# Change Log This document records all notable changes to Librarian. ## 24.9 - Remove old API for HTML, TXT generators. Move to new API. - The HTML format is changed: - text-elements now have 'wl' class and id attribute, - magic 'secN' ids are removed - 'a.target' elements removed - numbering anchors class changed to wl-num - Removed all code related to art gallery. - Drop Python 3.7. ## 24.5 - Smaller images in EPUB (600px width instead of 1200px). - Convert PNG to JPEGs if too large in EPUBs. ## 24.4 - Add fundraising inserts in PDF. ## 24.1 - Added cover logos support. ## 23.12 - Added debug version for epub and epubcheck utility. ## 23.10 - Remove direct verse styling from HTML. ## 23.08 - Move statistics counter to L2 WLgit lDocument. ## 23.07.1 - Add . - Support Python 3.7--3.11. ## 2.6.1 - Fix for better ignoring . ## 2.6 - Change default cover to marquise. - Add support for full predesigned covers to marquise. - Remove support for changing actual cover class via coverClass. ## 2.5.1 - Bugfix release. ## 2.5 - Add html-snippet builder. - Remove DateValue class. - Fix some texts and tests. - Drop Python < 3.6. Up to 3.9 is supported. ## 2.4.13 - Added thema meta field. ## 2.4.12 - Fix for marquise cover: allow scaling title text in all layouts. ## 2.4.11.1 - Added assigning and preserving id attribute . ## 2.4.10 - Added , , and . ## 2.4.9 - Added verse counters to document statistics. ## 2.4.8 (2022-07-23) # Changed - Shortening authors and translators list on new cover. - Update lxml. ## 2.4.7 (2022-07-18) # Added - Metadata types. # Fixed - Don't validate comment contents. - Ignore translators on new cover if too long. ## 2.4.6 (2022-07-15) # Fixed: - Bug in reading child books. ## 2.4.5 (2022-07-08) # Fixed - Bug in reading picture XML. - Bug in setting WLURI.slug ## 2.4.4 (2022-07-07) ### Fixed - Packaging error. ## 2.4.3 (2022-06-07) ### Fixed - Fix for new cover with custom size. ## 2.4.2 (2022-05-23) ### Fixed - Error reporting on metadata parse error. ## 2.4.1 (2022-05-12) ### Added - `category.legimi` metadata field ### Changed - Fix PDF support for 'Geometric shapes' characters. - Explicitly ignore unknown elements in v2 elements API. ## 2.4 (2022-05-06) ### Added - New 'marquise' cover style. - 'Factory' cover style with a QR code. ## 2.3.5 (2022-04-07) ### Fixed - Support for www tag. - Support for mat with inline text. ## 2.3.4 (2022-02-28) ### Fixed: - Document stats totaling bug. - Legimi cover background position. ## 2.3.3 (2021-12-23) ### Added: - Line numbering reset with numeracja. ## 2.3.2 (2021-12-21) ### Fixed: - Legimi cover colors. - EPUB/MOBI TOC styling. ## 2.3.1 (2021-12-20) ### Fixed: - Additional logos order. ## 2.3 (2021-12-20) ### Added: - New MOBI builder. - Document statistics. - Legimi cover classes. ### Changed - Default licensing info. ## 2.2 (2021-10-07) ### Added - Support for block annotations. - Option to use endnotes in PDF. ### Fixed - Avoid hanging conjunctions on covers. ## 2.1 (2021-07-08) ### Added - Basic document structure validation. - Support for headers inside quotes in epub. ## 2.0 (2021-07-08) ### Added - New Document API. - New EPUB builder. - New `librarian` entry point for converters. ## 1.15 (2021-03-02) ### Fixed - Epub: translator marked with MARC Relators role 'trl'. - Epub: authors and translators dc:creator elements had duplicate `id`. - Epub: 'Start' element in TOC was added inconsistently and in wrong order. ## 1.14 (2021-02-05) ### Changed - Image sources are now URLs. This changes the API: instead of paths given as `ilustr_path`, `transform` functions now accept a new `base_url` parameter. - Size limits introduced for images in all formats. ## 1.13 (2021-01-27) ### Changed - Responsive images in HTML. This changes the html.transform API. ## 1.12 (2021-01-27) ### Added - Content warnings. ## 1.11.3 (2021-01-25) ### Fixed - Missing items in EPUB TOC. ## 1.11.2 (2020-12-18) ### Fixed - MathML in EPUBs. - Subchapters without chapters in EPUBs. ## 1.11 (2020-12-09) ### Added - Experimental DAISY builder. ## 1.10 (2020-11-09) ### Added - Support for `ref` tags in HTML. ## 1.9 (2020-10-08) ### Added - Experimental class-based builders for HTML and TXT. ### Changed - Upgraded to EPUB 3, using ebooklib. ## 1.8.3 (2020-05-28) ### Fixed - XML entities in MathML breaking PDF generation. - Regression: verse numbering was lost. ## 1.8.2 (2020-04-03) ### Changed - New Ebookpoint logo. ## 1.8.1 (2020-02-18) ### Changed - Multiple authors on cover are now split into lines. ## 1.8 (2020-02-12) ### Added - Support for tags: podtytul_*. - Swappable CSS for HTML preview (as --css). - First version of a test file in tests/uat. ### Changed - Verses in HTML are now generated as div, not p, to prevent blank lines on copying. ## 1.7.8 (2020-02-05) ### Added - Preliminary support for new tag: animacja. ## 1.7.7 (2019-12-31) ### Changed - Allow multiple source.URL fields. ## 1.7.6 (2019-10-01) ### Changed - Epub: only attach images referenced in the text. ## 1.7.5 (2019-08-19) ### Changed - Updated information on the back of EPUB and PDF files. ## 1.7.4 (2019-08-01) ### Fixed - `html.tranform_abstract` bytes vs text confusion. - Tests. ## 1.7.3 (2019-07-31) ### Added - Support for wl:coverLogoUrl, for adding cover logos. ## 1.7.2 (2019-03-12) ### Fixed - Fix for Picture in Python 3: open image in binary mode. ## 1.7.1 (2019-03-01) ### Fixed - Bug in cover generation. ## 1.7 (2019-02-27) ### Added - Python 3.4+ support, to existing Python 2.7 support. - `coverter_path` argument in `mobi.transform`. - Proper packaging info. - This changelog. - Tox configuration for tests. ### Changed - `from_bytes` methods replaced all `from_string` methods, i.e. on: OutputFile, WorkInfo, BookInfo, WLDocument, WLPicture. - `get_bytes` replaced `get_string` on OutputFile. ### Removed - Shims for Python < 2.7.