Using Editor to mass delete anything contained in a Class?

[fogice]

I used the FanFicFare plugin to download a work from RoyalRoad. Near the beginning and/or the end of most of the chapters is an author's note, which I'd like to remove. There are over 600 chapters, I don't want to do it manually.

Each of these notes is within a section marked as follows:

< div class="portlet" >

Which seems like it would be easy to clear, but there are more < div > within the note, so I can't just regex with < /div > as the endpoint of the find.

Any thoughts on how to handle this?

Full code chunk with plaintext changed:

<div class="calibre1"><div class="portlet">
                    <div class="calibre1">
                        <div class="calibre1">
                            
                            <span class="calibre2">A note from Author</span>
                        </div>
                    </div>
                    <div class="calibre1"><p class="calibre4"><span>I want to thank all of my patrons, including:</span></p>
<p class="calibre4"><span>Person</span></p>
<p class="calibre4"><span>Person</span></p>
<p class="calibre4"><span>Person</span></p>
<p class="calibre4"><span>And my 2 other patrons!</span></p>
<p class="calibre4"></p>
<p class="calibre4"><span>Thank you!</span></p>
<p class="calibre4"></p>
</div>
                </div>

[theducks]

Use the following Calibre (editor) Plugin
https://www.mobileread.com/forums/sh....php?p=2980740
It accepts arguments on which to apply

[fogice]

Originally Posted by theducks

Use the following Calibre (editor) Plugin
https://www.mobileread.com/forums/sh....php?p=2980740
It accepts arguments on which to apply

Nice tool! But I don't think it does what I want it to. It successfully removed the div in question. However, it left all the text within the div. I want to remove the whole block (in my example, the list of patrons).

[Karellen]

This will take a few separate regex's to accomplish.

First you need to "Fix HTML - All files" and then "Beautify all files" which are tools in the editor. The spacing, tabs and layout look a mess. You need to get consistency across all the tags.

You also need to check that the classes being used in the tags are unique to these Authors Notes otherwise you will also be removing parts of the book you want to keep.

Then you need to step through the regexes. Remember to also capture the tabs and line before so you don;t end up with lot's of spacing.
Start with removing the empty <p class="calibre4"></p> tags.
Then remove the <p class="calibre4"><span>(.*?)</span></p> tags.
Then remove the <div> tags using Edit Spans & Divs, which I think is part of Diap's Toolbag plugin.
Then run the "Fix HTML - All files" and then "Beautify all files" again to remove all the extra line feeds.

[fogice]

Originally Posted by Karellen

This will take a few separate regex's to accomplish.

Thanks for this very clear writeup!

Unfortunately, as I dug further into the edits, things keep changing in the layout, and sometimes the classes are reused for things that need to be kept. But I've learned a lot today, and it will be useful for future edits.

[JSWolf]

Originally Posted by fogice

Thanks for this very clear writeup!

Unfortunately, as I dug further into the edits, things keep changing in the layout, and sometimes the classes are reused for things that need to be kept. But I've learned a lot today, and it will be useful for future edits.

It sounds like a manual edit of all 600 chapters is in order.

[Brett Merkey]

Is the .portlet class reused? If not, you can skip working with the html and use the stylesheet:

div.portlet {display: none;}

[Karellen]

Originally Posted by fogice

things keep changing in the layout, and sometimes the classes are reused for things that need to be kept.

I am wondering if you could be more strict with the search for those classes reused elsewhere in the book.

Maybe you could be strict on the spacing. Something like...
\s{6,12}<p class="calibre4"><span>
The numbers in the brackets are min,max spaces to search for.

Without seeing the book, maybe the authors note, with multiple div wrappings, might bump the spacing more than what is used in the rest of the book.

[kovidgoyal]

Just FYI the correct tool to use for this is the Transform HTML tool which is under Tools->Transform markup

in it you can create a rule like if the tag has class whatever, remove it.

[Karellen]

If the same class is being used in the block to be removed, as well as in areas not to be removed, how do you ensure with the Transform tool that you do not remove the wrong elements?

[kovidgoyal]

Originally Posted by Karellen

If the same class is being used in the block to be removed, as well as in areas not to be removed, how do you ensure with the Transform tool that you do not remove the wrong elements?

depends on what information is present to discriminate between the scenarios in the first place. My point was more that for the task of editing HTML rather than relying on regex which is fragile use the transform html tool.

[fogice]

Originally Posted by kovidgoyal

Just FYI the correct tool to use for this is the Transform HTML tool which is under Tools->Transform markup

in it you can create a rule like if the tag has class whatever, remove it.

Oh, that was lovely.

If the tag has class portlet
Remove tag and children

That was exactly what I needed. One shot and 500 chapters are fixed. It's amazing how much depth and power Calibre contains, you are a genius, Kovid.

[fogice]

Originally Posted by Karellen

If the same class is being used in the block to be removed, as well as in areas not to be removed, how do you ensure with the Transform tool that you do not remove the wrong elements?

In my case, it was some classes like calibre4 and calibre2 that became other things in other chapters. The 'portlet' tag was consistent.

[thiago.eec]

Originally Posted by fogice

I used the FanFicFare plugin to download a work from RoyalRoad. Near the beginning and/or the end of most of the chapters is an author's note, which I'd like to remove. There are over 600 chapters, I don't want to do it manually.

Each of these notes is within a section marked as follows:

< div class="portlet" >

Which seems like it would be easy to clear, but there are more < div > within the note, so I can't just regex with < /div > as the endpoint of the find.

Any thoughts on how to handle this?

The regex below should work.
Choose 'Regex' mode and 'All text files'.
Remember to check the 'Wrap' and 'Dot all' options.

<div class="portlet">.*?<p class="calibre4"><span>Thank you!</span></p>.*?</div>

______
Edit: Nevermind. I didn't notice you already solved the problem.