exceedingly ugly, whatever it is) and overrides the Nook's font type selection.

There doesn't seem to be a Calibre conversion tool to remove these definitions. Anyway, I know I can do it manually by editing the CSS file directly, but I just can't believe there isn't already a conversion tool for that <-- that's why I think I must be missing something fundamental.

I use Sigil with the following REGEX S&R to strip all the excess (MSWORD) font definitions.

Search:  \@font-face \{.+\}\s\s\s  Replace: (nothing)

The problem with this is: there may be a single Nice font used for Initial caps or alternate characters.
You just reduced the book to a step above mono-spaced text

[Rob Lister]

Originally Posted by theducks

I use Sigil with the following REGEX S&R to strip all the excess (MSWORD) font definitions.

Search:  \@font-face \{.+\}\s\s\s  Replace: (nothing)

The problem with this is: there may be a single Nice font used for Initial caps or alternate characters.
You just reduced the book to a step above mono-spaced text

Thanks for that. Still, given the heuristic processing already available in Calibre, I'm just surprised Font normalizing wasn't included...it seems so very basic. There should at least be a global substitution tool, even if it included flashing red 40pt warnings about killing the author-intended styles.

The fact that it is not there, and with no [obvious] mention of it in the Calibre FAQs, makes me think twice about creating a script or plug-in to do automatically.