On Wed, Aug 21, 2013 at 8:04 AM, Dmitry Pashkevich <dpa...@gmail.com> wrote:
> I think I found a bug, tested with SQLite v3.7.15.2 and v3.7.9. > Looks like sqlite3_finalize() always returns SQLITE_OK, even if most recent > execution of prepared statement failed, which contradicts the > documentation<http://www.sqlite.org/c3ref/finalize.html> > : > > If the most recent evaluation of the statement encountered no errors or if > > the statement is never been evaluated, then sqlite3_finalize() returns > > SQLITE_OK. If the most recent evaluation of statement S failed, then > > sqlite3_finalize(S) returns the appropriate error code< > http://www.sqlite.org/c3ref/c_abort.html> > > or extended error code< > http://www.sqlite.org/c3ref/c_abort_rollback.html> > > . > > > > I'm using sqlite3 via lsqlite3 <http://lua.sqlite.org/> Lua wrapper > library > (so maybe the bug is there but I can't confirm). > I'm unable to reproduce the problem using C. Maybe it is in lsqlite3. My test case: #include <stdio.h> #include "sqlite3.h" int main(int argc, char **argv){ sqlite3 *db; sqlite3_stmt *pStmt; int rc; sqlite3_open(":memory:", &db); sqlite3_exec(db, "create table t(x unique);", 0, 0, 0); sqlite3_prepare_v2(db, "insert into t(x) values(?)", -1, &pStmt, 0); sqlite3_bind_int(pStmt, 1, 123); rc = sqlite3_step(pStmt); printf("first step returns %d\n", rc); sqlite3_reset(pStmt); rc = sqlite3_step(pStmt); printf("second step returns %d\n", rc); printf("error message = %s\n", sqlite3_errmsg(db)); rc = sqlite3_finalize(pStmt); printf("finalize returns %d\n", rc); sqlite3_close(db); return 0; } Output of the test program: first step returns 101 second step returns 19 error message = column x is not unique finalize returns 19 And this sort of thing is extensively checked in the SQLite test suites ( http://www.sqlite.org/testing.html), so it is difficult to imagine how something like this could go unnoticed. Hence, pending additional contrary evidence, I'm going to assume this is a problem with the lua bindings. > > Here's some sample code: > > > require "lsqlite3" > > db = sqlite3.open_memory() > > =db > sqlite database (0x238e858) > > =db:exec("create table t(x unique)") *-- create a table with a > constraint > * > 0 > > stmt = db:prepare("insert into t(x) values(?)") > > =stmt > sqlite virtual machine (0x23a23e8) > > =stmt:bind_values(123) > 0 > > =stmt:step() > 101 > > =sqlite3.DONE > 101 > > =stmt:reset() > 0 > > =stmt:step() *-- intentionally execute statement with same values to > violate the constraint* > 19 > *> =sqlite3.CONSTRAINT -- step() returns the expected error code* > *19* > > =db:errmsg() > column x is not unique > *> =stmt:finalize() -- finalize returns OK!* > *0* > *> =db:errmsg() -- error message is still persisted, though* > *column x is not unique* > > > Somebody please confirm this... > > -- > Dmitry Pashkevich > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > -- D. Richard Hipp d...@sqlite.org _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
