[GUI Plugin] EpubMerge

[JimmXinu]

Version 2.0.3 - 23 Sep 2014

• New translation for Swedish, courtesy of the volunteers at Transifex.

[JimmXinu]

There's been a calibre change (in 2.10) that requires changes to some of the plugins to correctly support keyboard shortcuts. Specifically, plugins that change their menus dynamically.

Unlike the other test versions I've posted recently, this one addresses the problem by removing the dynamic menu generation. It wasn't really needed.

Attached is a version of this plugin I've updated. I've done some testing with it on calibre 2.10, 2.9 and 1.48 and it seems good. I'd like to hear if anybody has problems with it.

Confirmation that it fixes the shortcut issue for you would also be appreciated.

[JimmXinu]

Version 2.0.4 - 18 Nov 2014

• Fix for calibre 2.10's changes to how keyboard shortcuts work. Removes unnecessary dynamic menu.

[sprite7]

Can you add a setting to not generate images for internal covers? There are problems when convert to mobi/azw3 (bigger file size, hi-res images make kindle slow down and sometimes reboot).

[JimmXinu]

Originally Posted by sprite7

Can you add a setting to not generate images for internal covers? There are problems when convert to mobi/azw3 (bigger file size, hi-res images make kindle slow down and sometimes reboot).

EpubMerge doesn't generate any images, all it does is include images that were already there--including covers. I'd rather not see changing things along the way for specific issues.

One approach would be removing the cover images before merging (see the Modify Epub plugin). Copy to a temporary library first if you still want to keep the original epubs.

There are some indicators in the calibre conversion docs (... images to be auto-resized to be fit to the screen in some output formats.) that conversion can sometimes resize images. I don't know any details myself, however.

[DoctorOhh]

Originally Posted by sprite7

Can you add a setting to not generate images for internal covers? There are problems when convert to mobi/azw3 (bigger file size, hi-res images make kindle slow down and sometimes reboot).

You can always open the book in the editor and remove those images before converting to mobi/azw3.

[sprite7]

I converted htm->epub with calible, and there was parameter to generate cover if none. Thanks for replies.

[imaginedtruths]

Would there be a way to combine EPub Merge with Marvin? Sort of a Epub + Epub + Epub etc. -> auto sets the merged Epub maintaining the original table of contents , with the best possible settings of Epubs for the app Marvin, -> that in process auto downloads any missing covers and metadata, and -> if connected to an iPhone/iPad via USB auto sends the finished combined Epub directly to Marvin. I know that would be a bit complex, but for those of us that collect and own legal copies of Epubs that no existing complete all in one Epub books exists. A great example would be, a complete collection of every book published by Stephen King. All merged together in published date, maintaining original book covers, original book comments and table of contents, made into a merged ePub and with settings that work best on Marvin. Might seem a bit ambitious but would be amazing.

[JimmXinu]

Originally Posted by imaginedtruths

Would there be a way to combine EPub Merge with Marvin? Sort of a Epub + Epub + Epub etc. -> auto sets the merged Epub maintaining the original table of contents , with the best possible settings of Epubs for the app Marvin, -> that in process auto downloads any missing covers and metadata, and -> if connected to an iPhone/iPad via USB auto sends the finished combined Epub directly to Marvin. ....

I believe tools already exist in calibre (with some plugins) to do everything you've mentioned.

I, personally, don't have any interest in trying to automate them to do it all in one step.

[miajoshua]

Hi! Still a relative newbie in this ebook thing. Learning to use plugins, converting ebooks etc. Very much appreciate this plugin - would like to ask some questions, hopefully not too silly.

Okay, now I noticed you can configure CUSTOM columns, but what about other regular columns? I lose all the info for: Published, Publisher, Ids.

Also a small request: can we have the "Anthology" automatically added to the title of the merged epub so there won't be two epubs with the same title (the pre-existing one and the merged one)

Or if there is a better way to manage that I'm unaware of, please let me know!
Thanks!
Mia

[JimmXinu]

Originally Posted by miajoshua

Okay, now I noticed you can configure CUSTOM columns, but what about other regular columns? I lose all the info for: Published, Publisher, Ids.

Ids cannot (or at least IMO should not) be copied because each source book has different Ids and calibre will only let a book have one isbn id, one url id, etc.

Setting the Publisher if the sources all have the same Publisher is not unreasonable--the field only allows one value.

