Books disappear after calibre crash - Page 2

kovidgoyal · 10-26-2010, 07:35 PM

Well unless you're willing to write the code for it, no hashing solution is ever going to be implemented.

Giuseppe Chillem · 10-26-2010, 07:53 PM

Quote:

Originally Posted by kovidgoyal

Well unless you're willing to write the code for it, no hashing solution is ever going to be implemented.

I am a REBOL programmer, not a Phyton one. So, I am only in your hands.

Hope Time will change things...

Regards,
Giuseppe Chillemi

kovidgoyal · 10-26-2010, 07:55 PM

Unfortunately, I have too many higher priority things on my hands. I have however updated the PDF libraries in calibre to the latest versions for the enxt release. That may fix your crashes.

DoctorOhh · 10-26-2010, 08:29 PM

Quote:

Originally Posted by Giuseppe Chillem

5) File Hash + Duplicate tag is the complete solution for any problem. When a duplicate file hash is found you have 99.99999999% chance you have a duplicate. Using other tools complicate things too much as new files needs the duplicate library to be rebuild, even for one file added.

Why don't you eliminate all duplicates from your source library (using hash) via third party software prior trying to import them into calibre?

Giuseppe Chillem · 10-26-2010, 09:25 PM

Quote:

Originally Posted by dwanthny

Why don't you eliminate all duplicates from your source library (using hash) via third party software prior trying to import them into calibre?

Last reply for this topic as the developer already told he won't implement this feature.

Obviusly you are right: using a duplicate finder on the original library removes duplicates but it is not the intended scenario.

When you have a crash of the application and the TAG bug has generated a lot of "Unknown" books there is no way to know which books have been added. At the second RUN of the import of the same files the TAG method will fail while The HASH method will remove those books.

Also, if you add many books with strange repeating tags as: "Amber CHM to PDF converter", "Unknown", "Cover.doc", "Untitled" they are shown as duplicates even if they belong to different books. They may or not be duplicates, only hash will say.

Also, if you edit the TAGs of the added books, adding again the same book(s) only checking for the original Hash will show duplicates.

As you see there are many scenarios where this feature is useful.

Regards,
Giuseppe Chillemi

DoctorOhh · 10-26-2010, 09:36 PM

Quote:

Originally Posted by Giuseppe Chillem

When you have a crash of the application and the TAG bug has generated a lot of "Unknown" books there is no way to know which books have been added. At the second RUN of the import of the same files the TAG method will fail while The HASH method will remove those books.

Thanks for explaining.

Kovid did say,

Quote:

Originally Posted by kovidgoyal

I have however updated the PDF libraries in calibre to the latest versions for the next release. That may fix your crashes.

When the next release comes out let us know if the crashes are less frequent. I stay away from PDF files but I would like to know if the latest versions of the PDF libraries make a difference.

LGN · 10-26-2010, 11:33 PM

I thought about Hashes for duplicates too, but then I discovered that when you open an ePUB in Calibre, Calibre actually makes changes to the ePUB itself, therefore chances are the Hash vallues will change. So if you would add a ePub, open it with the Calibre viewer, close it, then add the same ePUB again, Hashes will be different.

CRC is different as it checks the file's integrity, thats mostly used by copying.

Dopedangel · 10-26-2010, 11:44 PM

I don't know much about using calibre with pdf files. But was wondering is their any other data you need apart from the cover if not you could disable pdf metadata import to first import all the files.
And then you being a programmer can code something that would extract the covers from the pdf and save as a cover.jpg in the book folder All such covers would automatically be added to calibre so it should solve your problem.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Kobo Books Disappear	Themus	Kobo Reader	6	10-09-2010 09:19 AM
Preferences causes Calibre to crash	PKFFW	Calibre	2	06-07-2010 06:01 PM
Calibre 0.6.51 Preferences crash	GexMax	Calibre	0	05-02-2010 09:04 AM
Ebook DRM provider goes dark, the books you paid for disappear	orwell2k	News	17	02-04-2010 03:23 PM
Conversions crash calibre	slantybard	Calibre	0	07-20-2009 12:50 AM

10-26-2010, 07:35 PM	#16
kovidgoyal creator of calibre Posts: 45,377 Karma: 27230406 Join Date: Oct 2006 Location: Mumbai, India Device: Various	Well unless you're willing to write the code for it, no hashing solution is ever going to be implemented.

10-26-2010, 07:55 PM	#18
kovidgoyal creator of calibre Posts: 45,377 Karma: 27230406 Join Date: Oct 2006 Location: Mumbai, India Device: Various	Unfortunately, I have too many higher priority things on my hands. I have however updated the PDF libraries in calibre to the latest versions for the enxt release. That may fix your crashes.

10-26-2010, 11:33 PM	#22
LGN Connoisseur Posts: 88 Karma: 60 Join Date: Oct 2010 Device: ProMedia eReader, Sony PRS-600	I thought about Hashes for duplicates too, but then I discovered that when you open an ePUB in Calibre, Calibre actually makes changes to the ePUB itself, therefore chances are the Hash vallues will change. So if you would add a ePub, open it with the Calibre viewer, close it, then add the same ePUB again, Hashes will be different. CRC is different as it checks the file's integrity, thats mostly used by copying.

10-26-2010, 11:44 PM	#23
Dopedangel Wizard Posts: 1,794 Karma: 30548723 Join Date: Dec 2006 Location: Singapore Device: Boyue	I don't know much about using calibre with pdf files. But was wondering is their any other data you need apart from the cover if not you could disable pdf metadata import to first import all the files. And then you being a programmer can code something that would extract the covers from the pdf and save as a cover.jpg in the book folder All such covers would automatically be added to calibre so it should solve your problem.