Add an API endpoint which permits writing to review files
Currently, when a query is beyond the capabilities of Create Lists, it is usually possible to use a SQL query with Python to produce a report. Such a report might list record numbers of bibs/items/orders/patrons/etc which bear further examination and correction by library staff, but staff must copy record numbers or barcodes in the report and paste them into the SDA to pull up the records in question. It would be much easier if an API endpoint permitted those record numbers to be written to a review file so staff can go through them one-by-one in the SDA.
Another use case for such an endpoint would be the development of record import routines with features which do not exist in the current Create Lists record import facility.
Bob Gaydos
shared this idea
-
Bob Gaydos
commented
In response to Jeremy's comment - thank you, Jeremy - my vision of writing to review files was more simplistic and does not require the dynamic flexibility of programmatically locating an open review file to use. Instead, I imagined using one or more existing review files dedicated to the purpose of the reports mentioned as the primary use case in my first paragraph above.
Granted, a robust record import mechanism external to the SDA or other related applications could absolutely benefit from adding the missing features Jeremy points out, so please add "capacity" (bool_info.max) to the /v6/reviewFiles GET request as well as a new parameter named something like "showUnusedFiles".
-
Jeremy Goldstein
commented
I absolutely want this feature.
The one complication I see that would have to be resolved is that there would also have to be a mechanism for identifying available files that could be written to. Currently the reviewFiles endpoint will only return a list of files that are in use...so you could take that and I suppose just know to fill in the blanks, but that's hardly ideal. That endpoint also does not return the capacity for the files, just how many records are currently contained within one. This proposed feature would work a lot better in conjunction with the ability to search for empty review files of x size vis the API, so that it could then claim the appropriate file to write to.