[Editor Plugin] EpubCheck

[Doitsu]

Originally Posted by Karellen

When I ran EpubCheck earlier today, a message popped up stating that v5.2.6 was available and to upgrade. I dismissed the notice as I was in the process of working on a file, but now I can't seem to get that message back, nor is there an update in the first post of this thread.

Does anybody know where I can download the update?

In Calibre, select Preferences > Miscellaneous > Open Calibre Configuration Folder > Plugins. Locate EpubCheck.json and change this line to an earlier date and re-run the plugin. For example,

  "last_time_checked": "2023-12-25 13:47:16.570619",

Alternatively, download the official .zip file, unzip it and extract the jar file and the lib folder to the EPUBCheck folder in the plugin directory.

[Karellen]

Originally Posted by Doitsu

In Calibre, select Preferences > Miscellaneous > Open Calibre Configuration Folder > Plugins. Locate EpubCheck.json and change this line to an earlier date and re-run the plugin. For example,

Thank you @Doitsu, that worked perfectly. Now running...

EPUBCheck v5.2.0
Validating using EPUB version 3.3 rules.
No errors or warnings detected.
Messages: 0 fatals / 0 errors / 0 warnings / 0 infos

EPUBCheck completed

[ryder]

Hey everybody, when I use the epubcheck in the calibre editor I get an error massage:
FileNotFoundError: [WinError 2]
So unfortunately it's not working for me. I am on win11, calibre 7.23. Is this a known bug with the current calibre version or was my plugin installation not successful? Thanks a lot

[ryder]

Hey everybody, I can't get the epubcheck plugin to work. Running on win11, calibre 7.23. I get the error massage:
calibre, version 7.23.0
FEHLER: Unbehandelter Ausnahmefehler: <b>FileNotFoundError</b>:[WinError 2] Das System kann die angegebene Datei nicht finden

calibre 7.23 embedded-python: True
Windows-10-10.0.26100-SP0 Windows ('64bit', 'WindowsPE')
('Windows', '10', '10.0.26100')
Python 3.11.5
Windows: ('10', '10.0.26100', 'SP0', 'Multiprocessor Free')
Interface language: de
EXE path: C:\Program Files\Calibre2\calibre-parallel.exe
Successfully initialized third party plugins: DeACSM (0, 0, 16) && Gather KFX-ZIP (from KFX Input) (2, 19, 0) && DeDRM (10, 0, 9) && Package KFX (from KFX Input) (2, 19, 0) && EpubCheck (0, 2, 5) && KFX metadata reader (from KFX Input) (2, 19, 0) && From KFX (2, 19, 0) && KFX Input (2, 19, 0) && Set KFX metadata (from KFX Output) (2, 13, 0) && KFX Output (2, 13, 0) && KePub Output (3, 7, 3) && KoboTouchExtended (3, 7, 3) && Obok DeDRM (10, 0, 9)
Traceback (most recent call last):
File "calibre_plugins.epub_check.main", line 199, in ask_user
File "calibre_plugins.epub_check.main", line 123, in get_arch
File "calibre_plugins.epub_check.main", line 114, in jarWrapper
File "subprocess.py", line 1026, in __init__
File "subprocess.py", line 1538, in _execute_child
FileNotFoundError: [WinError 2]

What could be the problem of this? Thanks a lot

[ryder]

Edit: I found some interessting method to get rid of all epub errors. At least the one which are shown by the internal calibre bugchecker. Maybe somebody can confirm if this is also working for epubcheck as well, because as mentioned in this thread, epubcheck finds more bugs than the internal bugcheck. So all I do is this:
Select around 50 ebooks already in epub format (with various bugs), convert them from epub to mobi, delete all epub formats, convert them back from mobi to epub, delete all mobi formats and there will be no more bugs found in the internal bugcheck scan. I read somewhere here that it could help to convert epub to epub again, but there were still some bugs in the scan afterwards. With this method I see none. This method takes less than two minutes for 50 books (depending on your pc power of course). So let me know if this fixes most of the errors in epubcheck as well. That would be great. Thanks again

[PeterT]

Originally Posted by ryder

Hey everybody, I can't get the epubcheck plugin to work. Running on win11, calibre 7.23. I get the error massage:
calibre, version 7.23.0
FEHLER: Unbehandelter Ausnahmefehler: <b>FileNotFoundError</b>:[WinError 2] Das System kann die angegebene Datei nicht finden