Setting the Published date is also possible, but which date should be used? The oldest? The newest? And if I put it one way, somebody's going to want the other. I'll think about it.

Originally Posted by miajoshua

Also a small request: can we have the "Anthology" automatically added to the title of the merged epub so there won't be two epubs with the same title (the pre-existing one and the merged one)

Unless you've changed it, EpubMerge should already be doing that.

[JimmXinu]

Attached is a test version that fills in the Publisher when all merged books have the same Publisher.

I'm curious what people think about filling pubdate. Yes? No? Oldest? Newest? First listed book?

[davidfor]

Originally Posted by JimmXinu

Ids cannot (or at least IMO should not) be copied because each source book has different Ids and calibre will only let a book have one isbn id, one url id, etc.

There was a change recently to allow multiple ids of the same type. For most identifiers, it needs support from the metadata source plugin, but for URLs, you can have "url1", "url2" etc. So, you could take all unique identifier types plus the identifier from the first book and render the rest as URL and add them.

[embryo]

First of all, thank you for your efforts.

I don't use your plugin very often, but sometimes its really a time saver.

Among other things I use it to make anthologies of articles I have downloaded in epub format.
I have already made 3 of these collections with some articles from a specific site that I visit and everything was ok since the last time.

Today I tried to make a forth collection and I discovered a couple of problems.
Every epub I was trying to merge had 3-6 articles (chapters).
The xhtml inside them had names as article0.xhtml, article1.xhtml, article2.xhtml etc.
The images had names like img0.jpeg, img1.jpeg, img2.jpeg etc.
There were 12 epubs.

The first problem is that although the TOC in the final epub was generated ok, the articles that it was pointing to where wrong and the wrongness was random.
Looking with the Sigil discovered that the xhtml files were in the wrong order in the book (a different wrongness this one).
But that was not crucial.
I re-made the TOC (from the headers) to point to the right articles and the order was not important for that anthology.

The real problem was that a lot (but not all) of the pictures were mixed up.
I see the same picture in more than one article (taking the place of an other pictures).
Some of them are really famous and are in more articles than others.
There was not a pattern that I could recognize to that, so its really difficult to fix it.

The procedure for the epub creation was the same as the previous collections.

I tried different settings with no luck.
I use Calibre 1.48 (WinXP) with the latest version (2.0.4) of your plugin.
I always update it, so maybe some update break it for me.

Is there a version prior to 2.0 that I could check?

TIA

[JimmXinu]

Originally Posted by embryo

...
Every epub I was trying to merge had 3-6 articles (chapters).
The xhtml inside them had names as article0.xhtml, article1.xhtml, article2.xhtml etc.
The images had names like img0.jpeg, img1.jpeg, img2.jpeg etc.
There were 12 epubs.

That's not uncommon. Many epubs generated from the same system will have the same file names inside.

EpubMerge deals with it by putting each separate source epub in a separate directory inside the merged epub. Which is acceptable according to the standards and most readers, but causes some editors problems. (My other plugin, FFDL, produces epubs with the same file names and EpubMerge works fine with it.)

Originally Posted by embryo

The first problem is that although the TOC in the final epub was generated ok, the articles that it was pointing to where wrong and the wrongness was random.
Looking with the Sigil discovered that the xhtml files were in the wrong order in the book (a different wrongness this one).

Unless it's been significantly upgraded since the last time I looked, Sigil does not correctly handle epubs with multiple directories inside and the devs had no plans to fix it. I don't think you can trust what Sigil tells you about the structure of an EpubMerge epub. Nor can you trust an epub saved from Sigil that had multiple dirs to start with.

Furthermore, calibre's builtin epub editor, while it handles multiple dirs better, still tries to 'simplify' things for you. It also doesn't show you the true structure. I believe it does a better job of saving the structure, but I can't promise it.

Originally Posted by embryo

But that was not crucial.
I re-made the TOC (from the headers) to point to the right articles and the order was not important for that anthology.

The real problem was that a lot (but not all) of the pictures were mixed up.
I see the same picture in more than one article (taking the place of an other pictures).

If you changed the TOC in Sigil and saved it, that's probably what corrupted the image links.

Originally Posted by embryo

Is there a version prior to 2.0 that I could check

There haven't been any changes to the actually epub merging part of the code in 8 months, well before the changes for calibre 2.0.

If you can provide me with copies of the source epubs you merged and, ideally, the merged epub as it first came out of EpubMerge that shows the problems I might be able to help more.