OpenOffice to ePub export extension

[roger64]

Hi!

A little conversion problem with a heading:

from ODT
Part One 'THE WORLD WILL HOLD ITS BREATH!'
to EPUB
PART ONE 'THE WORLD WILL HOLD ITS BREATH!'

[ajaxleon]

My first post is for you: ¡a lot of thanks for you job! is very interesting.
Lucas,thanks for share your time with us.

[eBookLuke]

Originally Posted by roger64

Hi!

A little conversion problem with a heading:

sorry, there are a lot of bug to solve.
Unfortunately I will be absent for a month or two, I'll resume the development ASAP

Luke

[roger64]

I wish you all the best.

I used your macro to process some plain epubs and it worked very well. The main "bug" was the random ordering of xhtml text parts that I had to reorder again using drag and drop. No big deal and it's already a very useful tool.

One idea for the future. As I am not a technician programmer, I do not know if it would be easy to implement it. So forgive it if it's impossible. Here it is.

Allowing the use of an other external stylesheet

You propose a standard CSS stylesheet and I agree with most of your choices. Most of them but not all. Which means I must manually make some modifications.

I think it's normal that each one may have its own CSS preferences, regarding font, margin, whatever. But it's a bit tedious to modify your standard stylesheet after every book. Could it be possible to offer the possibility to use (select, copy and insert) some custom-made external style sheet with as few clicks as possible? This alternative stylesheet would be placed somewhere on the user's computer.

I imagine a pop up screen: do you wish to install a different stylesheet? OK: where is it? OK. Done.

It would complete your macro. I am sure some "advanced" users could propose you later different stylesheets to put on your website to widen the choice of other users...This way, you could very easily make available the work of some MobileRead CSS specialists without tinkering with the delicate setup of your macro.

We'll have time to think about it.

[anabelee]

Thank you! I'll try it next weekend, Luke.

[eBookLuke]

Hi all,

I released a new version of writer2epub, more stable and with many improvements.

In my signature you will find the link to download it

Luke

[roger64]

Hi,

This is really an excellent release.

I used your extension to convert two odt books, one including a TOC, another with some images but no structure. I did not have the opportunity to test endnotes.

The results for both books are excellent and your extension is now my best choice. It can't be more easy to use. All the images and the TOC were correctly processed, which is a major usabillity achievement.

I have now only two remarks:
- the major one: the user still needs to use Sigil to do some tweaking after processing , like reordering and/or splitting text parts and also to check meta-data.

- the minor one: my personal choice would be to use slightly smaller margins (left and right, up and down) for text styles.

Congratulations for this release and

[eBookLuke]

Hello Roger!

Originally Posted by roger64

- the major one: the user still needs to use Sigil to do some tweaking after processing , like reordering and/or splitting text parts and also to check meta-data.

about metadata, if the filename is in the form "author - title" the macro gets the basic data (please note the space before and after the dash).
In a future release I will add a second button to insert metadata and adjust some preferences before export.

About splitting, the macro splits the text at every page break. If you wants to split your text, simply add a break where you want.
Moreover, if in the "Heading 1" style you specify a page break before stile, you will get automatically a splitted epub.

About the reordering, I still don't understand what you mean. All the files are generated, ordered and compressed in alphabetical order, and anyway the file sequence depends from the content.opf file. You don't have to worry about the order.

In my current release, I don't need anymore Sigil to create ePubs

- the minor one: my personal choice would be to use slightly smaller margins (left and right, up and down) for text styles.

If you look inside the macro, you will find a module called CSS. You can change the quoted text as you prefer.
Maybe in a future release I will find a simplest method to manage these preferences. Suggestions?

Thanks for your testing!

[st_albert]

Hello Luke!

Thanks for your efforts with this program. It is SO close to being usable.

But I must admit that I am also having problems with the order of file display. When I put it into sigil, the component files of the epub are displayed in the order that they are put into the <spine...> section of content.opf, like this:

<spine toc="ncx">
<itemref idref="content005.xhtml"/>
<itemref idref="content002.xhtml"/>
<itemref idref="content004.xhtml"/>
<itemref idref="content003.xhtml"/>
<itemref idref="content001.xhtml"/>
</spine>

also, when the created epub is opened in Firefox epub reader, the TOC displays all chapters, but they all point to the beginning of the single file that contains all the text of the book, i.e. they point to chapter one.

I have not modified the input file (which is in rtf format) to provide page breaks before each chapter, I must admit. It didn't seem worthwhile if all the resulting (component) text files in the epub would turn out in scrambled order.

So, there's still a ways to go.

I really do like your clean and simple CSS file though. Much easier to manage than any other I've tried so far, such as the InDesign epub export.

Here's hoping you persevere!

[eBookLuke]

Originally Posted by st_albert

Hello Luke!
… the component files of the epub are displayed in the order that they are put into the <spine...> section of content.opf, like this:

<spine toc="ncx">
<itemref idref="content005.xhtml"/>
<itemref idref="content002.xhtml"/>
<itemref idref="content004.xhtml"/>
<itemref idref="content003.xhtml"/>
<itemref idref="content001.xhtml"/>
</spine>

strange, strange, strange…

In all my files the spine is always correct and in the right order…
Why happens this in your environment?
In your opininion, are there some criteria to order the files in the order you get? I.E.: creation date, dimension or other?

