ODFToEPub-1.1.3

Extension:
Extension ID:
Release date:
05/10/2010 - 09:31
Version:
1.1.3
Operating system:
Windows, Linux, MacOS X, Linux x86-64
Compatibility:
--
Binary type:
OOo extension (.oxt)
Release Notes: 

Release Notes ODFToEPub 1.1.3
5 October 2010

1. Requirements

You need OpenOffice 3.2 or higher to use the ePub extension. You may also run
it as a stand-alone program. Either double-click the JAR-file or write the
following on the command-line:

> java -jar odf_to_epub.jar

The supported platforms are Windows, Mac OS X and Linux. Java 1.5 or higher has
to be installed in OpenOffice.

2. Installation

Install the extension with the extension manager, which is behind the menu item
Tools->Extension Manager.

3. Changes

3.1 Changes Since Version 1.1.2

Bug fixes

- Justification and centering doesn't work for headings.
- The default master page is not used to determine the width of tables and
frames.
- Use a colgroup with the "width" attribute set to "0*" for better and more
portable spacing of itemized list labels.
- Exception when a master page doesn't specify an absolute width.
- Fix incorrect ID-format for embedded fonts.

Minor changes

- If an image has the name "Cover" it is used as the cover page instead of the
first entity in the generated document.
- The entities are generated in the root of the container instead of in the
folder "parts". This removes the need for relative URLs containing "../",
which doesn't work on all e-readers.

3.2 Changes Since Version 1.1.1

Bug fixes

- Escape special characters in style names.
- Put quotes around font-family names with spaces.
- Take the book identifier from the ODF-document.
- Compile for Java 1.5.

3.3 Changes Since Version 1.1

Bug fixes

- An Asian CSS style is not generated if there is only the "font-weight-asian"
property.
- If a "span" element has a style without a language code (directly or
indirectly through the parent style), the language code of the surrounding
element should be used.
- Missing Asian Unicode blocks.
- Language code "zxx" and country code "none" are not interpreted correctly.
- Exception while building the spine when the highest outline level is lower
than 1.
- The property "text-line-through-style" with value "none" is translated to
"line-through" instead of "none".

3.4 Changes Since Version 1.0.1

Bug fixes

- The "text-underline-style(-type)" and "text-line-through-style(-type)"
properties are not converted correctly when the value "none" is involved.
- A list style is not applied to the list labels.
- The Dublin Core language element should always be added to the meta data.
- An XHTML part must not have an empty body.
- A computer username with non-ASCII characters prevents the installation of
the license file.

New features

- Support for writing-modes (left-to-right and right-to-left).
- Support for Asian and Complex Text Layout ODF-styles.
- Index of tables.
- Index of illustrations.
- More control over the layout of the generated table of contents, index of
tables, index of illustrations and the alphabetical index.
- Frames can be floated, centered and left or right aligned.
- Improved frame handling. The width of a frame is now proportional to the
width of the surrounding frame or the page gallery.
- Improved image handling. The width of an image is now proportional to the
width of the surrounding frame.
- Improved table handling. The width of an outer table is now proportional to
the width of the page gallery. The column widths have the same proportions as
in the source document.
- Background images for paragraphs.
- Control over what goes into the spine of the book.
- Segmentation control through page breaks.

3.5 Changes Since Version 1.0

Bug fixes

- Generate the "xml:lang" attribute instead of "lang".
- Avoid gaps in the playOrder of the NCX-file.
- Take into account the anchor-type of a frame.
- Work around duplicate index mark ids in OpenOffice.
- Use "id" attribute instead of "name" on the "a" element.
- Generate a default "tbody" when there is no table row group in a table.
- Correct references to pictures in the OPF-file.
- Generate nested anchor elements for nested reference marks.
- Replace illegal characters when converting a reference name to an XML-id.

4. Contact Information

Comments and bug reports can be sent to support@pincette.biz. Information about
the application is available at http://www.pincette.biz/odftoepub/.