v2.4.12

Changes:

• Increased shortened_word_filter_percentage from 60% to 70% when matching through alternative matching (cover matching) a9bebd2
• Removed various colons from various discord embed messages for consistency a9bebd2
• Added an lru_cache to pick_webhook() a9bebd2
• Fixed the renamed folder not being added to transferred_dirs after rename a9bebd2
• Fixed rename issue with a multi-chapter release that starts with 0 a9bebd2
  • EX: series_name 000-025 - S1_P1 (2020).cbz
• Preparation for future feature a9bebd2
• Misc changes a9bebd2

v2.4.11

Changes

• Doubled sleep time for 104 error reattempt in get_komga_libraries() ce000a0
• Fixed only one library scan request being sent when multiple libraries had something added ce000a0

v2.4.10

Changes:

• Added thread locking to prevent simultaneous executions 83a1705
  • EX: Watchdog picks up files in folder 1, starts execution, then in the middle of folder 1's execution, watchdog picks up new files in folder 2 and starts executing. Now when watchdog picks up new files in folder 2 while folder 1 is still executing, it will be forced to wait until folder 1 is finished executing.

v2.4.9

New:

• Added multi-download-folder support to watchdog. (previously it only monitored the first path in the array) acd3941
  • pass in each one with -df in the same way you do with -p for paths, also in the same way for docker as you do with paths
• Added download_folders_with_types array and additional path classifications for a WIP feature acd3941
• Added index_number to the volume class acd3941
  • ex: volume 1 part 3 = 1.3
• Added a reattempt after a 15 second sleep on komga library pull when error 104 has occurred acd3941
  • common when komga is still starting up, but the script is already up and fails to request the libraries because komga hasn't finished starting up.

Changes:

• Cleaned up process_path() code and added additional checks for WIP feature acd3941
• Alternative cover matching now attempts on all series' files if 10 or less, regardless of volume_number and volume_part acd3941
  • a matching volume_number and volume_part are naturally at the front of the array and attempted first
  • allows a volume that uses a different naming standard to match, but still maintains accuracy as the alternative matching uses cover matching, and the wrong volume wouldn't have the same cover
    • ex: Ascendance of a Bookworm v05.3 can now match to Ascendance of a Bookworm Part 05 v03
• Improved volume/chapter-part named file matching and duplicate volume/chapter removal with index_number acd3941
  • ex: series_name v01 [Part 1] can now match to series_name v01.1
  • and series_name c05#5 to series_name c05.5
  • and series_name c05x5 to series_name c05.5
• Renamed path_types in Path class to path_formats acd3941
• Removed caching from set_num_as_float_or_int() acd3941
• Misc code changes acd3941

v2.4.8

REQUIRED:
   1. Manually make change to settings.py or update to new settings.py 992a9e9

New:

• Added experimental and optional auto-classification of watchdog paths 99786bf 992a9e9

Changes:

• User path processing cleanup 99786bf
• Removed multi-volume matching single volumes functionality 99786bf
• Misc cleanup 99786bf
• Misc changes 99786bf

v2.4.7

Changes:

• Download files' zip comments are now excluded from the cache
  • You could have a download file that doesn't match through similarity or alternative match (cover match), but could still match through an identifier match (zip comment), so caching a download file's (likely empty) zip comment would prevent such a match until a restart happened. The existing libraries zip comments however are far less likely to change as frequently as a file in the download folder. So a restart for those changes is acceptable.
  • With my metadata injection script that writes the identifier zip comment and other data still private, this change only really pertains to me. At least for now.

v2.4.6 - 7z Conversion Support

REQUIRED:
   1. Update and run new requirements.txt 3b7efdd

New:

• Added 7zip support to convert_to_cbz_toggle feature 4a34b28
• Added --watchdog_discover_new_files_check_interval and --watchdog_file_transferred_check_interval command-line arguments. 4a34b28 a392f43

Changes:

• Fixed failed removal of an improperly repacked archive in convert_to_cbz() 4a34b28
• Misc code cleanup 4a34b28

v2.4.5

Changes:

• Komga libraries are now pulled once at the beginning of the script for watchdog users. e9750f4
  • instead of pulling it every time we go to trigger a scan
• Improved overall script speed through the use of lru_cache e9750f4
  • most noticeable for watchdog users on subsequent runs
• Rewrote get_file_hash() and get_internal_file_hash() to read the file in buffers e9750f4
• Adjusted the two komga-related functions to let the user know when they're missing a required setting. e9750f4
• Code cleanup e9750f4
• Misc changes e9750f4

v2.4.4 - Chapter Rename Improvements pt.2

New:

• Added test_mode to rename_files_in_download_folders() 78ecca2
  • allows it to work with a list of filenames instead of actual files. (this one is for me)

Changes:

• Further improvements to chapter rename compatibility 78ecca2
• Improved alternative matching 78ecca2
• Improved overall script speed through the use of lru_cache 78ecca2
• Misc changes 78ecca2
• Misc fixes 78ecca2

v2.4.3 - Hotfix

Changes:

• Fixed regression in _extra renaming a0960d3