Release: 9.1 [31 Jan, 2026] BUG

fengli · 01-31-2026, 08:32 PM

Release: 9.1 [31 Jan, 2026]

A major bug existed between versions 8.10 and 9.10, causing imported OPML recipes to include a lot of irrelevant content. The versions were messy and didn't crawl as cleanly as before. For example, version 8.10 worked fine, crawling only articles without any extra clutter.

Additionally, there are issues with conversion. The new version encounters errors when converting multiple EPUBs to PDF simultaneously, but version 8.10 does not exhibit this problem.

Please help fix it. Thank you.

windows 11

kovidgoyal · 01-31-2026, 11:07 PM

There are no changes to recipe processing in calibre 9. As for conversion to PDF it will likely be an issue caused by Qt WebEngine being updated and not working well on your system. What is the actual error you get?

fengli · Yesterday, 01:27 AM

Quote:

Originally Posted by kovidgoyal

There are no changes to recipe processing in calibre 9. As for conversion to PDF it will likely be an issue caused by Qt WebEngine being updated and not working well on your system. What is the actual error you get?

<?xml version="1.0" encoding="UTF-8"?>
<opml version="1.0">
<head>
<title>Feed Subscriptions</title>
</head>
<body>
<outline title="Iceshi" text="Iceshi">
<outline text="Editorials, Editorial Opinions, Editorial News, The Hindu Opinion | The Hindu" title="Editorials, Editorial Opinions, Editorial News, The Hindu Opinion | The Hindu" type="rss"
xmlUrl="https://www.thehindu.com/opinion/editorial/feeder/default.rss" htmlUrl="https://www.thehindu.com/opinion/editorial/"/>
<outline text="National Review" title="National Review" type="rss"
xmlUrl="https://www.nationalreview.com/feed/" htmlUrl="https://www.nationalreview.com"/>
<outline text="The Atlantic" title="The Atlantic" type="rss"
xmlUrl="https://feeds.feedburner.com/TheAtlantic" htmlUrl="https://www.theatlantic.com/"/>
<outline text="Articles on Smashing Magazine Ã¢â‚¬â€ For Web Designers And Developers" title="Articles on Smashing Magazine Ã¢â‚¬â€ For Web Designers And Developers" type="rss"
xmlUrl="https://www.smashingmagazine.com/feed/" htmlUrl="https://www.smashingmagazine.com/"/>
<outline text="CBC - CBC | Top Stories News" title="CBC - CBC | Top Stories News" type="rss"
xmlUrl="https://rss.cbc.ca/lineup/topstories.xml" htmlUrl="https://www.cbc.ca/news/?cmp=rss"/>
</outline>
</body>
</opml>

Hey boss, I've provided a test OPML file. Could you please give it a try? Thanks! A lot of the scraped content is messy and disorganized.

Click image for larger version

Name: 20260201132425007.png
Views: 9
Size: 608.2 KB
ID: 220653

Converting EPUB to PDF fails when processing multiple files simultaneously, such as five at once, but works fine when converting a single file. Version 8.10 performs exceptionally well, with very few failures regardless of the number processed, even seven files. Please investigate whether this issue stems from the recent Qt WebEngine update in the new version.

fengli · Yesterday, 01:36 AM

Version 8.10 excels at both epub Conversion PDF and OPML extraction, delivering clean articles without any extraneous elements.

fengli · Yesterday, 01:57 AM

[ATTACH]

Click image for larger version

Name: 2026-02-01 135452.png
Views: 16
Size: 534.4 KB
ID: 220655

[/ATTACH]

Numerous disorganized elements, as shown in the figure, likely resulted from failure to remove other elements after scraping, leading to unclean and untidy article content.

fengli · Yesterday, 02:00 AM

I suspect that the updated version of Qt WebEngine has a bug when batch processing EPUB to PDF.

fengli · Yesterday, 02:05 AM

Versions from 8.10 through 9.1 have consistently exhibited these two issues. I tested each update, but ultimately had to revert to version 8.10—though it lacks new features.

1. OPML article elements are captured in a messy, disorganized manner.
2. Batch conversion of EPUB to PDF frequently fails, while individual conversions work fine.

