Created on 2008-07-06 22:56 by haypo, last changed 2008-07-19 09:37 by jcea. This issue is now closed.
| Files | ||||
|---|---|---|---|---|
| File name | Uploaded | Description | Edit | |
| bsddb_create.patch | haypo, 2008-07-06 22:55 | Proposition to fix the bug | ||
| Messages (3) | |||
|---|---|---|---|
| msg69358 - (view) | Author: STINNER Victor (haypo) *  |
Date: 2008-07-06 22:55 | |
newDBObject(), called by DB_construct(), doesn't check correctly the result of all to the external function db_create(). It checks if self->db is NULL, but db_create() doesn't change self->db value on error. So if self->db is uninitialized, the error is not catched. Two ideas to fix the bug: - check "if (err)" instead of "if (self->db != NULL)" - set self->db=NULL before calling db_create() I implemented the second proposition in the attached patch. Note: The bug occurs with PYDEBUG, I don't know if PyObject_New() fills new allocate memory to zero (I think no, but I'm not sure). |
|||
| msg69359 - (view) | Author: STINNER Victor (haypo) * |
Date: 2008-07-06 22:59 | |
The bug occurs on db_create() failure. Dummy example to reproduce it: "import _bsddb; _bsddb.DB(None, 29.515)" |
|||
| msg70008 - (view) | Author: Jesús Cea Avión (jcea) * |
Date: 2008-07-19 09:37 | |
Solved in my SVN repository. Revision 527. Testsuite updated. Will be available in bsddb 4.7.2. |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2008-07-19 09:37:50 | jcea | set | resolution: fixed |
| 2008-07-19 09:37:05 | jcea | set | status: open -> closed messages: + msg70008 |
| 2008-07-07 04:36:19 | gregory.p.smith | set | priority: normal assignee: jcea type: crash nosy: + jcea |
| 2008-07-06 22:59:13 | haypo | set | messages: + msg69359 |
| 2008-07-06 22:56:00 | haypo | create | |