[GUI Plugin] Count Pages

[edmun]

Hi Guys

I am sorry to trouble you but this will be probably the first time I am having an issues with CountPages plugin.

I have recently added several books to the newest version of Calibre (2.51), selected 5 books, clicked on Count Pages and accepted to update the columns. 4 books have been updated but one hasn't. I have tried several times but it does not update one book (source - mobi/epub):

Count Page/Word Statistics
-------------------------------
Logfile for book ID 563 (Przewieszenie)
563
Traceback (most recent call last):
  File "calibre_plugins.count_pages.jobs", line 141, in do_statistics_for_book
  File "calibre_plugins.count_pages.statistics", line 53, in get_page_count
  File "calibre_plugins.count_pages.statistics", line 88, in _open_epub_file
  File "site-packages\calibre\ebooks\oeb\iterator\book.py", line 140, in __enter__
  File "site-packages\calibre\ebooks\oeb\iterator\book.py", line 84, in run_extract_book
  File "site-packages\calibre\utils\ipc\simple_worker.py", line 220, in fork_job
  File "site-packages\calibre\utils\ipc\simple_worker.py", line 116, in communicate
WorkerError: Worker failed

edit: no way - found it while I was writing the post. What has brought my attention is that countpages started with counting with epub file (I have purchased book in both formats). Probably epub file is corrupted but mobi is fine.

[mariaclaudia]

Good afternoon. Sorry if this question is repeated, but English is not my forte, so if somebody asked the same before, it's sure I have not seen. Is it possible that the plugin run automatically each time you add one or more books?

Best regards from Lima, Perú
Blessings

[theducks]

Originally Posted by mariaclaudia

Good afternoon. Sorry if this question is repeated, but English is not my forte, so if somebody asked the same before, it's sure I have not seen. Is it possible that the plugin run automatically each time you add one or more books?

Best regards from Lima, Perú
Blessings

No

most Plugins are not able to be inside the Add Books pipeline. to much data has yet to be inserted into the Library.
Cart, but the horse has yet to be captured


You CAN TAG books during the Add.
Preferences: Adding Books: Tag to apply... <your tag>

[whismerhill]

what's the best algorithm for a kindle 4 non touch ?

reason I'm asking is because I was fine just using "apnx accurate"
but I found something strange.
*** a book with 106469 words & 392 pages
calibre viewer EPUB says "641"
& MOBI "622" pages

*** another book with 97 947 words & 1241 pages
calibre viewer EPUB says "2184" thingy (whatever that number is)
& for mobi "569" pages


obviously one of the two is wrong
I don't think the 2nd book has really 1241 pages that would really be a huge pocket book & I don't think it is.

so yeah I like this plugin, but the variance seems a bit too much, any way to make it more "sensible" maybe ?

thanks for reading.


EDIT: nevermind Adobe algorithm seems to be more consistent with the amount of real content sometimes dividing the number of pages by 3
new number of pages for the 1241 pages book is 486

[Pyjam]

Hello there

I'm lost!

I would like an accurate measure to compare book lengths. I don't expect this measure to match the actual number of pages. Which of the four methods should I choose?

Thanks for your help.

[theducks]

Originally Posted by Pyjam

Hello there

I'm lost!

I would like an accurate measure to compare book lengths. I don't expect this measure to match the actual number of pages. Which of the four methods should I choose?

Thanks for your help.

Consistency is probably more important.
Use the same method for all so that comparisons work

I use the ADE (adobe), since that is also my (and many others) reader device base
IIRC one is is Paragraphs

The Calibre viewer... fine for Calibre only uses

[Dopedangel]

Originally Posted by Pyjam

Hello there

I'm lost!

I would like an accurate measure to compare book lengths. I don't expect this measure to match the actual number of pages. Which of the four methods should I choose?

Thanks for your help.

If you just want numbers for comparison then I would recommend using word count instead of pages.

Sent from my SM-G925I

[Pyjam]

After many tests, I think I'm satisfied with the custom 1800 chars per page method. The result is often close to the mass market edition, or to the number of pages given by my Kindle.

Hyphenation has a weird effect on the result. After I hypenated a 1860-page long book it has lost 120 pages. OTOH a 1000-pages long book has gained almost 50 pages.

[JSWolf]

You should count the pages before adding in soft hyphens.

[Pyjam]

Indeed.

Most of the time, an evaluation with 1765 chars per page on an EPUB w/o hephen is very close to the number of pages of an AZW3 after hyphenation, with an accuracy of 98%. Really strange.

