Working with RediSQL another use case comes to mind for some implementation
of the interface we were discussing.
How to detect SELECT statements that return empty.
SQLite simply return SQLITE_DONE in all cases, and it makes impossible to
know if it is an empty SELECT or something else.
A
I would really support this other approach.
It seems more extensible and eventually would address on of my pain points
in SQLite, not knowing what type of statement is being executed.
--
Sent from: http://sqlite.1065341.n5.nabble.com/
___
Hi List,
I would like to propose a feature and I believe here is the best place.
Please keep in mind that I am ready to work on a patch for this feature if
needed, but I would like to discuss it here first.
I would like to propose a function (named `sqlite3_stmt_action` for the sake
of
Thanks!
Indeed I was already aware of the `sqlite3_stmt_readonly()` function. But it
is not enough in my case.
I really need to distinguish between UPDATE, DELETE and INSERT
Cheers,
Simone
--
Sent from: http://sqlite.1065341.n5.nabble.com/
___
Hi Simon,
Thanks! Indeed I thought about a similar solution but it seems a little
contrived.
Do you think it would be possible to add a function that does what I need?
How is the design coordinated? Where should I post my RFC?
Cheers,
Simone
--
Sent from: http://sqlite.1065341.n5.nabble.com/
Hi All,
I am already afraid that the answer to the question will be "no", but it
will save a lot of development effort on my side, so is worth to ask.
Given a statement, is there any way to know if it is an UPDATE, DELETE,
INSERT, SELECT or something else?
Thanks,
Simone
--
Sent from:
Sorry,
I was a little too optimistic.
Making the starts unique does help only for some queries, not for all.
Why?
Cheers,
Simone
--
Sent from: http://sqlite.1065341.n5.nabble.com/
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
Ok, after the message I thought a little bit more.
And it turns out that in the database the `start`s are not unique how they
should.
Making them unique, seems to solve the performance problem completely.
However, still, I am not sure why the `LIMIT 1` does not help at all.
Can you guys shed
Hi all,
thanks for your suggestions, unfortunately, I already tried all of them,
except for the rtrees.
Actually, my request for help wasn't complete.
The ranges I am storing in the table are not overlapping.
To make an example in SQL.
The following can be in the dataset:
INSERT INTO
Hi all,
I am facing an interesting optimization problem.
I have a table like this:
CREATE TABLE ranges (
start int,
end int,
value int,
);
The query that I am interested in optimizing is "select value from ranges
where (? between start and end)"
The max performance that I was able
am missing?
For reference the code is here:
https://gist.github.com/siscia/c6c8c6637f93e56a70a226e1ab643ab4
--
Sent from: http://sqlite.1065341.n5.nabble.com/
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.o
11 matches
Mail list logo