ay to corrupt your database.
>
> As to where uncommitted transactions live, I am not sure. I'll let someone
> else answer that.
>
> On November 23, 2017 9:13:20 PM EST, "林自均" <johnl...@gmail.com> wrote:
> >Hi folks,
> >
> >When I was learning about rollba
Hi folks,
When I was learning about rollback journal, I did the following tests:
(in shell 1)
$ sqlite3 /tmp/db.sqlite
SQLite version 3.21.0 2017-10-24 18:55:49
Enter ".help" for usage hints.
sqlite> PRAGMA journal_mode;
delete
sqlite> CREATE TABLE bank (name STR, money INT);
sqlite> INSERT INTO
and renaming is not really that important. We
keep the corrupt database only for the developers to debug, not for anyone
else to use.
Best,
John Lin
Simon Slavin <slav...@bigfraud.org> 於 2017年11月23日 週四 下午12:03寫道:
>
>
> On 23 Nov 2017, at 3:11am, 林自均 <johnl...@gmail.com> wrote
Hi Simon,
It's for logs. If it corrupts, we rename it with a suffix ".corrupt" and
write new logs into a new sqlite file. Does that affect the way I check it?
Best,
John Lin
Simon Slavin <slav...@bigfraud.org> 於 2017年11月23日 週四 上午11:04寫道:
>
>
> On 23 Nov 2017, at 3:00am
Hi Richard,
Thank you for the explanation!
In other word, to check if a sqlite file is corrupted, I have to check the
stdout instead of the exit value. Am I right?
Best,
John Lin
Richard Hipp <d...@sqlite.org> 於 2017年11月23日 週四 上午10:53寫道:
> On 11/22/17, 林自均 <johnl...@gmai
org> 於 2017年11月23日 週四 上午9:39寫道:
> On 11/22/17, 林自均 <johnl...@gmail.com> wrote:
> >
> > Usually, if the database corrupts, the exit value is non-zero. However, I
> > recently bumped into a corrupted sqlite file that makes the previous
> > command exit with z
Hi folks,
The document of "PRAGMA integrity_check" says:
> If the integrity_check pragma finds problems, strings are returned (as
multiple rows with a single column per row) which describe the problems.
But it doesn't specify the exit value of:
$ sqlite some-db.sqlite "PRAGMA integrity_check"
7 matches
Mail list logo