You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In email conversation with @h4yn0nnym0u5e, one somewhat frustrating thing that can crop up when using DB4S (manually, in the "Execute SQL" tab) is getting a join wrong, and the result set tries to return a huge number of rows. eg 1.6 mill rows
@MKleusberg @innermous @prutz1311 Any idea if there is some way we could catch that kind of thing - without needing a large amount of code change - and not let it go out of control?
Just a thought, but might be worth throwing that out to the sqlite users mailing group, as they might suggest some hook you could use .... you're not going to know how many rows you're going to get back until sqlite has done its biz (and consumed 97 GB of memory...) but you might be able to abort it after ... say, 20 seconds?
No need to ask them 😉 This does what we need. I'd suggest to not do anything automatically because some queries just take their time. See #409 for example. But what we can do is add a cancel button which just allows the user to stop the query at any time.
Cancel button is probably best, other possible approach I can think of is have a setting for query timeout and once that expires, ask user if he wants the query to run some more or cancel.
@h4yn0nnym0u5e@tilusnet Sorry for the long delay here. Are you still interested in this issue? I'm asking because we are currently preparing some changes which add, besides making some other changes, a cancel button to the Execute SQL area. Because it's changing a lot of our code we want to test this as good as we can. Are you interested in giving it a try? You can find the Windows and OSX build here:
Hi @MKleusberg Personally, my focus shifted away from SQLiteBrowser in the near past, moreover I am exclusively on Linux currently which makes testing impossible. Sorry I can't help.
Activity
chrisjlocke commentedon May 4, 2017
Just a thought, but might be worth throwing that out to the sqlite users mailing group, as they might suggest some hook you could use .... you're not going to know how many rows you're going to get back until sqlite has done its biz (and consumed 97 GB of memory...) but you might be able to abort it after ... say, 20 seconds?
justinclift commentedon May 4, 2017
Ahhh, good thinking Chris. You're already on that mailing list aren't you?
If so, would you be ok to ask? 😄
chrisjlocke commentedon May 4, 2017
Surely can. ✍️
MKleusberg commentedon May 4, 2017
No need to ask them 😉 This does what we need. I'd suggest to not do anything automatically because some queries just take their time. See #409 for example. But what we can do is add a cancel button which just allows the user to stop the query at any time.
justinclift commentedon May 4, 2017
Awesome. 😀
chrisjlocke commentedon May 4, 2017
Yup, Dr. Richard Hipp just replied back with the same suggestion:
justinclift commentedon May 4, 2017
Excellent, thanks Chris & D. Richard Hipp. 😄
justinclift commentedon May 4, 2017
For reference:
https://www.mail-archive.com/sqlite-users@mailinglists.sqlite.org/msg103468.html
mglatz commentedon Jun 29, 2017
Cancel button is probably best, other possible approach I can think of is have a setting for query timeout and once that expires, ask user if he wants the query to run some more or cancel.
mgrojo commentedon Jan 7, 2018
What's the state of this issue?
tilusnet commentedon Jul 25, 2018
I think a cancel button is badly needed. Would it be difficult to add?
MKleusberg commentedon Oct 18, 2018
@h4yn0nnym0u5e @tilusnet Sorry for the long delay here. Are you still interested in this issue? I'm asking because we are currently preparing some changes which add, besides making some other changes, a cancel button to the Execute SQL area. Because it's changing a lot of our code we want to test this as good as we can. Are you interested in giving it a try? You can find the Windows and OSX build here:
tilusnet commentedon Oct 18, 2018
Hi @MKleusberg Personally, my focus shifted away from SQLiteBrowser in the near past, moreover I am exclusively on Linux currently which makes testing impossible. Sorry I can't help.
MKleusberg commentedon Oct 18, 2018
No worries, @tilusnet 😄
justinclift commentedon Oct 18, 2018
As a data point, we definitely compile and run ok on Linux too. My main desktop these days is CentOS 7, and that runs DB4S without issue. 😄
chrisjlocke commentedon Oct 18, 2018
No it doesn't. There's 281 of them. 😉
justinclift commentedon Oct 18, 2018
😛
MKleusberg commentedon Feb 12, 2020
Let's close this issue. I'm pretty sure the cancel button is working as expected and we have other open issues for this too 😄
I wish this was still the case 😉