From: William P. <wil...@ya...> - 2010-02-24 03:43:58
|
On Feb 22, 2010, at 11:48 AM, Vladimir Gapeyev wrote: > I am setting up an SQL script to update TI info, following your suggestions in an earlier email (below). Besides what you mentioned there, would you want to update table fields taxonvaiant.tb1legacyid and taxon.tb1legacyid? In the case of the taxon_labels, there are actually two possible "legacy" ids. One is the TB1 legacy id (which, as you note, begins with a T), and the other is an integer that I use for the mapping between the tables -- this is, strictly speaking, not part of TB1 but part of an intermediate step in the process of mapping the taxon labels to the taxonomy. I guess I would prefer it if you would override the auto-increment for the taxonvariant_id field and the taxon_id field so that they would use my integers instead -- then ignore or drop the "tb1legacyid" fields in the taxonvariant and taxon tables. But if you'd rather not do that, then allow your fields to auto-increment and instead use my integer ids in the tb1legacyid fields of the taxonvariant and taxon tables. TB2's taxonlabel table does not have a tb1legacyid field, so we don't have a place to store the "T98303" legacy_taxon_label values. But that's okay -- the only person who might want to know what the new ids for TB1 legacy_taxon_labels is Rod Page, and I can email to him so conversion tables. So don't worry about trying to store the T98303 values. But please try to store the first columns in my taxon_variant and taxon files, either by suspending autoincrement and putting them in the equivalent fields in TB2, or by putting them in the integer-typed tb1legacyid fields. > Also, I'll set 1 in the 'version' field in all 3 tables, unless there is different advice. Sounds fine. thanks, bp |