Looks like feature bloat. If you need such features for a particular application just add the code to your application or create a modified version of Sqlite for just that application.
It makes no sense to burden all users with overhead for a feature required by only a handful of applications. Simon Slavin wrote: > On 21 May 2009, at 5:59am, Simon Slavin wrote: > > >> int sqlite3_change_count(sqlite3*) >> >> Returns an integer which is incremented whenn a change is made to any >> table in the database. May be the value stored in bytes 24..27 of >> file header or something else if that's not what I really want. >> > > Having thought about it some more, I want two different numbers. One > of them changes when a field changes: INSERT/DELETE/UPDATE. The other > changes when the schema changes: ALTER/DROP/CREATE. Rather than crowd > the function library with multiple functions, it might be possible to > implement it like so: > > int sqlite3_change_count(sqlite3*, int cTypes) > > when cTypes = 1, you get the number of changes to the schema > when cTypes = 2, you get the number of changes to fields > when cTypes = 3, you get the sum of the above two figures > > Or possibly 1 and 2 should be the other way around. > > Any ideas, criticism, or reasons it can't be done ? > > Simon. > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