[excaliber]

Hi all,

When I use this plugin I'm getting some times following error:

calibre, version 2.60.0
ERROR: Unhandled exception: <b>TypeError</b>bject of type 'NoneType' has no len()

calibre 2.60 embedded-python: True is64bit: True
Linux-4.4.0-22-generic-x86_64-with-debian-jessie-sid Linux ('64bit', 'ELF')
('Linux', '4.4.0-22-generic', '#40~14.04.1-Ubuntu SMP Fri May 13 17:27:45 UTC 2016')
Python 2.7.9
Linux: ('debian', 'jessie/sid', '')
Successfully initialized third party plugins: Save To Format (0, 0, 6) && Walk Search History (1, 3, 1) && Count Pages (1, 6, 9) && Wiki Reader (2, 1, 0) && Open With (1, 5, 9) && Get DJVU Metadata (1, 2, 0) && Author Book Count Hierarchy (1, 1, 6) && Favourites Menu (1, 0, 4) && Generate Cover (1, 5, 19) && Consolidate All Library Metadata (2, 0, 19) && Search The Internet (1, 7, 6) && Reading List (1, 6, 6) && Find Duplicates (1, 6, 2) && Zotero Metadata Importer (1, 0, 7) && Astro-ph (1, 1, 0) && English Noun Frequency (1, 0, 6) && Import List (1, 1, 5) && Clean Metadata (0, 0, 6) && Library Codes (1, 0, 27) && Job Spy (1, 0, 50) && Manage Series (1, 2, 8) && Google Scholar (0, 1, 1) && Annotations (1, 4, 7) && MultiColumnSearch (1, 0, 64) && User Category (1, 5, 3) && Quick Preferences (1, 4, 1) && Diaps Editing Toolbag (0, 3, 1) && Quality Check (1, 9, 11) && View Manager (1, 3, 2) && Goodreads (1, 1, 11) && Extract ISBN (1, 4, 4) && BOL_DE (0, 7, 2) && Drop Search Results (1, 0, 5) && Author Book Count (2, 1, 4) && Recoll Full Text Search (1, 0, 1) && Get PDF Metadata (1, 2, 0)
Traceback (most recent call last):
File "site-packages/calibre/gui2/__init__.py", line 418, in dispatch
File "calibre_plugins.count_pages.action", line 211, in _get_statistics_completed
TypeError: object of type 'NoneType' has no len()

I run the plugin for some hundred files. It starts normally but when it reaches 100% is throwing out the above error without updating the pages column and I cannot figure out if it's because of a particular file and if so which file is causing it.

Any ideas?

[davidfor]

Originally Posted by excaliber

Hi all,
I run the plugin for some hundred files. It starts normally but when it reaches 100% is throwing out the above error without updating the pages column and I cannot figure out if it's because of a particular file and if so which file is causing it.

Any ideas?

Just to check: you are selecting a hundred or more books and running the count. Then the error happens at the end?

From the code, the only way to get that error is if there was an earlier error. When it happens again, can you open the job window, find the count pages job in the list, select it and press the "Show job details" button. Then copy the text and post post it. Hopefully that will show the original error.

[excaliber]

@davidfor:
Thanks for your response!

Just to check: you are selecting a hundred or more books and running the count. Then the error happens at the end?
Yes.

From the code, the only way to get that error is if there was an earlier error. When it happens again, can you open the job window, find the count pages job in the list, select it and press the "Show job details" button. Then copy the text and post post it. Hopefully that will show the original error.
OK. I'm running another count pages job now. It will take some days to finish...
I think the reason is that I'm running out of space in my partition when the plugin puts files in the /tmp folder. I have now changed the path for it in a new bigger partition. I will report the result when the job has finished.

[davidfor]

Originally Posted by excaliber

@davidfor:
Thanks for your response!

Just to check: you are selecting a hundred or more books and running the count. Then the error happens at the end?
Yes.

From the code, the only way to get that error is if there was an earlier error. When it happens again, can you open the job window, find the count pages job in the list, select it and press the "Show job details" button. Then copy the text and post post it. Hopefully that will show the original error.
OK. I'm running another count pages job now. It will take some days to finish...

Days? Exactly how many books are you counting?

I think the reason is that I'm running out of space in my partition when the plugin puts files in the /tmp folder. I have now changed the path for it in a new bigger partition. I will report the result when the job has finished.

Yes, running out of space could do it. So might running out of memory if you are running that many in one go.

[excaliber]

@davidfor:
Days? Exactly how many books are you counting?
Around 15'000 documents. But most of them are not books. I was throwing into Calibre every document I had and now is a mess...That's why I use the plugin now. It helps me to to sort my documents.