2.0.1

Added

• User documentation now has an example showing how to install a library in Python
• The user interface now checks if the input file, the output folder and the log folder exist or can be created
  • If one of them does not exist, the user will stay in the main configuration window and an error pop-up will open
  • If one of the folder was created, the information will be written in the error pop-up or an informative pop-up (if there was no error)
• The main function now properly checks and displays in the terminal (and logs if available) if the input file, output folder or log folder do not exist or can not be created

Fixed

• User documentation now correctly uses BETTER_ITEM_MAPS instead of BETTER_ITEM_NO_MAPS
• Processings BETTER_ITEM_DVD & BETTER_ITEM_MAPS do not look anymore only for dates in the origin database physical description (method __special_better_item())
  • This behaviour is now an opt-in, the default one is keeping the data raw

2.0.0

Added

• Documentation now states which library versions were used for development (due to issues with PySimpleGUI in this project and pymarc in another project)
• Now uses pysisbn library
• Added error ISBN_979_CAN_NOT_BE_CONVERTED for ISBN 13 not starting with 978 if trying to convert them to ISBN 10
• Operations SEARCH_IN_SUDOC_BY_ISBN, SEARCH_IN_SUDOC_DVD, SEARCH_IN_SUDOC_NO_ISBN & SEARCH_IN_SUDOC_MAPS now have an additionnal last action that queries only on title on the title index (and document type filter)
• More specific erros were added if the title, the authors, the publisher or the dates were missing and they were required for the matching process and if the configured document type is not supported

Changed

• Now developped for :
  • Python 3.12 (3.11 before)
  • unidecode 1.3.8 (1.3.6 before)
  • python-Levenshtein 0.25.1 (0.23.0 before)
  • requests 2.32.3 (2.28.1 before)
  • python-dotenv 1.0.1 (1.0.0 before)
• Changed graphic user interface library from PySimpleGUI 4.2.2 to FreeSimpleGUI 5.1.0
• Light grey color was replaced by creme
• Replaced ISBN functions from Abes_id2ppn by pyisbn library
• Updated Abes_id2ppn with lastest version
• Splitted big files into smaller ones & reduce dependencies between them
  • Notably, cl_UDE.py can now be used alone as it was originnally intended
  • Added a classes_dependencies.md file explaining the dependencies
• Reworked request_action to avoid redundacy on Sudoc SRU, Koha SRU & Sudoc's id2ppn webservice
• Documentation (developper & user) was updated (user english documentation was added)

Removed

• Removed unecessary personnal Python test scripts and files

Fixed

• Put back the mention of Alexandre Faure original script (that I apparently deleted)
• Fixed various syntax warning with regular expressions
• Fixed fatal errors when trying to convert ISBN with more than just digits or X
• Fixed the functions returning instances from Enum (though the bug was not used in the main script I think)
• Fixed the interface not loading the saved language properly

1.17.5

Fixed

• Universal_Data_Extractor.extract_list_of_strings method does not crash anymore if some subfield had no values
• ASCII range (hexadecimal) 21-2F, 3A-40, 5B-60, 7B-7F added to the noise list

1.17.4

Added

• New action requesting Koha SRU only using title on the title index

Changed

• The operation SEARCH_IN_KOHA_SRU_VANILLA now queries as a last resort Koha SRU using only the title

1.17.3

Added

• Transformations before request actions using titles, authors or publishers now deletes every word that does not start with a letter or a number

Changed

• Reordered some columns in CSV output :
  • Document type is now between erroneous ISBN and titles
  • Authors are now between document type and titles
  • Matched IDs, database ID and current ID are moved after FCR prcessed ID

Fixed

• Added more dahs to the noise deletion process (Unicode 2010 to 2015)

1.17.2

Added

• New exportable data : piece (UNIMARC 463, internal name is linking_piece) and document type
• New action : query title, publisher and publication date in the any index in Koha SRU (added to SEARCH_IN_KOHA_SRU_VANILLA operation)

Changed

• Processing MARC_FILE_IN_KOHA_SRU now export :
  • Physical descriptions for both records
  • Items barcodes for both records
  • Document type for both records
  • Piece for both records
• Koha SRU now uses Filter 1 to filter items and items barcode information
• Mapping KOHA_ARCHIRES now uses b as filtering subfield for items and items barcode
• Added collectif & collectifs as empty words to delete

Fixed

• Added ° to noise list

1.17.1

Fixed

• Added UN to the lsit of empty words deleted
• Added + to the list of deleted noise
• Logs now properly write the target database value instead of the origin database twice

1.17.0

Changed

• JSON output exports much mor eprecise data, including raw data from records

Fixed

• Lists containing one element and empty elements are now properly output in CSV export

1.16.2

Fixed

• CSV output does not replace D by spaces anymore

1.16.1

Fixed

• List of list should not crash the CSV output function anymore