* Shared now exports fetchItem (instead of getItem) and openURL
* fetchItem has different parameters than getItem and includes
some logic formerly in the ODT and Docx writers
* getItem still used in SelfContained
Previously header ids were autogenerated by the writers.
Now they are generated (unless supplied explicitly) in the
markdown parser, if the `header_identifiers` extension is
selected.
In addition, the textile reader now supports id attributes on
headers.
* Added `embed_data_files` flag. (not yet used)
* Shared no longer exports `findDataFile`.
* `readDataFile` now returns a strict bytestring.
* Shared now exports `readDataFileUTF8` which returns a string like
the old `readDataFile`.
* Rewrote modules to use new data file functions and to avoid
using functions from Paths_pandoc directly.
* Previously there were three different templates involved in
epub production. There is now just one template, default.epub
or default.epub3.
* It can now be overridden using `--template`, just like other
templates.
* Here we go back to splitting the document at the Pandoc AST
level, and converting each chapter to HTML separately.
* Fixes bug in footnote processing introduced by
commit 01d109e2ef. The bug caused
all footnotes to appear in the last chapter, so that footnote
links from other chapters would not work. Closes#651.
* We also disable writerTableOfContents, since TOC is automatic
for epub.
* MathML math method now used always in EPUB3 (even if another
math method specified in options).
* epub:switch is used to specify a non-MathML default. This is supposed
to proide a good fallback behavior in older readers, though I'm
not sure how well it works in practice.
* We now convert to XHTML before cutting into chapter-sized chunks.
This fixes a number of problems.
* `--number-sections` now works properly.
* A proper three-level table of contents is now used in `toc.ncx`.
There is no longer a subsidiary table of contents at the beginning
of each chapter.
* New epub-page template without the `$title$` variable. Titles are
left in the chapter bodies as an initial h1.
* Closes#539.
* Depend on text.
* Expose Text.Pandoc.UTF8.
* Text.Pandoc.UTF8 now exports toString, fromString,
toStringLazy, fromStringLazy.
* These are used instead of the old utf8-string functions.
Otherwise we have a hard-coded English string, which looks
strange in ebooks written in other languages. Using the
title for this entry seems a good solution. Closes#572.
These now take WriterOptions and Pandoc only.
The extra parameters for epub stylesheet, epub fonts,
reference Docx, and reference ODT have been removed, since
these things are now in WriterOptions.
Note: breaking API change!
Previously they were ignored. Now all links are preserved,
but purely internal links are modified so that they point
to the proper place in the EPUB.
This is nontrivial, since the heading you refer to in your
markdown source with 'my-section-1' might end up as
'ch16.xhtml#my-section' in the EPUB.
Closes#76.
kindlegen doesn't like them - even '.
It should be safe to use the unescaped ' character, since
we know that all attributes are double quoted in the relevant
files.