Thread: Trouble with some concepts
Hi all, As announced previously I am currently working with other people on the translation of PostrgreSQL to French, applications and manuals. We have stumbled on a couple of concepts which are not very clear to us and thus do not know very well how to translate. Can someone explain the following: msgid "the target relation %u is nailed" One translator said that "nailed" was used for permanent network connection. Could this message mean that the relation is locked by a permanent connection to the database? msgid "improper qualified name (too many dotted names): %s" What's a qualified name? msgid "database name may not be qualified" Does this mean that the database with some given name cannot be found, or that the given name is not valid or something else entirely? msgid "" "cannot cluster on expressional index when index access method does not " "handle nulls" Is an expressional index an index built on a calculated field (say two concatenated field or whatever) or something else? Thanks for the help. --------------- Francois Home page: http://www.monpetitcoin.com/ "Would Descartes have programmed in Pascal?" - Umberto Eco
Francois Suter <dba@paragraf.ch> writes: > Can someone explain the following: > msgid "the target relation %u is nailed" It's referring to a nailed-in-cache index, which is to say something that we don't allow to drop out of the relcache (because we'd be unable to load it back in, because it's used during relcache entry loading). I think that this message is actually a can't-happen item because of upstream checks, so there's probably no need to consider it translatable. I'll look at turning it back to a plain elog. > msgid "improper qualified name (too many dotted names): %s" > What's a qualified name? Name with dots in it. The complaint would come out for something like a.b.c.d, which is syntactically illegal. > msgid "database name may not be qualified" As above. A database name cannot look like a.b. > Is an expressional index an index built on a calculated field (say two > concatenated field or whatever) or something else? Yes. "Expressional index" isn't a really good term --- it's something I invented on the spur of the moment to replace "functional index". If you have a suggestion for a better term, I'm all ears. regards, tom lane