[Conversion Output] KePub Output Plugin - Page 32

davidfor · 04-15-2021, 02:56 AM

Quote:

Originally Posted by silverbangles

NO NO NO. Do not upgrade BoschBabe. It will all break. I get error messages right left and center now.

Honestly, if you are getting error messages "right left and center", then you are doing something wrong. Calibre version 5 has been stable for a long time. I use it heavily with a lot of plugins. I don't get errors except when I do something stupid. You error above is almost definitely something interfering. You would have gotten the same error with older versions of calibre.

jgoguen · 07-05-2021, 09:06 PM

There's a new version posted which fixes errors when sending massive books. I didn't consider massive books when spinning up threads for each content file, this version uses a process pool to limit how many children there can be.

This version also removes Python 2 compatibility, so you'll need calibre 5.0.0 or above.

redlob · 07-17-2021, 03:46 AM

Hi all, I have this book that I can't convert to kepub. I'm running Calibre version 5.23 and plugin version 3.5.1. Reading the log it's not clear to me what the problem is. Any ideas? Thanks!

Log here

davidfor · 07-17-2021, 08:15 AM

Quote:

Originally Posted by redlob

Hi all, I have this book that I can't convert to kepub. I'm running Calibre version 5.23 and plugin version 3.5.1. Reading the log it's not clear to me what the problem is. Any ideas? Thanks!

Log here

The error is that the book, or at least one of the file in it, already as the Kobo spans in it. An error is thrown when that is processed, and it is not being handled properly. With the previous version of the plugin, this was being ignored. It is erroring on file "Section-0001.xhtml". Can you check if the spans exist in that file?

jgoguen · 07-17-2021, 05:17 PM

New version posted, fixing an issue with exceptions being thrown in subprocesses when functions should have simply returned instead.

davidfor · 07-17-2021, 11:52 PM

Quote:

Originally Posted by jgoguen

New version posted, fixing an issue with exceptions being thrown in subprocesses when functions should have simply returned instead.

I couldn't decide if they should be changed to returns, or throw more specific exceptions so that they could be handled if necessary.

Also, neither this or the previous version of the plugins were picked up by calibre. @chaley commented that he thought it was the indirection of the version number as the plugin checker was not handling that. I don't see any recent changes to that, so I don't know why it is not working now.

redlob · 07-18-2021, 12:52 PM

Quote:

Originally Posted by davidfor

The error is that the book, or at least one of the file in it, already as the Kobo spans in it. An error is thrown when that is processed, and it is not being handled properly. With the previous version of the plugin, this was being ignored. It is erroring on file "Section-0001.xhtml". Can you check if the spans exist in that file?

Yes, spans exist in the file. Thanks.

Minotaur · 08-01-2021, 02:12 AM

Having a really strange problem with converting to kepub. I switched to having my TEMP folder (with environment variables TEMP and TMP) on a RAMDrive with 4GB capacity. Converting to kepub would always fail. Switching back to a TEMP folder on C: resolved the issue.
There's over 2GB of free space on the RAMDrive. Any thoughts?

-EDIT- I've been able to set a different dir for Calibre as an environment variable (CALIBRE_TEMP_DIR) which works. Still unsure why converting to KEPUB was failing when converting to EPUB, LIT etc. was working fine.

davidfor · 08-01-2021, 04:01 AM

Quote:

Originally Posted by Minotaur

Having a really strange problem with converting to kepub. I switched to having my TEMP folder (with environment variables TEMP and TMP) on a RAMDrive with 4GB capacity. Converting to kepub would always fail. Switching back to a TEMP folder on C: resolved the issue.
There's over 2GB of free space on the RAMDrive. Any thoughts?

-EDIT- I've been able to set a different dir for Calibre as an environment variable (CALIBRE_TEMP_DIR) which works. Still unsure why converting to KEPUB was failing when converting to EPUB, LIT etc. was working fine.

Can you post a log from when it failed? I can't think of a reason it would not work, but, the log might tell me.

Minotaur · 08-02-2021, 02:40 AM

Quote:

Originally Posted by davidfor

Can you post a log from when it failed? I can't think of a reason it would not work, but, the log might tell me.

Error log attached. This was converting a kepub to a kepub. Quitting Calibre, changing CALIBRE_TEMP_DIR to a dir on my SSD and retrying the same conversion worked perfectly.

davidfor · 08-02-2021, 09:26 PM

Quote:

Originally Posted by Minotaur

Error log attached. This was converting a kepub to a kepub. Quitting Calibre, changing CALIBRE_TEMP_DIR to a dir on my SSD and retrying the same conversion worked perfectly.

The last few lines of the log are:

Code:

  File "concurrent\futures\thread.py", line 57, in run
  File "calibre_plugins.kepubout.container", line 174, in __flush_cache_impl
  File "calibre\ebooks\oeb\polish\container.py", line 1033, in commit_item
OSError: [Errno 22] Invalid argument: 'R:\\TEMP\\calibre_fqq_t_ua\\1o_yka___epub_container\\text\\part0020.html'

At first glance, there doesn't seem to be a problem with that. It looks like a valid path with escaped backslashes. But, Python tends to like forward slashes. It should be the same no matter the location as the same code path is used. I'll have to experiment. How did you created the RAMDrive? It's been years since I have used one.

