Update on November 22, 2019: At yesterday's RA SIG meeting it was agreed that this feature does not need to be in the MVP. For the MVP we will sort the Call Number column of FF History the same as the other columns.
Update on November 18, 2019: Inventory is creating and saving a normalized call number. Fees/Fines will use that call number for sorting when it is available. We don't need to do our own normalizing.
This Comment from January 25, 2019 was moved from
UXPROD-393 to this issue:
Holly met with Frances Webb to discuss how Cornell Library has addressed the call number sorting issue. Here are Holly's notes from the meeting:
The call number starts with a letter or letters, and then a number or numbers follow. Have to sort alphabetically due to the letters, but this messes up the numbers. For example, the sort results in this…
When it should be this…
Need to sort the letter(s) alphabetically, the number(s) numerically, then the rest alphabetically to get to this correct order…
Some fields needs to be ignored when sorting, like prefixes (Rare Books, Oversized, Archives, etc.). Frances has a file of prefixes to ignore. How will this work for FOLIO? Will we have to let tenet set what to ignore?
Solr will make a normalized call number based on a customer sort filter Frances wrote. All punctuation is cleaned up except for decimal points (periods with a number right before it and right after it) and prefixes in the file mentioned above are excluded. For example...
INPUT: Oversize ML410.1.M619 G56.1 2012 ++
OUTPUT: ml 000000410.1 m 619 g 56.1 2012
The output string is then simply sorted alphabetically.