3 This document records all notable changes to Librarian.
10 - Support for `ref` tags in HTML.
17 - Experimental class-based builders for HTML and TXT.
22 - Upgraded to EPUB 3, using ebooklib.
29 - XML entities in MathML breaking PDF generation.
30 - Regression: verse numbering was lost.
36 - New Ebookpoint logo.
42 - Multiple authors on cover are now split into lines.
48 - Support for tags: podtytul_*.
49 - Swappable CSS for HTML preview (as --css).
50 - First version of a test file in tests/uat.
53 - Verses in HTML are now generated as div, not p, to prevent blank lines on copying.
59 - Preliminary support for new tag: animacja.
65 - Allow multiple source.URL fields.
70 - Epub: only attach images referenced in the text.
76 - Updated information on the back of EPUB and PDF files.
82 - `html.tranform_abstract` bytes vs text confusion.
89 - Support for wl:coverLogoUrl, for adding cover logos.
95 - Fix for Picture in Python 3: open image in binary mode.
101 - Bug in cover generation.
107 - Python 3.4+ support, to existing Python 2.7 support.
108 - `coverter_path` argument in `mobi.transform`.
109 - Proper packaging info.
111 - Tox configuration for tests.
114 - `from_bytes` methods replaced all `from_string` methods,
115 i.e. on: OutputFile, WorkInfo, BookInfo, WLDocument, WLPicture.
116 - `get_bytes` replaced `get_string` on OutputFile.
119 - Shims for Python < 2.7.