On Wed, Dec 22, 2010 at 8:59 AM,
The general request (composing partial indices to cover the whole sequence) is interesting and a considered it some years ago, though I never found the time to implement it (and I'm suspicious it might be a little too "smart" for the general public).
Sorry to wake up an old thread, but I've just encountered a problem where partial indices would be the perfect solution. I'm working on a problem where I have several unique identifiers for each object, but I'm not guaranteed that I have all of them at the same time. I can get updates from two sources, which don't necessarily don't use the same ID, yet they will sooner or later be linked. Using postgresql a simplified version could look like this: CREATE TABLE test( a text, b text, data text); CREATE UNIQUE INDEX test_a_par_uidx ON test(a) WHERE a IS NOT NULL; CREATE UNIQUE INDEX test_b_par_uidx ON test(b) WHERE a IS NOT NULL; Are we thinking of the same sort of partial indices? ... and is this something that is feasible to implement in multi_index_container? -- Eld på åren og sol på eng gjer mannen fegen og fjåg. [Jøtul] <demo> 2011 Tore Halvorsen || +052 0553034554