> As I understand it, this should be enough to make it fail.

No. That's what I gathered in the real-world application too: that
sometimes it works, and other times it SIGSEGVs. I would think that
this JNIEnv thing is some kind of thread-local, which is implemented
with some global state, and sometimes it points to a (in)valid
environment, and sometimes not (then it points to NULL, as the core
evaluation shows).
In any case, I've tested your code
http://hashbin.com/827.html
and it doesn't trigger the SEGV.


--~--~---------~--~----~------------~-------~--~----~
Mailing List: http://groups.google.com/group/sqlitejdbc?hl=en
To unsubscribe, send email to [EMAIL PROTECTED]
-~----------~----~----~----~------~----~------~--~---

Reply via email to