Re: [sqlite] Quick way to determine optimal page size?

2019-08-02 Thread Simon Slavin
On 2 Aug 2019, at 8:55pm, Jen Pollock wrote: > Compressing the backups would likely save you a lot more space, and I > suspect it wouldn't be affected that much by page size; presumably empty > space at the ends of pages will compress very well. It might be an interesting exersize to compare

Re: [sqlite] Quick way to determine optimal page size?

2019-08-02 Thread Jen Pollock
Compressing the backups would likely save you a lot more space, and I suspect it wouldn't be affected that much by page size; presumably empty space at the ends of pages will compress very well. Jen Pollock On Thu, Aug 01, 2019 at 12:48:27AM +0300, Tony Papadimitriou wrote: > Tens of databases

Re: [sqlite] Large database backup

2019-08-02 Thread Tammisalo Toni
Thanks for your help! Especially the comment about cache size helped a lot. When I reduced the cache size to about 100 pages I actually get it do the writing incrementally in multiple sqlite3_backup_step() calls. With bit more finetuning I think it will be ok. I also had unrelated problem of

Re: [sqlite] Issue report: sqlite3_set_authorizer triggers error 4/516 (SQLITE_ABORT_ROLLBACK) during statement iteration

2019-08-02 Thread Gwendal Roué
I totally agree with your answer. But this wasn't really the question. > You have hacked around this security feature I beg you to try to look at my "hacks" with a fresh eye. The service they provide is a genuine one: be able to run raw SQL requests, and also to be notified when one has

Re: [sqlite] Issue report: sqlite3_set_authorizer triggers error 4/516 (SQLITE_ABORT_ROLLBACK) during statement iteration

2019-08-02 Thread Richard Hipp
On 8/2/19, Gwendal Roué wrote: > > Do you think this can still be seen as a misuse of the library? > Forcing a reprepare after an authorizer change is a security feature. It helps to prevent people from adding an authorizer, but then mistakenly using a statement that was prepared before the

Re: [sqlite] Issue report: sqlite3_set_authorizer triggers error 4/516 (SQLITE_ABORT_ROLLBACK) during statement iteration

2019-08-02 Thread Gwendal Roué
For the record, I have successfully refactored my code so that authorizer is set only once, and the issue has disappeared. Yet, this authorizer not always returns the same value for the same input. Precisely speaking, it always return SQLITE_OK, but when asked if it should allow deletion. In

Re: [sqlite] mode insert dumps

2019-08-02 Thread Olivier Mascia
> Le 2 août 2019 à 10:04, Luca Ferrari a écrit : > > I've got a lot of database files that I would like to dump using > INSERT statements. > unluckily, .mode insert does not work for me, or I'm not able to > understand how it works. > > sqlite> select mude_anno, mude_numero from catdaemo; >

Re: [sqlite] [EXTERNAL] mode insert dumps

2019-08-02 Thread Hick Gunter
You need to .mode insert SELECT * FROM ; Repeat for all your tables. -Ursprüngliche Nachricht- Von: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] Im Auftrag von Luca Ferrari Gesendet: Freitag, 02. August 2019 10:04 An: SQLite Betreff: [EXTERNAL] [sqlite] mode

Re: [sqlite] mode insert dumps

2019-08-02 Thread Dan Kennedy
On 2/8/62 15:04, Luca Ferrari wrote: Hi, I've got a lot of database files that I would like to dump using INSERT statements. unluckily, .mode insert does not work for me, or I'm not able to understand how it works. sqlite> select mude_anno, mude_numero from catdaemo; INSERT INTO table

[sqlite] mode insert dumps

2019-08-02 Thread Luca Ferrari
Hi, I've got a lot of database files that I would like to dump using INSERT statements. unluckily, .mode insert does not work for me, or I'm not able to understand how it works. sqlite> select mude_anno, mude_numero from catdaemo; INSERT INTO table VALUES(2019,1161); My questions are: 1) why is