Changelog for Writer2xhtml version 1.6.1 -> 1.7.1

All changes are classified as
- New feature: New functionality has been added to the extension
- Improved: A feature was changed to work better than in the previous version
- Bugfix: A bug in the previous version was fixed
- Changed: Something has changed due to a design decision

---------- version 1.7.1 ----------

- Bugfix: Lists that does not start with the top level are now indented correctly

- Bugfix: Removed the no longer functional checkbox "Ignore table dimensions" from options (custom configuration)

- New feature: The HTML export dialogs offers two new configurations: "Original formatting with navigation panel" and "Clean HTML".

- Changed: The standard option cleanxhtml.xml has been reworked to not include options that are available in the export dialog. Also, the
  content has been reordered to match the order in the configuration.

- Improved: In exported names (e.g. CSS class names), non-ASCII letters were previously ignored. Now they are replaced by the
  corresponding base character making names more readable. Also the Danish letters ae and o with stroke are replaced by ASCII characters.

- New feature: New option filenames to define the file naming scheme when file splitting is used.
  Possible values are values name-number (default, use base file name+number), name-section (use base file name+heading/sheet name)
  and section (use heading/sheet name)

- Improved: Page references now inserts the "real" page number, that is the file number in the sequence (only
  relevant if split_levels>0)

- Bugfix: Custom templates and style sheets now also work for spreadsheets

- Improved: Panels now include all indexes (previously only table of contents and alphabetical index)

- Improved: Panels now include footnote and endnote pages. Title is either automatic based on the document language
  or taken from the options footnotes_heading and endnotes_heading

- Bugfix: Now only includes header or footer if the corresponding id in template_ids is non-empty

- Bugfix: Now always uses the first global table of contents for the navigation bars

- New feature: Added support for rotated text, letter spacing and hyphenation

- Improved: Extended support for text decoration (underline, strike through and overline) and removed support for
  blinking text (deprecated in CSS)

- Improvement: Calculation of list indentations and label widths are now correct also in these two cases:
  (1) The list style is attached to a paragraph style.
  (2) Hard paragraph formatting is applied to the list items.

---------- version 1.7 ----------

- Changed: Removed old experimental code to export presentations to HTML

- Improved: Export of list formatting is now using modern CSS features like counters.

- Changed: The option list_formatting now supports the values ignore_all (do not export list formatting),
  convert_labels (only convert the labels, not indentations), convert_all (convert all list formatting).
  The old values css1, css1_hack and hard_labels have been removed.  

- Bugfix: Fix crash in Calc export when last row/column in view is hidden

- Improved: No longer uses deprecated Java constructions

- Improved: Pretty printing now works better with spreadsheets.

- Bugfix: The option use_mathjax now works correctly with TeXMaths equations. Also TeXMaths equations are now always
  typeset in inline mode.

- Bugfix: The converter now always uses the current page layout when calculating relative widths. Previously the first
  page layout in the document was used, which caused miscalculations if page layout changes in the document.

---------- version 1.7 beta ----------

- Bugfix: The converter no longer crashes on spreadsheets with diagrams.

- New feature: New boolean option original_page_numbers. If this option is set to true, the original page numbers will
  be included in EPUB export. This feature is currently experimental.

- Improved: The EPUB filter dialog has been extended to support the options table_size (previously only available in
  the custom configuration settings), units (replacing convert_to_px) and the new option original_page_numbers.

- Changed: The configuration file cleanxhtml.xml no longer sets custom_stylesheet.

- Improved: The HTML filter dialog in Writer has been extended and redesigned to a two column layout. The options
  multilingual, embed_svg (HTML5 only), embed_img, table_size and formulas (XHTML 1.0 and 1.1 only) can now be set
  in the filter dialog, and are thus removed from the configuration settings. Also all three values for the options
  table_size and image_size can now be set in the filter dialog.
  The same applies to the HTML filter dialog in Calc, which now supports the options embed_svg (HTML5 only),
  embed_img and image_size. (The other aforementioned options multilingual, table_size and formulas are not relevant
  for spreadsheet documents.)
  
- Changed: Removed localizations of filter dialogs (only some strings were translated anyway)

- New feature: New option units with values rem (default), px and original. This replaces the options convert_to_px,
  which corresponds to the values px and original. For backwards compatibility, convert_to_px is still accepted.

- Changed: The option formulas (for XHTML 1.0 and 1.1 strict only) no longer supports the values latex and image+latex

- Bugfix: The configuration option formulas=image+starmath now works correctly if embed_img=true

- Improved: Simplified export of column widths in tables (the old method set width on table cells rather than table
  columns. This was due to differences in browsers, and should not be relevant anymore).
  
- Improved: No longer sets absolute table width and the CSS attribute table-layout:fixed when exporting Writer documents.
  The old behavior was too strict and could lead to overfull table cells. On the other hand when exporting Calc
  documents the two attributes are now both set, because overflow of table cells is o.k. in a spreadsheets document.

- New feature: Added support for the ODF attribute table:border-model (collapsing or separating borders)

- Bugfix: The toolbar settings are now also visible in Calc.

- Improved: Added Help button to all configuration dialogs (because the "official" help button leads to online help).

- New feature: Added support for image bullets in lists.

- Bugfix: The converter no longer crashes if embed_img=true and the document contains two identical pictures.

- Improved: Removed the redundant attribute type from <script> if producing HTML5.

- Bugfix: Removed the deprecated profile attribute (Dublin Core) from <html> if producing HTML5.

- Bugfix: Removed the deprecated type attribute from <style> if producing HTML5.

- Bugfix: The <title> must be non-empty. The file name is now used as title if the document has an empty title. 

- Changed: The batch conversion feature has been removed. Also removed the two associated options directory_icon and
  document_icon (uplink is kept because it might be useful even without batch conversion).

- Changed: Writer2xhtml has been separated from Writer2LaTeX: All LaTeX code has been removed, and packages etc. has
  been renamed to avoid conflicts if installed together with Writer2LaTeX.