Batch-Processing of CSS in epub

[Kiruhdu]

Hi there

I often have to work over a whole series of calibre-created epubs where I do the same corrections in the CSS over and over...

Now I want to ask if there is any way to do this in a batch or something similar?

Much thanks in advance for your help!

[KevinH]

That all depends on what you are trying to do exactly. Both Sigil and Calibre allow regular expression based search and replace which can be used to automate the tasks. Both Sigil and Calibre allow plugins that can auto process any loaded epub. But creating a Sigil or Calibre plugin requires some knowledge of python and some knowledge of the plugin interface mechanism. Sigil's is perhaps easier to learn by example and is limited to the contents of the ebook, whereas Calibre's plugins have a more complicated interface but can do more.

That said, if you know some python or how to use regular expressions (or .bat files and have a tool you use) and your books are all epubs, you should be able to write a script to open (unzip) each epub in a folder, process it as desired, re zip up the revised epub, and repeat.

So it is hard to help without some idea of the changes needed in detail.

[DiapDealer]

I was looking at ways to add an "All CSS Files" option to our F&R parameters, but my head very quickly spun when trying to grasp the logic of that code.

The only way to currently do regex processing on css files is one file at a time with "Current File" selected. I think many would like an "All CSS Files option." But from what I saw, it looked like adding that option would require a minor overhaul of the existing F&R codebase.

But even that's not going to help the OP "batch process" css from many individual epubs. Sigil is really not geared toward batch processing at all.

[najgori]

i open css with notepad++ and run multiple regex search&replace with notepad++ python script plugin.
for sure it can be done with sigils plugin engine but i am used to do it this way.
notepad++ python script plugin comes with couple sample plugins.

edit: it is just one epub with multiple changes. i don't know how to do the same thing on multiple epubs

[elibrarian]

Originally Posted by Kiruhdu

I often have to work over a whole series of calibre-created epubs where I do the same corrections in the CSS over and over...

Now I want to ask if there is any way to do this in a batch or something similar

As I see it, that would require a tool that's able to search & replace inside an epub (or a zip-archive, which is more or less the same thing), without touching anything else in the archive.

The only tool I know of that's able to do that (in Windows anyway), is PowerGrep https://www.powergrep.com/, but it comes with both a learning curve and a price tag, and both sort of steep …

I'm told that there exists a Linux commandline grep-tool to search and (maybe?) replace in zip files - zipgrep. I've not used Linux lately, but you might be able to use it in a shell script or something.

As allways: FWIW.

Regards,

Kim

[kovidgoyal]

Use the transform styles tool in the calibre conversion dialog -- it allows you to create rules to manipulate CSS and can work in bulk. The tool is also present int he calibre editor, but that works one book at a time.

[Turtle91]

You can also just use a template css file that has all of your standard css that you want.

Then just use whatever form of search & replace, powergrep, etc. to change the tags within the html documents to match what you already have in the css.

[AlanHK]

Since epubs are zip files, use a command line unzipper to unzip all to a folder.
Now you can use whatever S&R app or text editor you like that has a batch mode to edit the css and xhtml of all the files of interest at once. Even sed, perl, whatever.

Rezip-- I know that you need to zip without compressing the mimetype file, there are ways to do that, e.g. epubpack.