Luke

[roger64]

Looks like this text part reordering and/or splitting is the biggest remaining hurdle to overcome.

1. Splitting

Splitting must be done at some time. It can be done
- within the original odt file,
- with your extension provided the file has a proper structure
- if the user forgets or neglects to do it, Sigil can help correct it later.

I had to use Sigil for example because my original odt text had no structure. It was my mistake. It was not your extension's fault but somewhere the splitting had to be done.

There is one OpenOffice extension which can nicely split a text, if it can be of any help: it's called writer2xhtml (last version v 1.0.2). It can mechanically split a text into workable chunks if you select this option. Maybe you could have a look at how it's done because if your goal is to process a file without Sigils' help, you will have to give your extension this capability .

2. Reordering

This is the most puzzling behaviour on my Linux box. I have no clues up to now. Only the fact that it does happen a mixup in the text chunks. I would suggest implementing a supplementary check within your macro with a kind of reordering mechanism. I do not know though if it is doable.

3. Metadata

There is a small problem with OpenOffice. The author's name is already specified even if you process a Dickens' novel. The language is also frequently forgotten or wrong.

I think it's not a big problem just to check the output on Sigil.

[st_albert]

Originally Posted by eBookLuke

Why happens this in your environment?
In your opininion, are there some criteria to order the files in the order you get? I.E.: creation date, dimension or other?

Luke

I don't know. When I unzip the epub, the files in OEBPS/text are:

$ ls -l
total 552
-rw-r--r-- 1 u83 u83   1007 Jun 28 02:42 content001.xhtml
-rw-r--r-- 1 u83 u83   1434 Jun 28 02:42 content002.xhtml
-rw-r--r-- 1 u83 u83 543032 Jun 28 02:42 content003.xhtml
-rw-r--r-- 1 u83 u83   3549 Jun 28 02:42 content004.xhtml
-rw-r--r-- 1 u83 u83   2135 Jun 28 02:42 content005.xhtml

As you can see, they all have the same creation date/time. The content is as follows:

content001.xhtml is the title page
content002.xhtml is the copyright page
content003.xhtml is the body of the book
content004.xhtml is the "about the author" page
content005.xhtml is the last page (advertisements)

Given the file breaks based on pagebreaks in the document, this is the correct order for the files. That is, the alphanumeric sort order for the file names in the epub matches the order in the original book.

The only problem is they don't appear in the correct order in the <spine> section of content.opf, as noted above.

There is one other thing: I started by opening an .rtf file in OOO-writer. Could that have made a difference?

::time passes:: Apparently, not. I just saved the .rtf as an .odt and re-exported the epub. Same result in content.opf.

Also, by the way, the toc.ncx is incorrect. All the chapters are listed, but they all point to the TOP of the .xhtml file which contains the body of the book, rather than pointing to targets within the file where the chapters actually start. (This gets automatically fixed by sigil, if you open and save from sigil.)

Sorry I couldn't be more help.

Edited to add: I am running OOO writer (ver. 3.2.1.4) and sigil (ver 0.2.3) on Debian Sid Linux. writer2epub is ver. 0.3.0

[eBookLuke]

Hi Roger, Hi Albert…

I solved the ordering problem, by force the order when create content.odf file.
Please test the attached macro, and remember to change the extension from .zip to .oxt before install it.

This version adds also a basic table management and is almost all rewritten to be faster.

Please give me your feedback.

About sliting, I will add a forced split after some kBytes of generated file in the final version. Anyway, the macro splits the file each page break inserted along the text.

Thanks for your support

Luke

[st_albert]

Oops. No joy. Upon hitting the "export" button I get an error: "Basic runtime error. Object variable not set."

The macro editor opens with a line highlighted in "function fnExportParagraph(oParagraph As Object)" as follows:

				if sLink <> "" then                     'this portion is a hyperlink 
					For i = 0 to oBMarks.Count - 1 
						sNam = oBMarks.getByIndex(I).Name 
						if InStr(sLink,sNam) > 0 then      'only internal links are recognized! 
'							sPortion = "\q=""#" & sNam & """" & sPortion & "\q" 
							exit for 
						end if 
					Next 
				end if

where the line " For i = 0 to oBMarks.Count - 1" is highlighted.

[st_albert]

OK, I just tried with a different .rtf file to start with, and everything worked! Very nice splits on chapter headers (not sure, but probably they were associated with pagebreaks in the original file -- it wasn't mine). The TOC works, too.

Only error it showed with epubcheck-1.0.5 was:

"length of first filename in archive must be 8, but was 22"

meaning, I suppose, that the file "mimetype" wasn't correctly zipped (?). I re-zipped the subdirectory left behind by writer2epub, and this time no errors were detected.

Sweet!

But the file I started with still causes the same BASIC error.