On 29/03/17 01:35, Simon Slavin wrote:
On 28 Mar 2017, at 11:02pm, Mark Brand <mabr...@mabrand.nl> wrote:
create temp table t (db, val);
insert into t select 'temp', 'original';
create table t (db, val);
insert into t select 'main', 'original';
Here’s your problem:
SQLite version 3.16.0 2016-11-04 19:09:39
Enter ".help" for usage hints.
sqlite> create temp table t (db, val);
sqlite> insert into t select 'temp', 'original';
sqlite> .schema
CREATE TABLE t (db, val);
sqlite> create table t (db, val);
sqlite> insert into t select 'main', 'original';
sqlite> .schema
CREATE TABLE t (db, val);
CREATE TABLE t (db, val);
sqlite> SELECT * FROM t;
temp|original
main|original
sqlite> SELECT * FROM temp.t;
temp|original
main|original
sqlite> DROP TABLE temp.t;
sqlite> SELECT * FROM t;
sqlite>
While temp.t exists it gets in the way of main.t. When you refer to "t" you’re
talking about temp.t, not main.t.
The point isn't about which table one expects to receive the update,
it's that *both* tables get updated. In fact, now I realize that the
effect can be demonstrated with a simpler demo than my original:
create temp table t (db, val);
insert into t select 'temp', 'original';
create table main.t (db, val);
insert into t select 'main', 'original';
update t set val = 'touched';
select * from temp.t;
select * from main.t;
/* output
temp|default table
main|default table
*/
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
