http://bugs.koha.org/cgi-bin/bugzilla/show_bug.cgi?id=2691 ------- Comment #6 from joe.atzberger@liblime.com 2009-02-02 16:29 ------- OK, this is a partial description of the problem, but only the "after" (i.e. target) state. What we need are mappings from the single-string "itemcallnumber" representation to the appropriate separate parts. For example, here are some test cases I propose: HE8700.7 .P6T44 1983 => HE 8700.7 P6T44 1983 BS2545.E8 H39 1996 => BS 2545 .E8 H39 1996 BS2545.E8 .H39 1996 => BS 2545 .E8 H39 1996 NX512.S85 A4 2006 => NX 512 .S85 A4 2006 Are those correct mappings? Are they even "valid" LCCN to begin with? Note the "H39" vs ".H39" difference. What characters or sections should get "dropped"? There does not appear to be any reliable rule that would apply to all libraries. --Joe (In reply to comment #3)
This from a cataloger who uses LC: ....
BJ466 .S7 1986 Copy 1
BJ466 .S7 1986 Copy 2
BJ319 .A46 1989 Copy 1 Asian Japan
BJ455 .B3S6 1986 vol.3 Copy 2 MRR Alc
BJ455 .B3S6 1986 Suppl. Copy 3 Sci RR
When volume numbering is necessary, use the abbreviated English designations vol., pt., and no. for volume, part, and number (cf. G 615).
You can see that here we did not follow LC exactly. We drop the numeric portion of the class number, for example:
BJ 466 .S7 1986 Copy 1
------- You are receiving this mail because: ------- You are the QA contact for the bug, or are watching the QA contact.