DNB exports the ISBN with dashes in 020$9, which we import via SRU while cataloguing. Therefore we added a subfield 9 to field 020. But when 020 contains no data except the dashed ISBN in $9, the value of $9 is deleted. I assume this is caused by some special cases needed to delete the koha_id of an authority (which is often stored in $9). To reproduce: * Go to a framwork, field 020, add a subfield 9, visible in tab 0, OPAC, Intranet and edito * Go to a book, edit it, enter something in 020$a and 020$9, save * The values are stored * Edit again, remove the value from 020$a, save * the value from 020$9 is gone, too I can maybe take a look tomorrow to try to figure out what's happening..
Created attachment 139043 [details] A valuebuilder workaround Here is a valuebuilder that works around that problem by autofilling 020$a with ISNB without dashes after a value is entered into 020$9 (ISBN with dashes) Not optimal, but better than loosing data... Copy the file into `intranet/cgi-bin/cataloguing/value_builder` and enable it as a plugin to 020 subfiled 9.
I am wondering - doesn't the DNB also deliver 020$a without hyphens in addition to $9? At least that is the case for K10plus.