Please thoroughly investigate these issues as they are highly disruptive. Thank you very much.

kovidgoyal · Yesterday, 02:09 AM

batch conversion to PDF is working fine for me. Again, what actual error do you get and on what system? As for your OPML issues its likely a lxml update breaking readability. Neither of these have anything to do with calibre 9.

kovidgoyal · Yesterday, 03:22 AM

And this fixes the OPML cleanup issue. As suspected it was because of an lxml change and happened in version 8.11 not version 9

https://github.com/kovidgoyal/calibr...f0ece7bd379a20

Comfy.n · Yesterday, 04:00 AM

calibre < 8.11

Click image for larger version

Name: 2026-02-01 04_32_08-Iceshi [dom, 01 fev 2026] [EPUB] — E-book viewer.png
Views: 22
Size: 835.5 KB
ID: 220660

calibre current

Click image for larger version

Name: 2026-02-01 04_33_36-9Iceshi [dom, 01 fev 2026] [EPUB] — E-book viewer.png
Views: 25
Size: 870.8 KB
ID: 220659

calibre < 8.11

Click image for larger version

Name: 2026-02-01 04_34_54-Iceshi [dom, 01 fev 2026] [EPUB] — E-book viewer.png
Views: 21
Size: 686.8 KB
ID: 220658

calibre current

Click image for larger version

Name: 2026-02-01 04_35_00-9Iceshi [dom, 01 fev 2026] [EPUB] — E-book viewer.png
Views: 17
Size: 654.6 KB
ID: 220657

Great find.

What I found interesting is that calibre < 8.11 didn't fetch the first image of the example article: https://www.cbc.ca/radio/asithappens...068855?cmp=rss

Thank you for the fix, Kovid.

JSWolf · Yesterday, 06:21 AM

Quote:

Originally Posted by fengli

Versions from 8.10 through 9.1 have consistently exhibited these two issues. I tested each update, but ultimately had to revert to version 8.10—though it lacks new features.

1. OPML article elements are captured in a messy, disorganized manner.
2. Batch conversion of EPUB to PDF frequently fails, while individual conversions work fine.

Please thoroughly investigate these issues as they are highly disruptive. Thank you very much.

Update to 8.16.2 as that's the last version 8 release.

fengli · Yesterday, 10:19 AM

Quote:

Originally Posted by kovidgoyal

And this fixes the OPML cleanup issue. As suspected it was because of an lxml change and happened in version 8.11 not version 9

https://github.com/kovidgoyal/calibr...f0ece7bd379a20

Yes, you have found the problem. The problem I reported is that there is no problem with version 8.10, but the versions after 8.10, namely 8.11-9.1, have these two problems. I tried to update the version every time, but nothing worked, and then returned to the old version.

fengli · Yesterday, 10:23 AM

Is the problem not that the content cannot be captured。 It is that the captured content has many cluttered elements, such as various titles.

fengli · Yesterday, 10:27 AM

Quote:

Originally Posted by Comfy.n

calibre < 8.11
Attachment 220660

calibre current
Attachment 220659

calibre < 8.11
Attachment 220658

calibre current
Attachment 220657

Great find.

What I found interesting is that calibre < 8.11 didn't fetch the first image of the example article: https://www.cbc.ca/radio/asithappens...068855?cmp=rss

Thank you for the fix, Kovid.

Yes, your screenshot has already reflected the problem, that is, the captured content is very clean in version 8.10 and has no messy elements, but after 8.11, there are many messy elements.

Comfy.n · Yesterday, 10:29 AM

Quote:

Originally Posted by fengli

cluttered elements

yes that's what I tried to show in my first two comparison screenshots, the other two show that the first image of the article wasnt fetched by calibre prior to 8.11.

01-31-2026, 08:32 PM	#1
fengli Zealot Posts: 111 Karma: 39846 Join Date: Aug 2022 Device: PC	Release: 9.1 [31 Jan, 2026] BUG Release: 9.1 [31 Jan, 2026] A major bug existed between versions 8.10 and 9.10, causing imported OPML recipes to include a lot of irrelevant content. The versions were messy and didn't crawl as cleanly as before. For example, version 8.10 worked fine, crawling only articles without any extra clutter. Additionally, there are issues with conversion. The new version encounters errors when converting multiple EPUBs to PDF simultaneously, but version 8.10 does not exhibit this problem. Please help fix it. Thank you. windows 11

Yesterday, 01:36 AM	#4
fengli Zealot Posts: 111 Karma: 39846 Join Date: Aug 2022 Device: PC	Version 8.10 excels at both epub Conversion PDF and OPML extraction, delivering clean articles without any extraneous elements. Last edited by fengli; Yesterday at 01:42 AM.

