Endless Conversion lit->epub

[slantybard]

Hmm, not sure if this has been answered yet since a search did not provide good results. I have noticed that some .lit books seem to take a long time to convert or actually never convert. I am using 0.5.7 on Mac OS X and the .lit files are not DRM'd.

From the console app, it seems that the jobs are actually not even started. For example, I started a conversion and waited over 30 minutes with no change. When I look in the console, it seems that nothing has happened. Then I stop the conversion, and the console shows this:

20/04/09 8:23:15 PM SyncServer[8106] SyncServer: Truth vacuumed. Next vacuum date 2009-05-04 20:23:15 -0700
20/04/09 8:34:18 PM [0x0-0x410410].net.kovidgoyal.calibre[8072] Job: Convert book: *******
20/04/09 8:34:18 PM [0x0-0x410410].net.kovidgoyal.calibre[8072] **JobKilled**: Job stopped by user
20/04/09 8:34:18 PM [0x0-0x410410].net.kovidgoyal.calibre[8072] Exploding LIT file: /var/folders/cn/cncd5ho2GXOwQFm1q4K7l++++TI/-Tmp-/calibre_0.5.7_m4XZP8.lit
20/04/09 8:34:18 PM [0x0-0x410410].net.kovidgoyal.calibre[8072] Creating EPUB file...
20/04/09 8:34:18 PM [0x0-0x410410].net.kovidgoyal.calibre[8072] [INFO] Parsing -Tmp-/calibre_0.5.7_YpEZEQ_any2epub2/******

Yet the book conversion was actually started over 30 min earlier. The lit files are about 700-800 KB in size. Any thoughts? Am I not waiting long enough and should just leave these ones overnight?

[kovidgoyal]

Looks like there's something weird about these lit files. Opena ticket and atatch one of them.

[slantybard]

files emailed.

[artbatista]

Having the same problem here. I have opened the files in MS reader and extracted them with CLIT and they seem ok. I just tried a batch of 16 LIT files and it hung after the 3rd file.

CALIBRE-PARALLEL is running at 100% CPU and has a huge memory footprint (185MB). I stopped the conversion, but none of the other files converted. I had to restart calibre.

Art

[Student1]

Originally Posted by artbatista

Having the same problem here. I have opened the files in MS reader and extracted them with CLIT and they seem ok. I just tried a batch of 16 LIT files and it hung after the 3rd file.

CALIBRE-PARALLEL is running at 100% CPU and has a huge memory footprint (185MB). I stopped the conversion, but none of the other files converted. I had to restart calibre.

Art

Had similar problems with a few lits... will have more info once my collection is in... will have a bunch of lit to convert!

[artbatista]

Originally Posted by Student1

Had similar problems with a few lits... will have more info once my collection is in... will have a bunch of lit to convert!

Indeed, I have converted a few hundred MOBI files now, and a handfull (a dozen or so) required extraction and cleaning of the html to compile into epub successfully.

By contrast I tried about 3 dozen LIT files yesterday and gave up after a couple of hours or so since it seems every other file went 30 minutes or more without any signs of finishing, a few actually crashing the app.

I have better success extracting the lit files with CLIt and importing the resulting HTML directly.

A funny things is that if I extract the lit file, and compile the resulting OPF with Mobi Creator, I can usually then import the resulting MOBI file to Calibre and then convert to EPUB. Much longer process, hence I just extract to html and import directly.

I would say that there is something hinky with the LIT conversion routines. Not surprising really as LIT is a closed standard?, I figure most conversion processess are created by reverse engineering ( am I wrong?).

Cheers

Art