Hi, Using the latest amalgamation build sqlite-autoconf-3080801 I'm seeing the following inconsistent behaviour:
$ ./sqlite3 sqlite> .headers on sqlite> pragma short_column_names; 1 sqlite> pragma full_column_names; 0 sqlite> create table tbl1 (id1 integer, s1 text); sqlite> create table tbl2 (id2 integer, s2 text); sqlite> insert into tbl1 values (1, 'v1'); sqlite> insert into tbl2 values (1, 'v2'); sqlite> select x.id1, x.s1, y.s2 from tbl1 x inner join tbl2 y on x.id1=y.id2; id1|s1|s2 1|v1|v2 So far so good, everything as expected. If I rewrite the above select statement to do a join with a subquery instead the resulting output changes in an unexpected way. sqlite> select x.id1, x.s1, subq.s2 from tbl1 x inner join (select * from tbl2 y where y.id2=1) subq on x.id1=subq.id2; x.id1|x.s1|subq.s2 1|v1|v2 Here we get unexpected column prefixes on all fetched columns. If I rewrite the query again to a subq.* query the behaviour is different again where only the first two columns have prefixes. sqlite> select x.id1, x.s1, subq.* from tbl1 x inner join (select * from tbl2 y where y.id2=1) subq on x.id1=subq.id2; x.id1|x.s1|id2|s2 1|v1|1|v2 Expected behaviour: returned columns should not contain prefixes in any of the above scenarios. Kind regards, Marcus Bergner _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
