[Plugin] ODTImport

[Julie34]

Works fine for me. But I have a question : How to import Dublin Core metadata with ODTImport ?
With the add of : "<xhtml-style-map name="custom-body-text" family="paragraph"
element="p" css="custom-bodytext" />" and some elements in a secondary epub.css file (or not), it add fine the styles or "normal" metadatas of the ODT but ignore the "dc:..." meta.
Many thanks.

[Doitsu]

Originally Posted by Julie34

How to import Dublin Core metadata with ODTImport ?

The plugin is using writer2latex.jar for the conversion, which has only limited metadata support. In LibreOffice, select File > Properties > Description to add metadata. Title, will be mapped to <dc:title>, both Subject and Keywords will be mapped <dc:subject> and Comments will be mapped to <dc:description>.
You'll need to manually add all other metadata entries in Sigil via Tools > Metadata Editor...

[Julie34]

Many thanks for the answer.
I allready did it manually but was interested to know if it was possible.
<subject> is not separated when severall items with coma. I had to do it manually too.

Is there another possibility or plugin to export from odt or word and keep the styles like here and metas ?

[Doitsu]

Originally Posted by Julie34

Is there another possibility or plugin to export from odt or word and keep the styles like here and metas ?

I don't understand why you keep referring to styles, because styles and metadata entries are unrelated. If you were referring to the formatting of metadata entries, what formatting problems have you encountered?

As for other conversion options, you could try and save your file as a .docx file and then either convert it with Calibre or the DOCXImport Sigil plugin.

However, since it's much easier to add and edit metadata entries in Sigil, you might want to consider adding them there.

[Julie34]

You're right, here i asked for the metas.
Thanks a lot ! I Have all the informations I needed.
I mentioned styles because, of course i'm interested in keeping them too. That the ODTImport do perfectly from the odt.
And, i think too that it's easier to add a meta with the Metas Editor Tool in Sigil even if it's possible to add them directly in the content.opf.
So i'll keep to ODTImport for a part even if i'm still using the old "Amanuensis" sometime when in windows.

[Doitsu]

I've attached a new plugin version in two "flavors" to the first post. The first one, ODTImport_0.4.0.zip, contains the latest version of Writer2xhtml 1.7 with several bug-fixes by the original author. For more information, please see the official release notes.

The second one, ODTImport_0.4.1.zip, contains an unofficial version of Writer2xhtml 1.7 with several cosmetic changes for better epub3 compatibility.

My version:

• removes the linear="no" attribute from the Cover page spine entry
• adds the missing epub3 properties="cover-image" manifest entry attribute
• adds Cover, TOC and Bodymatter landmarks semantics entries to nav.xhtml

(In accordance with the GPL license, the 2 modified Java files are included in the plugin folder.)

Unless you're really interested in these particular epub3 features, you might want to stick with the official release (ODTImport_0.4.0.zip).

[retiredbiker]

Thank you! I installed this last night and gave it a quick test...seems fine. In a day or two I'll have an odt of an old novel to really try it out.

[retiredbiker]

I'm new to Sigil and new to this import plugin. Just used it for a full-length novel newly OCR'd and proofed in Writer. Works very well. (Much better than the Calibre odt conversion, which is unmaintained.)

Main thing I like: it keeps my css selectors intact, exactly as in the odt file. It does not generate any css with them, but I simply dump the example css, pull in my custom css, and it works. It uses <i> for italics, not some invented <span> code like Calibre does.

Splits the file on <h1> tags. That is sufficient; in the future I'll add dummy <h1>s to get any extra splits I want--for a dedication page, for example.

Very quick, very clean.

[roger64]

Indeed.

Many thanks for this welcome release!

Most of the work on the software can be controlled thru the use of a unique text configuration file. A recommended step before first using it regularly would be to have a look at it, and if need be, to select your custom options (for example, how to split files .)

This file, named config.xml can be found in the ODTImport folder. This folder is among the plugins Sigil folder. If you wish so, you can also insert in this latest folder your own epub.css file or a resources folder containing for example fonts to be embedded or any other file you may wish to include in your ePub.

The underlying java engine (writer2xhtml) has now been dissociated from writer2latex. For the time being, the relevant writer2latex manual contains also the documentation for writer2xhtml (English) v. 1.6.1.

https://writer2latex.sourceforge.net...r2xhtml10.html

Enjoy!

[retiredbiker]

@roger64, thanks for the pointers. I had not gotten to the config file yet, and looking at it with your link to the writer2xhtml manual makes it very easy to see what is going on. I've made a couple of small changes, and will have fun playing with them!

[roger64]

You're welcome.

To correct a mistake of mine: when you download writer2xhtml v. 1.7, you'll find in the zip (within the doc folder) a user manual in odt format for version 1.7. It's quite an exhaustive document (about 30 pages).

[philja]

I've upgraded the plugin to 0.4.1 in order to make some tests with the config.xml to see if I can improve on the results I obtained previously (in 2022) with the conversion of lists from the odt to the epub.

On reading the latest "user-manual-Writer2xhtml17.odt" with special reference to anything mentioning lists, I saw the option "list_formatting" with three possible values - ignore_all, convert_labels, convert_all. This is at odds with line 20 in the config.xml supplied with the plugin 0.4.1 which quotes "value=css1"

The footnote (7) in the writer2xhtml17 documentation says that option "list_formatting" is a new option in v17 and replaces a previous option called "list_hack". The option was already called "list_formatting" back in writer2latex v1.4 (page 51) and at that time, a possible value was indeed css1. That value no longer exists so line 20 of config.xml could be updated in the current plugin version.

Incidentally, when updating the plugin in Sigil, I made a rookie error. Using Sigil's gui for Manage Plugins, I removed the old version and installed the new one. And my customised epub.css and config.xml files were lost, replaced by the basic versions in the new zip file. Fortunately, I had ~ backups from which to recover them—a useful reminder to keep copies in a safe place.

[Doitsu]

Originally Posted by philja

On reading the latest "user-manual-Writer2xhtml17.odt" with special reference to anything mentioning lists, I saw the option "list_formatting" with three possible values - ignore_all, convert_labels, convert_all. This is at odds with line 20 in the config.xml supplied with the plugin 0.4.1 which quotes "value=css1"

I wasn't aware of that. I'll update config.xml for the next version. I'll also add support for configuration files in the preferences folder, which isn't deleted if the plugin is uninstalled.