Converting to HTMLZ, two copies of cover?

shandrakor · 04-11-2012, 06:32 PM

All my htmlz conversions are ending up with two copies of the cover image file, one in the root as cover.jpg, the other in the images subdirectory as 0001.jpg.

Is this by design? Is there any way to avoid it? "Ignore first image" doesn't do it.

itimpi · 04-11-2012, 10:40 PM

In it's internal file store Calibre always stores a cover.jog. This is in addition to whatever is sorted inside any particular eBook file. This is done as many ebook formats do not support string the cover image, and calibre stores all formats for a given book in the same folder. The cover.jpg file is the one used in the calibre GUI. It need not be the same as the cover image stored inside any particular eBook file.

You cannot change this behaviour. However since you are not meant to be accessing the internals of the calibre library directly this should not concern you.

shandrakor · 04-12-2012, 12:00 AM

What's looking in the library got to do with it? When creating an htmlz file, it includes both versions of the cover jpg. It's still there after doing a "save to disk" of the htmlz file.

itimpi · 04-12-2012, 01:24 AM

If you do a save to disk, then whether the cover.jpg gets saved as well as a separate file is optional (as is the .opf file).

theducks · 04-12-2012, 07:58 AM

HTMLZ has no way of knowing that 0001.jpg is a cover. Unlike EPUB, there is no defined 'reference' section (in the content.opf) to consult.

I will assume that 'remove first image' applie to the Location of usage in the HTML and not its filename or position in a file list.

shandrakor · 04-12-2012, 01:32 PM

Quote:

Originally Posted by itimpi

If you do a save to disk, then whether the cover.jpg gets saved as well as a separate file is optional (as is the .opf file).

I'm clearly not conveying my issue well. It is optional to save copy of the cover and opf when doing a save to disk _in addition to all exported formats_. I'm not talking about a loose jpg. I'm talking about the contents of the created htmlz file.

The purpose of htmlz was to export a single, flat html file for editing more easily than the contents of an epub. But in addition to that, calibre is including parts of its internal library structure in the archive--specifically, the cover.jpg file.

If I export an epub, only one jpg is included. If I export to mobi, the file size suggests that only one copy of the jpg is included. But if I export to htmlz, it includes both the "cover according to the epub" file and the "cover according to calibre" file, even if those files are identical.

Quote:

Originally Posted by theducks

HTMLZ has no way of knowing that 0001.jpg is a cover. Unlike EPUB, there is no defined 'reference' section (in the content.opf) to consult.

True, but that should only be an issue when importing from htmlz. When creating the htmlz in the first place (converting from epub or mobi) calibre does know whether the referenced file is a cover.

04-11-2012, 06:32 PM	#1
shandrakor Junior Member Posts: 4 Karma: 10 Join Date: Apr 2012 Device: ipad	Converting to HTMLZ, two copies of cover? All my htmlz conversions are ending up with two copies of the cover image file, one in the root as cover.jpg, the other in the images subdirectory as 0001.jpg. Is this by design? Is there any way to avoid it? "Ignore first image" doesn't do it.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
keep cover picture when converting several ebooks	HiRsch	Conversion	1	04-01-2012 12:54 AM
HTMLZ	Ortep	Calibre	21	05-09-2011 10:27 PM
blank page for cover after converting	crestfalleen	Calibre	2	06-03-2010 12:08 PM
DR800 Converting a planner to a cover	The Straven	iRex	0	05-02-2010 12:04 AM
Problems with Cover Images after converting	NASCARaddicted	Calibre	7	05-06-2009 03:14 AM

04-11-2012, 10:40 PM	#2
itimpi Wizard Posts: 4,553 Karma: 950151 Join Date: Nov 2008 Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)	In it's internal file store Calibre always stores a cover.jog. This is in addition to whatever is sorted inside any particular eBook file. This is done as many ebook formats do not support string the cover image, and calibre stores all formats for a given book in the same folder. The cover.jpg file is the one used in the calibre GUI. It need not be the same as the cover image stored inside any particular eBook file. You cannot change this behaviour. However since you are not meant to be accessing the internals of the calibre library directly this should not concern you.

04-12-2012, 12:00 AM	#3
shandrakor Junior Member Posts: 4 Karma: 10 Join Date: Apr 2012 Device: ipad	What's looking in the library got to do with it? When creating an htmlz file, it includes both versions of the cover jpg. It's still there after doing a "save to disk" of the htmlz file.

04-12-2012, 01:24 AM	#4
itimpi Wizard Posts: 4,553 Karma: 950151 Join Date: Nov 2008 Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)	If you do a save to disk, then whether the cover.jpg gets saved as well as a separate file is optional (as is the .opf file).

04-12-2012, 07:58 AM	#5
theducks Well trained by Cats Posts: 30,897 Karma: 60358908 Join Date: Aug 2009 Location: The Central Coast of California Device: Kobo Libra2,Kobo Aura2v1, K4NT(Fixed: New Bat.), Galaxy Tab A	HTMLZ has no way of knowing that 0001.jpg is a cover. Unlike EPUB, there is no defined 'reference' section (in the content.opf) to consult. I will assume that 'remove first image' applie to the Location of usage in the HTML and not its filename or position in a file list.

Advert

Advert