calibre 7.23 embedded-python: True
Windows-10-10.0.26100-SP0 Windows ('64bit', 'WindowsPE')
('Windows', '10', '10.0.26100')
Python 3.11.5
Windows: ('10', '10.0.26100', 'SP0', 'Multiprocessor Free')
Interface language: de
EXE path: C:\Program Files\Calibre2\calibre-parallel.exe
Successfully initialized third party plugins: DeACSM (0, 0, 16) && Gather KFX-ZIP (from KFX Input) (2, 19, 0) && DeDRM (10, 0, 9) && Package KFX (from KFX Input) (2, 19, 0) && EpubCheck (0, 2, 5) && KFX metadata reader (from KFX Input) (2, 19, 0) && From KFX (2, 19, 0) && KFX Input (2, 19, 0) && Set KFX metadata (from KFX Output) (2, 13, 0) && KFX Output (2, 13, 0) && KePub Output (3, 7, 3) && KoboTouchExtended (3, 7, 3) && Obok DeDRM (10, 0, 9)
Traceback (most recent call last):
File "calibre_plugins.epub_check.main", line 199, in ask_user
File "calibre_plugins.epub_check.main", line 123, in get_arch
File "calibre_plugins.epub_check.main", line 114, in jarWrapper
File "subprocess.py", line 1026, in __init__
File "subprocess.py", line 1538, in _execute_child
FileNotFoundError: [WinError 2]

What could be the problem of this? Thanks a lot

Have you followed the install instructions (ie see the first post in this thread).

Java IS required.

[ryder]

I thought java is always preinstalled on windows since win10. Turns out it's not. Thanks it's working now

[JSWolf]

Originally Posted by ryder

I thought java is always preinstalled on windows since win10. Turns out it's not. Thanks it's working now

You need 64-bit Java.

[Karellen]

Originally Posted by ryder

So all I do is this:
Select around 50 ebooks already in epub format (with various bugs), convert them from epub to mobi, delete all epub formats, convert them back from mobi to epub, delete all mobi formats and there will be no more bugs found in the internal bugcheck scan. I read somewhere here that it could help to convert epub to epub again, but there were still some bugs in the scan afterwards. With this method I see none. This method takes less than two minutes for 50 books (depending on your pc power of course). So let me know if this fixes most of the errors in epubcheck as well. That would be great. Thanks again

Sorry to butt in, but this is a terrible thing to do. Your epubs must be a mess code-wise.

[ryder]

Originally Posted by Karellen

Sorry to butt in, but this is a terrible thing to do. Your epubs must be a mess code-wise.

Why is that? I checked them already on my ereader and the format style looks better than before. No problems discovered while reading either. Why do you think this could be code-wise a mess? Maybe I'm missing something here

[Karellen]

Sure, I have no doubt that they will work in an ereader, but I mean the code inside the book I imagine to be excessive and redundant. Have you opened one of the chapters in the editor? How does it look? Maybe post a screenshot?

[JSWolf]

Originally Posted by ryder

Why is that? I checked them already on my ereader and the format style looks better than before. No problems discovered while reading either. Why do you think this could be code-wise a mess? Maybe I'm missing something here

Who knows what code is not buggered because it's something Mobi does not support and it's been changed.

Do not convert your ePub to Mobi. It's a really bad idea. The problem is that Mobi is an old obsolete format and does not support everything that ePub does. So you could be screwing up some of your ePub.

It's best to learn HTML/CSS and edit any ePub you need to make changes to.

[theducks]

Originally Posted by Karellen

Sorry to butt in, but this is a terrible thing to do. Your epubs must be a mess code-wise.

↑ ↑ ↑ ✔

Mobi is an old format and did not support modern stylesheets.

The conversion makes no attempt (because it is very difficult) to do all the tricks, those early publishers did to fake styling in Mobi. So once you pass thru Mobi, you lose all those styling. You might as well convert to plain text

[ryder]

Originally Posted by Karellen

Sure, I have no doubt that they will work in an ereader, but I mean the code inside the book I imagine to be excessive and redundant. Have you opened one of the chapters in the editor? How does it look? Maybe post a screenshot?

Here are screenshots. While there are no bugs found by the internal calibre scan, there are more bugs afterwards found by epubcheck. The thing is I don't like to fiddle around with every epub in sigil. I just like to read books and since calibre does a great job converting epubs I don't see the point in learning stuff about css. But I have noticed some severe battery drain on my Clara 2E and there was a thread on reddit where some guys talked about bad css in epubs could cause a massive drain of the ereader battery somehow. So the question is: are the bugs mentioned by epubcheck huge or just mild to be ignored (index_split for example)? I mean if you look at the outcome of the epub style: there is no wasting space after the chapter number, the formating looks clean. And maybe the battery drain will be gone too. I hadn't time to check about that yet. Do you see a huge problem with this method other than the mess code-wise?

[Karellen]

Originally Posted by ryder

So the question is: are the bugs mentioned by epubcheck huge or just mild to be ignored (index_split for example)?

I don't know. I can't see what the error is. Nor can I really see the code in the body of the chapter.

Originally Posted by ryder

I mean if you look at the outcome of the epub style: there is no wasting space after the chapter number, the formating looks clean. And maybe the battery drain will be gone too. I hadn't time to check about that yet. Do you see a huge problem with this method other than the mess code-wise?

Yes, I see a problem. But I am a bit OCD about tidy and logical code. If it works for you, and you would rather read instead of tidying up formatting, then ignore what I said. Just be mindful, you will eventually come across issues with some books. Things like text that should be in tables, but the tables are missing, maybe ordered or unordered lists have been stripped, or blockquotes no longer showing.