Re: SPI_ERROR_CONNECT within pl/pgsql, PG 8.4
От | Tom Lane |
---|---|
Тема | Re: SPI_ERROR_CONNECT within pl/pgsql, PG 8.4 |
Дата | |
Msg-id | 20196.1247584515@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Re: SPI_ERROR_CONNECT within pl/pgsql, PG 8.4 (Frank van Vugt <ftm.van.vugt@foxi.nl>) |
Ответы |
Re: SPI_ERROR_CONNECT within pl/pgsql, PG 8.4
Re: SPI_ERROR_CONNECT within pl/pgsql, PG 8.4 |
Список | pgsql-bugs |
I think the attached patch will fix it for you. regards, tom lane Index: src/backend/utils/cache/plancache.c =================================================================== RCS file: /cvsroot/pgsql/src/backend/utils/cache/plancache.c,v retrieving revision 1.27 diff -c -r1.27 plancache.c *** src/backend/utils/cache/plancache.c 11 Jun 2009 14:49:05 -0000 1.27 --- src/backend/utils/cache/plancache.c 14 Jul 2009 15:04:38 -0000 *************** *** 45,50 **** --- 45,51 ---- #include "access/transam.h" #include "catalog/namespace.h" #include "executor/executor.h" + #include "executor/spi.h" #include "nodes/nodeFuncs.h" #include "optimizer/planmain.h" #include "storage/lmgr.h" *************** *** 502,509 **** --- 503,521 ---- { /* * Generate plans for queries. + * + * The planner may try to call SPI-using functions, which causes + * a problem if we're already inside one. Rather than expect + * all SPI-using code to do SPI_push whenever a replan could + * happen, it seems best to take care of the case here. */ + bool pushed; + + pushed = SPI_push_conditional(); + slist = pg_plan_queries(slist, plansource->cursor_options, NULL); + + SPI_pop_conditional(pushed); } /*
В списке pgsql-bugs по дате отправления: