On 2016/03/06 1:39 PM, Paul Sanderson wrote: > I understand this - but, there always a but, I still would like to do > something. Applying the limit anyway and then telling them the query > has been limited might be a solution. > > Time is usually not an issue but as the results are loaded into a grid > for display memory can be the issue.
Yes, you are quite right. I don't really have a "best way" solution for you. I could tell you how I solved it, or perhaps I should say, how I made it work, in SQLitespeed (Jury's out on whether it is a solution at all). I added an automatic (and user-configurable) LIMIT clause to every standard table query (to avoid loading tables with millions of rows) for simply speed purposes, but I display this limit to the user and they can adjust it. Once it is a user defined query, I don't add any LIMIT clause, but I limit the results set inside the GUI to 16 megabytes of output text (which displays whatever amount of rows fills up that many megabytes, usually quite a few more than can possibly be read in reasonable time-frames). Of course it is suggested that if more results are needed, to write the query to file since the GUI will get sluggish after that. (Hovering over the query buttons highlights this too). You can download SQLitespeed[1] for windows easily and test these assumptions, play for a few minutes to see if the idea works or suits you or perhaps work out a better idea (which I won't mind hearing about) :) As a note, the vast majority of user database queries seem to load up less than 16MB of query output for any query a user is interested in inspecting (i.e. might want to look at in a GUI). Any larger queries usually only gets used inside some system as a dataset - and 16MB is really easy to manage, even for a very basic GUI. [1] http://www.sqlc.rifin.co.za/ > > I'll give it some more thought, but thank you for your answers it has helped. > Paul Best of luck! Ryan