[GUI Plugin] WordDumb

[fetib]

Hi!

I got this error message whenever I would like to generate word-wise, what should I do in this case?

calibre, version 7.26.0 (darwin, embedded-python: True)
Tonnerre de Brest!: An error occurred, please copy error message then report bug at GitHub.

Starting job: Generating Word Wise for Alien Clay
Job: "Generating Word Wise for Alien Clay" failed with error:
Traceback (most recent call last):
File "calibre/gui2/threaded_jobs.py", line 85, in start_work
File "calibre_plugins.worddumb.parse_job", line 177, in do_job
install_deps(data.spacy_model, notifications)
File "calibre_plugins.worddumb.deps", line 42, in install_deps
pip_install("spacy", dep_versions["spacy"], notif=notif)
File "calibre_plugins.worddumb.deps", line 134, in pip_install
run_subprocess(args)
File "calibre_plugins.worddumb.utils", line 47, in run_subprocess
return subprocess.run(
^^^^^^^^^^^^^^^
File "subprocess.py", line 571, in run
subprocess.CalledProcessError: Command '['/opt/homebrew/bin/python3', '-m', 'pip', '--disable-pip-version-check', 'install', '-U', '-t', '/Users/*/Library/Preferences/calibre/plugins/worddumb-libs-py3.13', '--no-user', '--no-cache-dir', 'spacy==3.8.2']' returned non-zero exit status 1.

[xxyzz]

Originally Posted by beecom

One strange thing is that the same EPUB file looks different when opened on the Kindle Paperwhite and the Kindle Scribe, even though they are running the same version of KOReader (2024.11-203).

Sorry for the late reply...

I have fixed the double lines in the first screenshot: https://github.com/xxyzz/WordDumb/co...039790743ff7ce

no idea what's going on in the second pic

I also reduce the css line height value to 2, if the underline is still a problem when changing line space I'd prefer to remove it.

[esso123]

I'd like to kindly request the addition of Arabic language support in wordDumb plugin. Thanks.

[xxyzz]

Do you mean Arabic definition for other language or the other way around?

Arabic Wiktionary is not parsed so Arabic definition data could only be obtained from translation data in other Wiktionary like English Wiktionary(English to Arabic).

[xxyzz]

WSD feature is added for en-en Wiktionary data, please download at here: https://github.com/xxyzz/WordDumb/re...g/v3.34.0-beta

Word sense embeddings are created using a simplified LMMS code, and short definitions are summarized by LLM.

Nested senses are removed, koreader users can get the full Wiktionary definitions by downloading stardict files from here: https://github.com/xxyzz/wiktionary_stardict/releases

[kvothe]

Do you know if there is any way that we could mannually adjust the size of word wise hints? I notice that once my device (Kindle Oasis 3) updated to 5.18.2, the font size for word wise hints became much smaller than previous firmaware, which is extremely annoying.

[CamilleScholtz]

Is it possible to save the used settings (for example Wikimedia api url) per book? Now I manually have to remember and re-set it per book it seems.

[xxyzz]

Originally Posted by CamilleScholtz

Is it possible to save the used settings (for example Wikimedia api url) per book? Now I manually have to remember and re-set it per book it seems.

Feature added to the master branch, you can download the test version at here: https://github.com/xxyzz/WordDumb/ac...ns/17490314475

[trekky0623]

After a recent update, I was getting this error when attempting to generate WordWise/X-ray:

/usr/bin/python3.12 -m pip --disable-pip-version-check install -U -t /home/USERNAME/.config/calibre/plugins/worddumb-libs-py3.12 --no-user accelerate==1.10.1

...

ERROR: Could not install packages due to an OSError: [Errno 28] No space left on device

I was able to resolve this by making a temporary folder, ~/tmpdir, and then adding TMPDIR=~/tmpdir before the install command above.