Yesterday, 01:57 AM	#5
fengli Zealot Posts: 111 Karma: 39846 Join Date: Aug 2022 Device: PC	[ATTACH][/ATTACH] Numerous disorganized elements, as shown in the figure, likely resulted from failure to remove other elements after scraping, leading to unclean and untidy article content. Attached Thumbnails

Yesterday, 04:00 AM	#10
Comfy.n want to learn what I want Posts: 1,850 Karma: 7945227 Join Date: Sep 2020 Device: none	calibre < 8.11 calibre current calibre < 8.11 calibre current Great find. What I found interesting is that calibre < 8.11 didn't fetch the first image of the example article: https://www.cbc.ca/radio/asithappens...068855?cmp=rss Thank you for the fix, Kovid.

Yesterday, 10:23 AM	#13
fengli Zealot Posts: 111 Karma: 39846 Join Date: Aug 2022 Device: PC	Is the problem not that the content cannot be captured。 It is that the captured content has many cluttered elements, such as various titles. Last edited by fengli; Yesterday at 10:33 AM.

01-31-2026, 11:07 PM	#2
kovidgoyal creator of calibre Posts: 45,963 Karma: 29579516 Join Date: Oct 2006 Location: Mumbai, India Device: Various	There are no changes to recipe processing in calibre 9. As for conversion to PDF it will likely be an issue caused by Qt WebEngine being updated and not working well on your system. What is the actual error you get?

Yesterday, 02:00 AM	#6
fengli Zealot Posts: 111 Karma: 39846 Join Date: Aug 2022 Device: PC	I suspect that the updated version of Qt WebEngine has a bug when batch processing EPUB to PDF.

Yesterday, 02:05 AM	#7
fengli Zealot Posts: 111 Karma: 39846 Join Date: Aug 2022 Device: PC	Versions from 8.10 through 9.1 have consistently exhibited these two issues. I tested each update, but ultimately had to revert to version 8.10—though it lacks new features. 1. OPML article elements are captured in a messy, disorganized manner. 2. Batch conversion of EPUB to PDF frequently fails, while individual conversions work fine. Please thoroughly investigate these issues as they are highly disruptive. Thank you very much.

Yesterday, 02:09 AM	#8
kovidgoyal creator of calibre Posts: 45,963 Karma: 29579516 Join Date: Oct 2006 Location: Mumbai, India Device: Various	batch conversion to PDF is working fine for me. Again, what actual error do you get and on what system? As for your OPML issues its likely a lxml update breaking readability. Neither of these have anything to do with calibre 9.

Yesterday, 03:22 AM	#9
kovidgoyal creator of calibre Posts: 45,963 Karma: 29579516 Join Date: Oct 2006 Location: Mumbai, India Device: Various	And this fixes the OPML cleanup issue. As suspected it was because of an lxml change and happened in version 8.11 not version 9 https://github.com/kovidgoyal/calibr...f0ece7bd379a20

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Short Fiction Various: Past Masters 248, v1, 27 Jan 2026	Pulpmeister	Other Books	0	01-26-2026 09:01 PM
Short Fiction Various: Past Masters 247, v1, 18 Jan 2026	Pulpmeister	ePub Books	0	01-17-2026 07:47 PM
Short Fiction Various: Past Masters 246, v1, 5 Jan 2026	Pulpmeister	Other Books	0	01-05-2026 02:21 AM
Short Fiction Various: Past Masters 246, v1, 5 Jan 2026	Pulpmeister	ePub Books	0	01-05-2026 02:19 AM
Horror Various Authors: A Book of Ghosts — 44. v1. 04 Jan. 2026	GrannyGrump	Kindle Books	0	01-05-2026 12:14 AM

Advert

Advert