Of course, the behaviour is not actually "undefined" -- it is perfectly determinable and entirely predictable and reasonable. However, if one does not understand the factors which determine the behaviour then, for you, the behaviour is undefined. In other words, if one does not know what results are expected to be obtain and why, then the responsibility rests with the one who is "using the force" instead of understanding what they are doing.
> -----Original Message----- > From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users- > boun...@sqlite.org] On Behalf Of Richard Hipp > Sent: Friday, 12 July, 2013 14:03 > To: General Discussion of SQLite Database > Subject: Re: [sqlite] Another 2 questions about SQLite > > On Fri, Jul 12, 2013 at 3:01 PM, Igor Tandetnik <i...@tandetnik.org> > wrote: > > > On 7/12/2013 12:30 PM, James K. Lowden wrote: > > > >> The documented behavior is - if you modify the data as you iterate > >>> over that same data, the results are unpredictable. > >>> > >> > >> Where does it say that? > >> > > > > You got me here. The behavior doesn't appear to be documented, > > > Proposed documentation enhancement here: > http://www.sqlite.org/draft/isolation.html > > > > > and it probably should. The closest I could find to the official > word is > > this post by Dr. Hipp from 2009: > > > > http://sqlite.1065341.n5.**nabble.com/Python-sqlite-** > > binding-commit-inside-select-**loop- > td51927.html#a51941<http://sqlite.1065341.n5.nabble.com/Python-sqlite- > binding-commit-inside-select-loop-td51927.html#a51941> > > "The official policy is that if you modify a table (via INSERT, > UPDATE, or > > DELETE) in the middle of a SELECT, then what you get from the > remainder of > > the SELECT is undefined." > > > > -- > > Igor Tandetnik > > > > > > ______________________________**_________________ > > sqlite-users mailing list > > sqlite-users@sqlite.org > > http://sqlite.org:8080/cgi-**bin/mailman/listinfo/sqlite- > **users<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 _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
