Convert text formating from CSS to HTML

[abraum]

Hi,

i offen convert scanned books to epub using first abbyy finereader and then using sigil or calibre editor. abbyy finereader is using css to format the text (bold, italic, ...) and not html (<b>,<i>,...). Sometimes i have to make a lot of customize the epub file because abby finereader makes crazy things. a lot of css in the code doesnt make it easier. sigil and calibre can display bold or italic text as bold or italic directly in the code. html is much shorter then css and its clearer.

There is nbothing i can do in Finereader so that FR is using HTML instead of CSS when creating the ePub file. Can sigil or calibre do this? converting the most important css into html? something like:

<span style="font-weight: bold;">text</span>

into

<b>text</b>

Or can this be done with a normal html editor?

[BeckyEbook]

Use the TagMechanic plugin.

[Turtle91]

Tag mechanic certainly works!

I do these kinds of changes constantly so I just use a regex find/replace:

find: <span style="font-weight: bold;">(.*?)</span>
replace: <b>\1</b>
or
replace: <strong>\1</strong>

find: <span style="font-style: italic;">(.*?)</span>
replace: <i>\1</i>
or
replace: <em>\1</em>

Unfortunately it's rare to see two ebooks that use the same style (spaces and ; are different) so I just highlight the <span ....>...</span> and copy that down to the find field, then I replace the text between the span tags with "(.*?)".

You could even make the regex more robust like:
<span style="\s*font-style:\s*italic\s*;*\s*">(.*?)</span>
<span style="\s*font-weight:\s*bold\s*;*\s*">(.*?)</span>

and save it as a saved search (group). That fixes all of them with a single mouse click.

Beware: this process assumes that the ebook spans are simplistic. I have seen some books that throw spans around EVERYTHING and this regex would not correctly select the corresponding </span>. In that case it's probably easier to use tag mechanic.

Code example:

barf:
<p class="para"><span><span class="calibre9"><span><span class="calibre1">
Hello,</span></span><span><span class="italic">buddy!
</span></span><span><span class="calibre1">How are you?
</span></span></span></span></p>

clean:
<p>Hello, <em>buddy!</em> How are you?</p>

[abraum]

I'll test the regex first because i use calibre editor more then sigil. Thanks for that.

[KevinH]

Originally Posted by abraum

I'll test the regex first because i use calibre editor more then sigil. Thanks for that.

Then why ask your question in the Sigil forum. Why not ask it in the calibre forum?

[Tex2002ans]

Originally Posted by abraum

i offen convert scanned books to epub using first abbyy finereader and then using sigil or calibre editor. abbyy finereader is using css to format the text (bold, italic, ...) and not html (<b>,<i>,...).

I explained exactly this a few months ago in:

• "Regex Examples" (Post #689)

Go down to where I said:

Originally Posted by Tex2002ans

Back in 2020, I partially wrote about my "12-step Finereader Cleanup" (Sigil Saved Searches).

Here's the last 5 steps of my Saved Searches dealing with Finereader tables:

[...]

Like BeckyEbook said, TagMechanic is your best friend.

Step 1. In Sigil, press Tools > Saved Searches.

Step 2. Right-Click in the list and press New Group. Name the new Group "Finereader Cleanup".

Step 3. Then fill it with entries such as:

Name: Fix Bold
Find: <span style="font-weight:bold;">
Replace: <span class="bold">

Name: Fix Italics
Find: <span style="font-style:italic;">
Replace: <span class="italics">

[...]

Step 4. In your Saved Searches, you can click on the bold "Finereader Cleanup" to highlight it, then press Replace All.

This will run all those Search/Replaces on your selected files.

Now ugly Finereader HTML like:

This is <span style="font-weight:bold;">bold text</span>.

changes into the more human-readable:

This is <span class="bold">bold text</span>.

* * *

Then you can use TagMechanic to change those <span class="bold"> into <b>:

This is <b>bold text</b>.

For example, here's one mini TagMechanic tutorial I wrote in 2020:

• "Replacing All Tags Of One Kind With Another" (Post #19)

[CalibUser]

@ abraum: You could try the ePubTidyTool plugin [https://www.mobileread.com/forums/sh....php?t=264378].

This can remove unwanted tags and can change tags that contain, e.g.,

<span style="font-weight: bold;">text</span>

to

<b>text</b>