[rqdoom]

I have a relatively dumb question, I tried using Worddumb in the past but never got to make it work as I wanted it to. However, I remember selecting the "Customize Kindle Word Wise" and "Customize EPUB Wiktionary" options in the settings, would this change any default or normal file on the kindle?

Say I wanted to restore my kindle to it's default Wordwise, should I replace any file back to it's original one? Is there anywhere that worddumb would save the original files as backup in such a case?

[xxyzz]

Change word wise language to English in Kindle to use the original definition from Amazon. If the word wise data is created from the plugin using Wiktionary data you also need to regenerate word wise file with that option disabled.

[Rishan]

Hi,
I’m getting this error when using WordDumb on Calibre:

Starting job: Generating Word Wise and X-Ray for The Secret of Secrets: A Novel
Job: "Generating Word Wise and X-Ray for The Secret of Secrets: A Novel" failed with error:
Traceback (most recent call last):
File "calibre/gui2/threaded_jobs.py", line 85, in start_work
File "calibre_plugins.worddumb.parse_job", line 186, in do_job
install_deps("wsd", notifications)
File "calibre_plugins.worddumb.deps", line 58, in install_deps
pip_install(p, dep_versions[p], notif=notif)
File "calibre_plugins.worddumb.deps", line 155, in pip_install
run_subprocess(args)
File "calibre_plugins.worddumb.utils", line 49, in run_subprocess
return subprocess.run(
^^^^^^^^^^^^^^^
File "subprocess.py", line 571, in run
subprocess.CalledProcessError: Command '['python3', '-m', 'pip', '--disable-pip-version-check', 'install', '-U', '-t', '/home/linux/.config/calibre/plugins/worddumb-libs-py3.13', '--no-user', 'accelerate==1.10.1']' returned non-zero exit status 1.

Called with args: (ParseJobData(book_id=20, book_path='/home/linux/Calibre Library/Dan Brown/The Secret of Secrets_ A Novel (20)/The Secret of Secrets_ A Novel - Dan Brown.azw3', mi=<calibre.ebooks.metadata.book.base.Metadata object at 0x7f8ddeb6bad0>, book_fmt='AZW3', book_lang='en', useragent='WordDumb/3.35.1 (https://github.com/xxyzz/WordDumb)', plugin_path=PosixPath('/home/linux/.config/calibre/plugins/WordDumb.zip'), spacy_model='en_core_web_lg', create_ww=True, create_x=True, asin='BBCFQBSA5Q', acr='The_Secret_of_Secrets__A_Novel', revision='4acb6cc4', kfx_json=None, mobi_html=b'<?xml version="1.0" encoding="UTF-8"?>\n<html xmlns="http://www.w3.org/1999/xhtml" lang="en-US">\n <head>\n <title>Also by Dan Brown, The Secret of Secrets</title>\n <meta content="urn:uuid:36ff629c-76dc-4410-823b-ae83dbf42b70" name="Adept.expected.resource"/>\n <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>\n <link rel="stylesheet" type="text/css" href="kindle:flow:0002?mime=text/css"/>\n<link rel="stylesheet" type="text/css" href="kindle:flow:0001?mime=text/css"/>\n</head>\n <body lang="en-US" class="calibre" aid="0">\n\t\t<a id="d1-d2s0" class="calibre1" aid="1"></a>\n\t\t<div class="page_top_padding" aid="2">\n\t\t\t<h1 class="para-acmh-ac-pg" aid="3"><span type="pagebreak" id="page_iii" role="doc-pagebreak" title="iii" class="calibre2" aid="4"></span><span class="fv-allsmallcaps" aid="5">Also by Dan Brown</span></h1>\n\t\t\t<h2 class="para-ach" aid="6">Featuring Robert Langdon</h2>\n\t\t\t<p class="para-acl" aid="7">Angels &amp;
... bla bla bla
<link rel="stylesheet" type="text/css" href="kindle:flow:0003?mime=text/css"/>\n</head>\n <body lang="en-US" class="calibre8" aid="4GULG0">\n\t\t<div class="center" aid="4GULG1">\n\t\t\t<p class="calibre9" aid="4GULG2"><img alt="Penguin Random House publisher logo." class="image_h" src="kindle:embed:0005?mime=image/jpeg"/></p>\n\t\t\t<h1 class="calibre10" aid="4GULG3">\n\t\t\t\tWhat\xe2\x80\x99s next on<br class="calibre11"/> your reading list?</h1>\n\t\t\t<p class="calibre9" aid="4GULG4"><a href="http://links.penguinrandomhouse.com/type/prhebooklanding/isbn/9780385546928/display/1" class="pcalibre2 pcalibre4 pcalibre3 pcalibre1 calibre12" aid="4GULG5">Discover your next<br class="calibre11"/> great read!</a></p>\n\t\t\t<div aria-hidden="true" class="rule" aid="4GULG6">\xc2\xa0</div>\n\t\t\t<p class="x-small" aid="4GULG7">Get personalized book picks and up-to-date news about this author.</p>\n\t\t\t<p class="calibre9" aid="4GULG8"><a href="http://links.penguinrandomhouse.com/type/prhebooklanding/isbn/9780385546928/display/2" class="pcalibre2 pcalibre4 pcalibre3 pcalibre1 calibre12" aid="4GULG9">Sign up now.</a></p>\n\t\t</div>\n\t\t<p aria-hidden="true" class="identifier" aid="4GULGA">_153075629_</p>\n\t</body>\n</html>\n', mobi_codec='utf-8', book_settings={'mediawiki_api': ''}),) {'notifications': <queue.Queue object at 0x7f8ddeb55c90>, 'abort': <threading.Event at 0x7f8ddead2350: unset>, 'log': <calibre.utils.logging.GUILog object at 0x7f8ddeb55f50>}
ERROR: Could not install packages due to an OSError: [Errno 28] No space left on device

