Simon Slavin <slavins@...> writes:
> Would you care to explain what advantages Window functions would give us that VIEWs and sub-SELECTs don't > give us ? I'm not being contrary, I'd like to know. I have never compared lines of code between the various approaches, but window functions make it pretty simple to do rankings, moving averages, etc, while sub-SELECTS etc are much more convoluted. I think of window functions as syntactic sugar only, but still very nice and very useful for analytical work, less so for pure data store work. Compare the two SQL examples between Approach 2 and Approach 3 in the linked page: http://hashrocket.com/blog/posts/sql-window-functions Also, VIEWS are permanent and thus lead to clutter, though temporary views mitigate somewhat. I am not sure it would be the best use of developer time and energy, but like I say it would be great for us analysts who would like to do more SQL work without a server setup, and less SAS/ Excel/ R/ whatever. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users