* 0.7.42 is a re-release of 0.7.41, because conversion to MOBI was broken in 0.7.41
* Conversions: Replace the remove header/footer options with a more geenric search replace option, that allows you to not only remove but also replace text
* Conversion: The preprocess html option has now become a new ‘Heuristic Processing’ option which allows you to control exactly which heuristics are used
* Conversion: Various improvements to Heuristic Processing (used to be preprocess HTML)
* When adding empty books to calibre, optionally set the author to the author of the currently selected book
* Device drivers for the Archos 101, SmatQ T7 and Acer Lumiread
* Catalog generation: Make By Authors optional
* Allow bulk editing of Date and Published columns.
* Add a little button to clear date and published values to the edit metadata dialogs
* When adding books by ISBN, allow the specification of special tags that will be added to the new book entries
* Completion on multiple authors
* Add AZW to default list of internally viewed formats, a I am tired of getting tickets about it
* Nicer error message when catalog generation fails
* Add capitalize option to context menus in the edit metadata dialog
Bug Fixes
* RTF Input: Fix regression in 0.7.40 that broke conversion of some old style RTF files
* Fix Tag editor forgets position
* When converting books in the calibre GUI, override metadata from the input document, even when empty.
* Fix memory leak when switching libraries
* RTF Output: Fix incorrent spacing between letters.
* Catalog generation: Add composite columns to Merge Comments eligible types
* Add a confirmation when closing the add a custom news source dialog.
* Another workaround for LibraryThing UA sniffing that was preventing series metadata download, sigh.
* PD Novel driver: Put books on the SD card into the eBooks folder
* When shortening filepaths to conform to windows path length limitations, remove text from the middle of each component instead of the ends.
* Make completion in most places case insensitive
* Fix regression that caused the N key to stop working when editing a Yes/no column
* Email: Fix bug when connecting to SMTP relays that use MD5 auth
* MOBI Output: Fix bug that could cause a link pointing to the start of a section to go to a point later in the section is the section contained an empty id attribute
* When auto converting books and the device is unplugged, do not raise an error.
* Ebook-viewer: Display cover when viewing FB2 files
* MOBI Input: Special case handling of emptu div tags with a defined height used as paragraph separators.
* Fix sorting of author names into sub categories by first letter in the Tag Browser when the first letter has diacritics
* Fix regression in 0.7.40 that caused commas in author names to become | when converting/saving to disk
* Fix view specific format on a book with no formats gives an error
I can’t help but wonder if this is still broken. I was successfully doing epub-to-mobi conversions yesterday morning, using 0.7.40 but this morning, the conversion started failing. Thinking perhaps Calibre had updated itself to 7.41 but not to 7.42, I went and manually updated to 7.42. Still no mobi conversions.
I’m poking around among the bug reports over there, but I thought I’d bring it to your attention.