[Editor Plugin] Editor Chains

[nephtys59]

Originally Posted by theducks

Is this an Editor saved search?
Even if you did a change, just abort (Exit without save).
I really want to test out most complex S&R over more than 1 case.

BUT ....

You can just fill out the S&R box . Right-click in the FIND box: Save current search

I have 2 saved searches in the editor. They are complex and I need them to find some special things. But the replacement is manual, I use them only to search. I'd like to use them with editor chains, so with just one click I make both searches, but don't want anything to be replaced. How do I set editor chains to just find an do nothing else? If there is a way ...

[theducks]

Originally Posted by nephtys59

I have 2 saved searches in the editor. They are complex and I need them to find some special things. But the replacement is manual, I use them only to search. I'd like to use them with editor chains, so with just one click I make both searches, but don't want anything to be replaced. How do I set editor chains to just find an do nothing else? If there is a way ...

Sorry, but that makes little sense (as a 1 click)
FIND search1, But don't pause or replace.
FIND search2, But don't pause or replace.

A simple search for 1 of multiple values: [red!black] , but again, Editor chains is for an AUTOMATED task and you want a STOP for manual

[skil-phil]

Hi,
Thanks for all your work.
I would like to request an additional action.

Remove all fonts.
I do not use any of the fonts in a book and rely on the font I select on my reading device.
Just a help in the cleanup.

[theducks]

It is a conversion option: (at the bottom)

On the Styling tab

or
Preferences: Common Options: if you want it as the default

[skil-phil]

Thanks for the tip.
Didn't work for me though.
Tried adding an epub- still there.
Converted epub to epub - still there.
Will try some more. Maybe there is another setting somewhere.

[skil-phil]

Found it does remove fonts that are not imbedded.
Imbedded fonts remain. Now to try to automate unembeding them.
Thanks

[capink]

The plugin has a generic "Remove files" action. It supports regex, so you can use it to remove fonts with this regex

(?i).+\.(otf|ttf)$

You can then use a S&R action to remove @font-face from style files.

Attached below is a sample chain. Please test and verify before saving your books.

[skil-phil]

Thanks capink,
That worked and removed them from the file browser.
They still show in the manage fonts dialogue though.
Is this a problem? Can I just ignore it?
Thanks

[capink]

Manage font dialog will show any font referenced in any CSS rule, with the embedded font having a "✓" sign next to it, as opposed to non embedded fonts. If you want to delete all fonts and all references to them as done in the Manage fonts dialog, you can use this custom action which is based on the remove fonts functionality in "Manage fonts dialog":

from calibre.ebooks.oeb.polish.fonts import change_font, font_family_data

from calibre_plugins.editor_chains.actions.base import EditorAction

class RemoveAllFonts(EditorAction):

    name = 'Remove All Fonts'
    headless = True

    def run(self, chain, settings, *args, **kwargs):
        container = chain.current_container
        font_data = font_family_data(container)
        for font in font_data.keys():
            change_font(container, font)

    def validate(self, settings):
        return True

You can add the custom action by clicking on Editor Chains menu > Manage Modules > add > copy/paste.

Note: Using the above method, you will lose all the font formatting even if the font is installed on your device, as the function deletes all font-family properties from the epub. This apply to previously embedded as well as non-embedded fonts.

[skil-phil]

Thank you.
All good now.
Your hard work certainly makes life easier for users.
Cheers.

[skil-phil]

Hi,
I also found that I needed to add the following to
the Remove all fonts chain.

Search and replace "font-family: (.*?);" with "" in All Style Files.

Cheers

[theducks]

Originally Posted by skil-phil

Hi,
I also found that I needed to add the following to
the Remove all fonts chain.

Search and replace "font-family: (.*?);" with "" in All Style Files.

Cheers

Just delete the line. "" trigger an error on a validate (""; might slide)
Calibre (editor debug) will issues a warning if no generic font is declared in a "font-family"

[skil-phil]

Thanks, will look into that.

[nqk]

I have this python script to rename image.webp to image_width_height.webp, which runs fine on Windows

python path_to_script.py path_to_image.webp

In the Chain Editor - Run Command, the set up is like this
- Choose Binary: python
- arg: "path_to_script" {inputfile}
- extension: webp

It returns no error but nothing is changed.

I tried pasting the code to the "Run Python Code" section, but it doesn't provide the "selected files" option so it returned error.

What should I do? I want the selected webp images to include width and height values in their filenames.

The code is

from PIL import Image
import sys, os

f = sys.argv[1]

if not os.path.exists(f):
    print(f"File not found: {f}")
    sys.exit(1)

im = Image.open(f)
w, h = im.size
im.close()

base, ext = os.path.splitext(f)
new_name = f"{base}_{w}_{h}{ext}"
os.replace(f, new_name)

PS: Merry Christmas and Happy New Year

[capink]

Originally Posted by nqk

I have this python script to rename image.webp to image_width_height.webp, which runs fine on Windows
......

The "Run Command" action operates on the file contents without changing their names. It has an option to change file extension which is sueful when converting to different file format. Other than that it is supposed to work on the content and not the filenames.

The reason for that is that the structure for files in the container is different than the usual directory structure, so the usual code to work on these files like os.replace() is not going to work. Which is why your code does not work as expected.

To work around this, you can mimic the way the "Run Command" action works, by copying the content of the files into temporary files that exist in a classic directory, and run the commands on the temporary file, after which it replaces the corresponsing files in the container with the temporary files.

Here is how to adapt your code into a "Run Python Code" action using the same idea, but since your code does not change the content of the files, we will not need to replace the files, but only to rename them using calibre's rename_files()

from PIL import Image
import os

from calibre.ptempfile import better_mktemp
from calibre.ebooks.oeb.polish.replace import rename_files

supported_formats = ['epub']

def run(chain, settings):
    container = chain.current_container
    name_map = {}

    for name, media_type in container.mime_map.items():
        dirname = os.path.dirname(name)
        basename = os.path.basename(name)
        base, ext = os.path.splitext(basename)
        if media_type.startswith('image'):
            temp_file = better_mktemp(suffix=ext)
            with open(temp_file, 'wb') as f:
                f.write(container.raw_data(name, decode=False))
            im = Image.open(temp_file)
            w, h = im.size
            im.close()
            new_name = f"{base}_{w}_{h}{ext}"
            name_map[name] = os.path.join(dirname, new_name)
            os.remove(temp_file)

    rename_files(container, name_map)

Originally Posted by nqk

PS: Merry Christmas and Happy New Year

Thanks. Merry Christmas to you.

Edit: The above code will work on all image files in the book, if you want only the selected files run this:

from PIL import Image
import os

from calibre.ptempfile import better_mktemp
from calibre.ebooks.oeb.polish.replace import rename_files

supported_formats = ['epub']

def run(chain, settings):
    container = chain.current_container
    name_map = {}

    for name in chain.gui.file_list.file_list.selected_names:
        dirname = os.path.dirname(name)
        basename = os.path.basename(name)
        base, ext = os.path.splitext(basename)
        temp_file = better_mktemp(suffix=ext)
        with open(temp_file, 'wb') as f:
            f.write(container.raw_data(name, decode=False))
        im = Image.open(temp_file)
        w, h = im.size
        im.close()
        new_name = f"{base}_{w}_{h}{ext}"
        name_map[name] = os.path.join(dirname, new_name)
        os.remove(temp_file)

    rename_files(container, name_map)