Calibre hangs on importing books

[KaintukeeBob]

I downloaded books from Baen (CDs from the Fifth Imperium) - they are structured one book to a folder, all collected in a root folder. The book folders contain (DRM-free) .lit, .epub, .rtf, etc files, some images, and .htm files (for reading the books in a browser).

When I try to import these into calibre (latest release - 0.7.7) for conversion/upload to my Motorola Droid (using FBReader) it hangs. It will scan the root directory, then apparantly jumps to the last directory (I downloaded CD22 - Mission of Honor, and it hangs on 'Worlds of Honor', the final directory if sorted alphabetically).

When I tried removing the Worlds of Honor folder, it began hanging on the next folder, 'We Few'.

Steps to reproduce:
1) Download the Zip file here
2) Unzip the file to a folder
3) Open calibre (v 0.7.7)
4) Click the arrow beside 'Add books', select 'Add books from directories, including sub-directories (one book per directory)'
5) Choose the folder you unzipped the zip file into
6) Watch Calibre hang on 'Worlds of Honor'.

Can anyone help me figure out what's going on?

[kovidgoyal]

Run calibre as

calibre-debug -g

from a comamnd prompt and post any output you see

[KaintukeeBob]

UI Displays:
---------------------------
ERROR: Adding failed
---------------------------
The add books process seems to have hung. Try restarting calibre and adding the books in smaller increments, until you find the problem book.
---------------------------
OK Copied
---------------------------


CMD box displays:
Starting up...
link hasn't been detected!
link hasn't been detected!
Using proxies: {'http': '127.0.0.1:8888', 'https': '127.0.0.1:8888'}
Started up in 2.31200003624
Job: 1 Read metadata (0 of 1) finished
Exception in thread Thread-13:
Traceback (most recent call last):
File "threading.py", line 525, in __bootstrap_inner
File "site-packages\calibre\ebooks\metadata\worker.py", line 142, in run
File "site-packages\calibre\__init__.py", line 151, in prints
IOError: [Errno 12] Not enough space

The 'proxies' line is due to the fact that I was running Fiddler when I started this.

[itimpi]

That last .ine of Not enough Space looks a little ominous.

[KaintukeeBob]

I agree, though I'm not sure what it could reference: the computer has ~25 gigs free space, and at least 2 gigs of RAM.

edit: 3.25 GB RAM, Intel core 2 CPU T7600 (2.33GHz) dual-core processor.

[Starson17]

Originally Posted by KaintukeeBob

Steps to reproduce:
1) Download the Zip file

That zip file is 300+ MB in size. It's hard to tell what it is, but it looks like a few dozen books, associated graphics all linked together as a massive html. Perhaps "Not enough space" is due to the size of what you're adding?

[KaintukeeBob]

You can 'View' the CD from that site as well, which shows you the index. It's about 41 books, each in a folder that contains approx 8 formats, an HTML copy (split 1 chapter per HTML page), ~300-600KB of image files (for the HTML) and a single .jpg cover (varies in size).

The individual directories range from 5-30 MB, everything included.

I had initially suspected something similar, Starson, but the initial directory that hangs (Worlds of Honor) is one of the smaller ones (few images, only 5 'chapters', 23 files total) at 5.94 MB.

At the root of the CD's directory is some HTML that indexes all of the books, providing links to each subdirectory's index page (which contains links to the HTML version of the books and download links to each version).

I don't know if that's the cause of the issue, since I see the same thing when I drill down and choose just 'Worlds of Honor' from the 'Select Directory' prompt.

[Starson17]

It wouldn't surprise me if all the directories and all the books are crosslinked via the html. No matter where you start, it would try to build one humongous html book. Try adding the books by dragging into the Calibre GUI all of the books in the format you want. I doubt if the books themselves will cause any problem. It's most likely the html that surrounds them.

[theducks]

Originally Posted by Starson17

It wouldn't surprise me if all the directories and all the books are crosslinked via the html. No matter where you start, it would try to build one humongous html book. Try adding the books by dragging into the Calibre GUI all of the books in the format you want. I doubt if the books themselves will cause any problem. It's most likely the html that surrounds them.

You are mostly (95%) correct.

Baen CD's have a Front door page, so the links are one way only.

Remove the non-folder content in the root of the extraction should mostly take care of the "Links to everything else"

Other possible folders to "blow away": "images","screensaver","audio", "game"
Any folder in lower case Book folders are "mixed case"

I only extracted the format (file types) I wanted, when I did that.

[kovidgoyal]

Go to preferences->plugins->File type plugins and disable the html2zip plugin temporarily for this import

[KaintukeeBob]

Thanks! That worked like a charm.

[vitalichka]

I have been experiencing the same problem on a very similar computer set-up but running the current version of calibre. What's odd in my case is that the particular folder I am adding does not contain any HTML files, only .pdf .lit and .epub but am seeing the same hang up issues.

[kovidgoyal]

disable the pdf metadata reader plugin.

[vitalichka]

when i disable html to zip, it seems to work