Minotaur · 08-02-2021, 10:13 PM

Quote:

Originally Posted by davidfor

The last few lines of the log are:

At first glance, there doesn't seem to be a problem with that. It looks like a valid path with escaped backslashes. But, Python tends to like forward slashes. It should be the same no matter the location as the same code path is used. I'll have to experiment. How did you created the RAMDrive? It's been years since I have used one.

I used the Radeon RAMDisk Config Utility. I've just confirmed that the Ramdisk is formatted to NTFS - thought for a sec that maybe it had formatted to something else.
It's definitely not filing up so it's not erroring due to a disk full issue.

tamarissa · 08-04-2021, 07:46 PM

I have an ePub file that won't convert to kePub at all... it does convert to AZW, MOBI, etc.

I have also tried converting from the other files I was able to create; that often is my workaround from AZW files to kePub (they're often troublesome), but nothing will create a kePub.

I've attached both the details in the failure to convert notice as well as the debug reports for the failed conversions. This is for both starting from the ePub file and from the AZW file.

As far as I am aware, everything is up to date; there are no update alerts in Calibre, but I don't know if any of these required plugins needs manual updates.

jgoguen · 08-14-2021, 01:41 AM

Quote:

Originally Posted by tamarissa

I have an ePub file that won't convert to kePub at all... it does convert to AZW, MOBI, etc.

I have also tried converting from the other files I was able to create; that often is my workaround from AZW files to kePub (they're often troublesome), but nothing will create a kePub.

I've attached both the details in the failure to convert notice as well as the debug reports for the failed conversions. This is for both starting from the ePub file and from the AZW file.

As far as I am aware, everything is up to date; there are no update alerts in Calibre, but I don't know if any of these required plugins needs manual updates.

The error message you posted says there's an unknown field "#kobo". I'm not sure where you may have set that or may be referencing it, maybe take a look through the plugin settings.

jgoguen · 08-14-2021, 01:48 AM

New version 3.5.3 posted, uses calibre's smartyPants() to smarten punctuation instead of stupidly regexing the HTML

08-01-2021, 02:12 AM	#473
Minotaur Enthusiast Posts: 40 Karma: 3798 Join Date: Nov 2014 Location: Adelaide, Australia Device: Kobo Forma	Having a really strange problem with converting to kepub. I switched to having my TEMP folder (with environment variables TEMP and TMP) on a RAMDrive with 4GB capacity. Converting to kepub would always fail. Switching back to a TEMP folder on C: resolved the issue. There's over 2GB of free space on the RAMDrive. Any thoughts? -EDIT- I've been able to set a different dir for Calibre as an environment variable (CALIBRE_TEMP_DIR) which works. Still unsure why converting to KEPUB was failing when converting to EPUB, LIT etc. was working fine. Last edited by Minotaur; 08-01-2021 at 02:18 AM. Reason: Workaround found

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
catalogue builder output columns are not in the same order in the output	KWhytte	Library Management	5	12-04-2012 02:03 AM
Plugin not customizable: Plugin: HTML Output does not need customization	flyingfoxlee	Conversion	2	02-24-2012 02:24 AM
Plugin which uses net as input and output	medve	Development	0	12-04-2011 03:20 PM
Creat new Output plugin (LaTeX and/or txt2tags)	Lebewesen	Development	3	11-22-2011 11:10 AM
AZW Conversion Output Plugin	tylau0	Plugins	59	08-14-2011 10:36 PM

07-05-2021, 09:06 PM	#467
jgoguen Generally Awesome Person Posts: 1,061 Karma: 2178845 Join Date: Jan 2013 Location: /dev/kmem Device: Kobo Clara HD, Kindle Oasis	There's a new version posted which fixes errors when sending massive books. I didn't consider massive books when spinning up threads for each content file, this version uses a process pool to limit how many children there can be. This version also removes Python 2 compatibility, so you'll need calibre 5.0.0 or above.

07-17-2021, 03:46 AM	#468
redlob Connoisseur Posts: 61 Karma: 10 Join Date: Jul 2010 Location: The Netherlands Device: Kobo Glo HD, Kobo Aura H2O edition 2, Kobo Libra 2	Hi all, I have this book that I can't convert to kepub. I'm running Calibre version 5.23 and plugin version 3.5.1. Reading the log it's not clear to me what the problem is. Any ideas? Thanks! Log here

07-17-2021, 05:17 PM	#470
jgoguen Generally Awesome Person Posts: 1,061 Karma: 2178845 Join Date: Jan 2013 Location: /dev/kmem Device: Kobo Clara HD, Kindle Oasis	New version posted, fixing an issue with exceptions being thrown in subprocesses when functions should have simply returned instead.

08-14-2021, 01:48 AM	#480
jgoguen Generally Awesome Person Posts: 1,061 Karma: 2178845 Join Date: Jan 2013 Location: /dev/kmem Device: Kobo Clara HD, Kindle Oasis	New version 3.5.3 posted, uses calibre's smartyPants() to smarten punctuation instead of stupidly regexing the HTML