[Plugin] RemoveUnusedBookmarks - removes all unused bookmarks from epubs

[slowsmile]

Update: The following changes have been made to the plugin(v0.1.4):

* The href ids are now more widely gathered from the following files: xhtml files, toc.ncx, opf manifest and opf guides. These href ids are then compared to existing ids/bookmarks in the xhtml files. The xhtml ids are only removed if they don't exist in the href id list.

-- NCX pagelists and page-map.xml files are not supported by this plugin. If either are detected in an epub then there will be a warning with bailout.

[slowsmile]

Update: Changes to RemoveUnusedBookmarks v0.1.5

1. Fixed several bugs which, under certain conditions, would give epubcheck errors.
2. This plugin can now process epub 2 and epub 3 files. My thanks to @democrite for the suggestion.

[slowsmile]

Update: Changes to plugin(v0.1.6)

• Improved html output(layout).

[slowsmile]

Update: Changes to plugin(v0.1.7):

• Fixed a bug whereby if the epub contained SVG wrappers then the SVG code would be corrupted by the plugin, causing epubcheck errors.

[democrite]

Thank you very much for the update. A request: any chance there could be an option to not reformat HTML?

[slowsmile]

@democrite...I'm not sure what you mean by "an option not to reformat the html". It's usually inevitable that the html output will always change after you run any plugin. In HTML view the output layout from the plugin looks similar to when you click Tools > Reformat HTML > Mend and prettify All HTML Files in Sgil. Can you please be more specific concerning your request or problem?

[democrite]

I am aware that bookmarks (IDs) will be removed but the reformat like as does Sigil with the option you mention I would prefer not to have. I kind of like my EPUBs formatted a particular way and would like to keep it if possible.

[slowsmile]

After running the plugin, this should not effect the text in Preview. And, in Sigil, if you click on Tools > Reformat HTML > Mend and prettify... this will standardize the layout of the html. That's really the best that I can offer.

I'm afraid I disagree with your view. If you would like me to standardize html output layout from the plugin to just your liking then I do not regard that as a reasonable request. I also think that it's more reasonable for the html output layout to be standardized to the normal epub layout(i.e. the layout that you get after running mend and and prettify).

[democrite]

slowsmile,

I am not asking for my or some arbitrary standard reformating of html. If one wants to reformat, there is Sigil itself as you mention. Such a command I do not use. I format my EPUBs to my liking, making easier for me to see hierarchy and various element types as I edit, make corrections, etc. I do not use Sigil's command unless needed and easier than reformatting some commercial EPUB, such as in the case of some publishers that add intra-element line breaks in the case of line breaks at some line length as is done with programming code. I prefer no reformatting at all unless requested and done by the user. I think you use BeautifulSoup? I imagine such a mature library should be able to manipulate code while leaving it as is, just removing IDs. To repeat to be clear, no reformatting at all, just removal of IDs. If such is the case, perhaps that should even be default, as you are aware if one wishes to reformat, they can simply use Sigil. If BS is not capable of such, maybe it is more suitable to request the library to be able to do such.

Your plugin is invaluable. So often, at least somewhat regularly, I have some self-made EPUB or commercial with thousands of IDs for references, footnotes, hyperlinks, etc. After ensuring that they are all properly linked, I like removing unused IDs. Does it make some difference on modern hardware? If there are 10,000+ IDs and many can be removed? Who knows. I feel good knowing performance and RAM usage is as best as can be, even if marginal. Yet maybe it does make a difference. So I use your plugin. The ability to not reformat would be a most welcome addition.

[JSWolf]

delete post

[slowsmile]

@democrite...When I initially copy all your epub html files to be worked on by my plugin I use bk.text_iter() to fetch the files and use bk.readfile() to read the files into data. Then I run the data from each file through BeautifulSoup in the normal way. And it is there that your html layout will get changed -- it's not my plugin code that deliberately changes your html layout. And since I have no control over that, it is for that reason that I cannot "leave your html layout alone". That's also the reason why I usually always reformat the html output layout to look like it does after you run Tools > Reformat HTML > Mend and prettify... which, as I've already mentioned, helps to standardize the html output layout for each file(which most plugin users seem to prefer anyway).

[democrite]

Thank you for the explaination. From looking around a bit, it seemed perhaps possible to use BS without it reformatting. I may look into it further someday.

Btw, since your plugin operates on bookmarks, would you consider removal of duplicate IDs within each file? If for example I use regex to add IDs to footnotes for later linking, sometimes paragraphs can be such that there are dupes.

[KevinH]

You may want to take a look at the extras in sigil_bs4 versus normal bs4.

See: https://github.com/Sigil-Ebook/Sigil...s4/__init__.py

For "serialize_xhtml" near line 470 which tries to rebuild the source as close as possible to what was parsed, similar to how Sigil itself does it.

[slowsmile]

@democrite...Regarding the removal of duplicates, I doubt if you could do that automatically with a plugin because the plugin will not be able to automatically decide which duplicate IDs to remove. You could perhaps do it using a plugin via a dialog window that specifically allows you to choose the duplicate IDs for removal. But, to be honest, removing duplicate IDs is probably best achieved with a new plugin.

[Vanguard3000]

Hi, thanks for creating this plugin - it really helps me clean up retail epubs that use a lot of unnecessary ids. However, I've noticed it isn't able to clean out all unused ids. For example:

<h1 id="bm4">TITLE</h1>

<h2 id="bm4-s01">SECTION</h2>

In this case, "bm4" is unused, but "bm4-s01" is. If I rename "bm4" to, say, "banana" your plugin will catch it, so it seems to be thinking that, because the "bm4" string is present in the used "bm4-s01" it thinks that they're both in use.

Anyway, I know this plugin is a bit on the old side but is there a chance this could be addressed? Alternatively, is there another similar plugin that will find these ids? Thanks in advance.