Renaming filename *.epub based on meta data fields?

[DoctorOhh]

Originally Posted by ibu

OK, thanks for the hint. I thought, that Calibre needs the original path, because it does not copy the original file physically in it's database.

Calibre does copy the original file physically to it's database structure. Calibre creates the directories and file names based on the metadata. If the metadata changes so does the associated files or directories in calibre's library.

[ibu]

Than it's even easier to perform my task.

Calibre is really an incredible powerful and smart tool.

Thanks to all of you who contribute to it

Can anybody tell me a good information source (webpage), where it is explained, which epub optimizations are useful in a way which is easy to understand for a newbie in epub editing?

What I'm specially looking for:

Heuristics, which eliminate all source code of the html files, which is only for layout and not for semantic.
My wish is to have epubs in plain semantic xhtml. Free of redundant and ugly tags, classes, ids, inline-styles.

In other words:
Books without any layout at all.
Just structure.

Another thing:
Do modern eReaders have problems with ePubs in a single HTML file?

[BetterRed]

Originally Posted by ibu

Can anybody tell me a good information source (webpage), where it is explained, which epub optimizations are useful in a way which is easy to understand for a newbie in epub editing?

Did you try the User Manual - press Function Key #1 next time you're in Calibre, or go here http://manual.calibre-ebook.com/

Originally Posted by ibu

What I'm specially looking for:

Heuristics, which eliminate all source code of the html files, which is only for layout and not for semantic.
My wish is to have epubs in plain semantic xhtml. Free of redundant and ugly tags, classes, ids, inline-styles.

There's a very good EPUB editor called Sigil https://www.mobileread.com/forums/forumdisplay.php?f=165

BR

[theducks]

Originally Posted by ibu

@DoctorOhh

OK, thanks for the hint. I thought, that Calibre needs the original path, because it does not copy the original file physically in it's database.

Just the opposite.
Calibre copies the Book into its (database) structure. (the structure is the DB and the PRECISE Folder-File calibre stored the book into)

The original can go Bye-Bye after that.

[ibu]

[Optimizations]

Originally Posted by BetterRed

Did you try the User Manual - press Function Key #1 next time you're in Calibre, or go here http://manual.calibre-ebook.com/


There's a very good EPUB editor called Sigil https://www.mobileread.com/forums/forumdisplay.php?f=165

BR

The focus of manuals of tools is typically on functionality.

I'm looking for an article with basic thoughts with the focus on minimalistic code.
I think there are only few user who like the radical wish to have an epub without any "styles", just pure semantic.

When I understand
http://idpf.org/epub/30/spec/epub30-overview.html
correctly, there are 4 obligatory files of a minimalistic epub:
/META-INF/container.xml
/mimetype
/directory.opf
/content.xhtml

And 2 more files for the TOC and the cover:
/cover.jpg
/toc.ncx

[Adoby]

One way could be to convert the ePub to text or rtf and then back to ePub. Not much left of formatting after that, I would imagine. But not headings either. Add some minimal formatting in Sigil after that, like chapter headings.

[ibu]

@Adoby
Converting to TXT deletes all semantics like Headers, emphasized text, ordered lists etc.

For an analysis of dispensable elements, classes, ids, inline-styles you need a tool, which lists all of them and allows to dispense them with "a click". Without dispensing the content of these elements inside the tags.

[theducks]

While neither a Elegant nor a streamlined codewise solution:
Simply make the style empty (or comment it out).

.foo { }
.bar ( /* Display: Block; font-size: 36pt; */}

Results is a plain, default rendering

[ibu]

@theducks

I know how to edit CSS

But indeed I'm interested in a "normalization of epub code" with as less effort as possible (therefor I asked for heuristics and scripts which help cleaning dirty code) and as much clarity in the code as possible as well.

Little question:
Does anyone of you has tried to use an epub of a typical ebook in form of a single html file inside the container (instead of splitted HTML files)?
Can Kobo Glo digest it?

[PeterT]

@ibu: Why don;t you try and let people know?

[ibu]

@PeterT
Because I'm completly new to eReaders and ePubs.
Without experiences it's hard to estimate the result of experiments.

That's why I ask here

And: I didn't figure out how to merge multiple HTML files in sigil. I cannot mark more than one file in the book browser.
http://web.sigil.googlecode.com/git/...d_merging.html

[PeterT]

Just highlight all (x)html files in the book browser view on the left hand side, right click them and select merge!

Of course, normally you would want to ensure all the files used the same style sheet before starting, but I sense you want to remove all styles in any case, so this should not be much of a concern.

[theducks]

What version of Sigil? 7.x on windows does not have that problem.

Marking is the standard keyboard way for multiple file selection. (Drag selection does not function)
Select the first
Then hold the shiftkey and select the last

or
hold the control key and click each


then
Ctrl-M for Merge

Huge sections and device performance suffers, even if it does not choke.

Many older devices have a ~200K limit before they choke.

[ibu]

@PeterL and theducks

Shame on me.
My version of Sigil was to old.
Sorry, please.

In the 0.7.3 I can highlight multiple files.
Works fine with the normal "chapter html files".


Now I have figure out, how to merge even files like cover.xhtml, title.xhtml etc. properly with the "chapter files".

Whow, Sigil is a really powerful editor. A lot to discover.

[theducks]

Originally Posted by ibu

@PeterL and theducks

Shame on me.
My version of Sigil was to old.
Sorry, please.

In the 0.7.3 I can highlight multiple files.
Works fine with the normal "chapter html files".


Now I have figure out, how to merge even files like cover.xhtml, title.xhtml etc. properly with the "chapter files".

Whow, Sigil is a really powerful editor. A lot to discover.

You should be able to merge any file in the Text section.

That said: Beware
Not all files use the same stylesheets (or any. Cover being one.)
Note: Sigil (HTML)TOC has its own stylesheet and depends on the semantics being set
Which would be the same issue with Cover. semantics (sets the guide) are at the file level. KISS keep them Simple and Separate