On 1/13/14 9:21 AM, "big stone" <stonebi...@gmail.com> wrote:
>Hello, > >I read the source code of sqlite and I find a lot of : > > >CREATE TABLE t1(x INTEGER); >INSERT INTO t1 VALUES(1);INSERT INTO t1 VALUES(2); >.... > >Wouldn't it be both an economy of coding time and a more quick upload >process to allow Multiple inserts SYNTAX ? > >like : >INSERT INTO t1 VALUES(1),(2), .... (9999); > >I'm not 100% certain it's in SQL standard, but the syntax seems ok for >SQL Server , Oracle , and POSTGRESQL at least. I can't speak to the syntax; but in my experience, preparing a single INSERT statement and executing it multiple times in a loop is much faster than building a script of many INSERT statements and then preparing and executing it once. I would imagine the same thing would apply for a multi-row INSERT: for large amounts of data, the parsing time would kill you. Will >_______________________________________________ >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