My system and Kindle both have plenty of free space.
Details:
• OS: Manjaro Linux (latest)
• Calibre: latest version
• Python: 3.13.7
• WordDumb: latest release
I’ve tried reinstalling the plugin, restarting Calibre, reconnecting the Kindle, and confirmed all partitions have free space (df -h). Still the same error.
Please suggest a solution for this problem
Thanks!

[xor_]

Hello. I'm trying to use a customized worddumb-custom-x-ray.json file for a book that has been posted on this site.

I copy that file to that book's folder in my local calibre library. WordDumb seems to update and upload the epub ebook just fine. But then when I hit a link in KOReader, I see errors like this:

Invalid or external link: #_doc_fragment_91_2

And in calibre when I select Customize X-Ray for that book, it brings up the table, but in the foreground there's an unhandled exception error box:

calibre 8.12  embedded-python: True
Traceback (most recent call last):
  File "calibre_plugins.worddumb.custom_x_ray", line 154, in data
    value = self.x_ray_data[row][column]
            ~~~~~~~~~~~~~~~~~~~~^^^^^^^^
IndexError: list index out of range

I don't see any obvious errors in that json file, and in that thread noone seemed to have a problem, so I'm guessing I'm doing something wrong?

[xxyzz]

Originally Posted by Rishan

all partitions have free space (df -h). Still the same error.

Maybe /tmp space is not enough. Should be fixed by https://github.com/xxyzz/WordDumb/co...6ccd5b4af6f41e, you can download at here: https://github.com/xxyzz/WordDumb/ac...ns/18308977814

[xxyzz]

Originally Posted by xor_

calibre 8.12  embedded-python: True
Traceback (most recent call last):
  File "calibre_plugins.worddumb.custom_x_ray", line 154, in data
    value = self.x_ray_data[row][column]
            ~~~~~~~~~~~~~~~~~~~~^^^^^^^^
IndexError: list index out of range

That JSON file probably was created for an very old version of the plugin. Each x-ray data should have six elements but that file has five.