CHM Conversion fails with memory error

nicolasblank · 08-29-2011, 05:57 AM

Hi all
I'm trying to convert the Exchange 2010 .CHM help file into .mobi
The CHM file can be found here:
http://www.microsoft.com/download/en...s.aspx?id=1573
I've tried this and a previous version of calibre but the conversion fails with the same error, which is "MemoryError: out of memory"
The error log is attached

Manichean · 08-29-2011, 09:00 AM

That error indicates that there's too little memory (RAM) available to finish the conversion. There's no fix except to upgrade your computer.

nicolasblank · 08-29-2011, 09:14 AM

I'm running this on a 64bit machine with 8GB of RAM, with 5GB free while the conversion process is running. The process is 32 bit, which theoretically won't take more than a few GB's and the process doesn't even reach one GB at runtime.

nicolasblank · 08-29-2011, 10:18 AM

took a while to rerun - every run takes nearly an hour to fail:
I confirm that I had well over 3GB free
the process maxed out at 1867640 KB (1.78GB).
My Machine had pleanty of available RAM, I expect the process is contrained by a 32bit limit, or 32 bit Python might have hit a limit.
Any ideas on how to work around this issue? Any tweaks I can try in the product?

Manichean · 08-29-2011, 10:21 AM

That sounds like it might be a 32 bit limit issue. I don't know if and how you can work around that, but I'm pretty sure someone from the dev team will be along later.

kovidgoyal · 08-29-2011, 10:59 AM

calibre is designed to keep the entire contents of an ebook that it is
processing (except for images) in memory. This makes
transformations on the text and formatting easy enough to implement
that they actually end up getting implemented. The downside of this
approach is that calibre cannot handle very large documents. I
suggest you find a way to split up your document outside calibre,
or use some other tools to convert it.

nicolasblank · 08-29-2011, 11:42 AM

Fair enuf. I don't suppose you'reworking on a 64bit build since CHM's can't be decompiled into anything decent that keeps a structure.

kovidgoyal · 08-29-2011, 11:50 AM

64bit build are available for linux and OS X, just not windows. I currently have no plans for a 64bit windows build.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
PDF conversion fails	farble1670	Calibre	2	10-09-2010 07:28 PM
Conversion memory error (HTML->EPUB)	doremifaso	Calibre	4	06-25-2010 10:56 PM
Conversion error: CHM to PDF or "anything else" conversions FAIL. Help.	mm07	Calibre	3	05-16-2010 09:41 AM
Catalog Function Fails With Memory Error	Ambermonk	Library Management	4	03-08-2010 01:10 PM
Catalog Function Fails With Memory Error	Ambermonk	Calibre	0	03-07-2010 08:54 PM

08-29-2011, 09:00 AM	#2
Manichean Wizard Posts: 3,130 Karma: 91256 Join Date: Feb 2008 Location: Germany Device: Cybook Gen3	That error indicates that there's too little memory (RAM) available to finish the conversion. There's no fix except to upgrade your computer.

08-29-2011, 09:14 AM	#3
nicolasblank Junior Member Posts: 4 Karma: 10 Join Date: Aug 2011 Device: Kindle	I'm running this on a 64bit machine with 8GB of RAM, with 5GB free while the conversion process is running. The process is 32 bit, which theoretically won't take more than a few GB's and the process doesn't even reach one GB at runtime.

08-29-2011, 10:18 AM	#4
nicolasblank Junior Member Posts: 4 Karma: 10 Join Date: Aug 2011 Device: Kindle	took a while to rerun - every run takes nearly an hour to fail: I confirm that I had well over 3GB free the process maxed out at 1867640 KB (1.78GB). My Machine had pleanty of available RAM, I expect the process is contrained by a 32bit limit, or 32 bit Python might have hit a limit. Any ideas on how to work around this issue? Any tweaks I can try in the product?

08-29-2011, 10:21 AM	#5
Manichean Wizard Posts: 3,130 Karma: 91256 Join Date: Feb 2008 Location: Germany Device: Cybook Gen3	That sounds like it might be a 32 bit limit issue. I don't know if and how you can work around that, but I'm pretty sure someone from the dev team will be along later.

08-29-2011, 10:59 AM	#6
kovidgoyal creator of calibre Posts: 43,835 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	calibre is designed to keep the entire contents of an ebook that it is processing (except for images) in memory. This makes transformations on the text and formatting easy enough to implement that they actually end up getting implemented. The downside of this approach is that calibre cannot handle very large documents. I suggest you find a way to split up your document outside calibre, or use some other tools to convert it.

08-29-2011, 11:42 AM	#7
nicolasblank Junior Member Posts: 4 Karma: 10 Join Date: Aug 2011 Device: Kindle	Fair enuf. I don't suppose you'reworking on a 64bit build since CHM's can't be decompiled into anything decent that keeps a structure.

08-29-2011, 11:50 AM	#8
kovidgoyal creator of calibre Posts: 43,835 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	64bit build are available for linux and OS X, just not windows. I currently have no plans for a 64bit windows build.