Re: [sqlite] Does wal-file support MMAP？

2017-11-08 Thread Dan Kennedy
On 11/09/2017 08:59 AM, advancenOO wrote: I am trying to understand how SQLITE_DEFAULT_MMAP_SIZE works and I think db files can use MMAP by using this compilation option, as Sqlite3PagerGet() mapped to getPageMMap(). And I know "/the current VFS implementations use a mmapped file for the

[sqlite] Does wal-file support MMAP？

2017-11-08 Thread advancenOO
I am trying to understand how SQLITE_DEFAULT_MMAP_SIZE works and I think db files can use MMAP by using this compilation option, as Sqlite3PagerGet() mapped to getPageMMap(). And I know "/the current VFS implementations use a mmapped file for the wal-index/" But is there a way that wal-files

Re: [sqlite] Automatic indexes don't work on without rowid tables

2017-11-08 Thread Richard Hipp
On 11/8/17, korablev wrote: > What is the timeline of this feature? Will it be implemented in the nearest > future? Low priority. I do not like automatic index feature, because I think that app developers should do an appropriate CREATE INDEX instead. Of course, it is not

Re: [sqlite] Automatic indexes don't work on without rowid tables

2017-11-08 Thread korablev
What is the timeline of this feature? Will it be implemented in the nearest future? And what optimizations doesn't also work on without rowid tables? Is the query planner good at optimizing stuff on without rowid tables at all? -- Sent from: http://sqlite.1065341.n5.nabble.com/

Re: [sqlite] Most efficient way to detect on-disk change

2017-11-08 Thread Wout Mertens
Thank you all! I really have to do a thorough read of all the pragmas, there are so many useful things in there! The user version sounds exactly like what I should be using for storing the db version, and presumably the data_version is a little faster still than reading the user version. @Keith,

Re: [sqlite] Can I determine the collation of a text argument to a function?

2017-11-08 Thread Richard Hipp
On 11/8/17, Jens Alfke wrote: > Is it possible for a user-defined function to determine the collation > associated with its argument(s)? I'm implementing some custom string > matching/comparison functions, and I want their case-sensitivity to depend > on the collation of the

[sqlite] Can I determine the collation of a text argument to a function?

2017-11-08 Thread Jens Alfke
Is it possible for a user-defined function to determine the collation associated with its argument(s)? I'm implementing some custom string matching/comparison functions, and I want their case-sensitivity to depend on the collation of the first argument (the LHS). I’ve searched sqlite3.h for

Re: [sqlite] CLI: .dbinfo does not work in version 3.21

2017-11-08 Thread Richard Hipp
On 11/8/17, Richard Hipp wrote: > > Expect to see fresh precompiled binaries that fix this problem within > a few hours. > New builds for the "sqlite-tools-*-321.zip" packages are now available from the https://sqlite.org/download.html webpage -- D. Richard Hipp

Re: [sqlite] Automatic indexes don't work on without rowid tables

2017-11-08 Thread Richard Hipp
On 11/8/17, Peter Da Silva wrote: > On 11/8/17, 8:20 AM, "sqlite-users on behalf of Richard Hipp" > > wrote: >> The technical reason for this is that, from the point of view of the query >>

Re: [sqlite] CLI: .dbinfo does not work in version 3.21

2017-11-08 Thread Richard Hipp
On 11/8/17, Klaas Van B. wrote: > In 3.20.1 the dot-command dbinfo still worked, but since 3.21.0 not anymore This was a boo-boo on our precompiled-binary build script. The ".dbinfo" command was modified to require a special compile-time option (specifically

Re: [sqlite] Automatic indexes don't work on without rowid tables

2017-11-08 Thread Peter Da Silva
On 11/8/17, 8:20 AM, "sqlite-users on behalf of Richard Hipp" wrote: > The technical reason for this is that, from the point of view of the query > planner, a WITHOUT ROWID table is really a covering index So basically

Re: [sqlite] Automatic indexes don't work on without rowid tables

2017-11-08 Thread Richard Hipp
On 11/8/17, korablev wrote: > sqlite> create table t1(a primary key, b) without rowid; > sqlite> create table t2(a primary key, b) without rowid; > sqlite> explain query plan select * from t1, t2 where t1.b = t2.b; > 0|0|0|SCAN TABLE t1 > 0|1|1|SCAN TABLE t2 > sqlite> create

[sqlite] CLI: .dbinfo does not work in version 3.21

2017-11-08 Thread Klaas Van B.
In 3.20.1 the dot-command dbinfo still worked, but since 3.21.0 not anymore sqlite32001 e0.sqb SQLite version 3.20.1 2017-08-24 16:21:36 Enter ".help" for usage hints. sqlite> .dbinfo database page size:  512 write format:    1 read format: 1 reserved bytes:  0 file change counter:

Re: [sqlite] Most efficient way to detect on-disk change

2017-11-08 Thread Dan Kennedy
On 11/08/2017 03:55 PM, Dominique Devienne wrote: On Wed, Nov 8, 2017 at 7:45 AM, Dan Kennedy wrote: On 7 Nov 2017, at 6:53pm, David Raymond wrote: I think pragma data_version is what you're looking for.

Re: [sqlite] Most efficient way to detect on-disk change

2017-11-08 Thread Dominique Devienne
On Wed, Nov 8, 2017 at 7:45 AM, Dan Kennedy wrote: > On 7 Nov 2017, at 6:53pm, David Raymond wrote: >> >> I think pragma data_version is what you're looking for. >>> http://www.sqlite.org/pragma.html#pragma_data_version >>> >> > I think it's the