Any particular reason why? This creates a longer query string than the one I listed and they both give the same result:
select h.Day, h.Description, h.Month, h.Year, g.Name, r.Name from Holidays h join HolidayGroup g on h.HolidayGroup = g.id join HolidayRegion r on g.HolidayRegion = r.id
Does using a join make the query execute faster, or is it just a 'more accepted' query standard? I've been so used to our custom syntax that I haven't written a 'real' SQL query in more than a decade.
Even if this is something that would be committed (I have my doubts and generally don't think it should be) you'd still end up writing it because I am near certain no one else would be so inclined.
Given you are not adding any new capabilities and making significant changes to parsing code, the barrier for entry is very high. If the syntax was standard then work on using meta-data to auto-resolve the joins might be worthy of inclusion.
I would have thought that being able to reduce the complexity of long SQL statements would be seen as a new capability, but if it isn't desirable then I'll download the code and start looking into it. At the very least I'll probably have to provide a homebrew alternative to psql as we've got heaps of scripts written in this syntax which will take a very long time to port across to standard SQL.
I can't find any information on the postgres website about a way to submit feature requests/enhancements, only to report bugs. Is there a formal mechanism to request new functionality?
Most requests that are not bugs should go to -general unless one of the other lists seems more appropriate. Discussions about patches occur on -hackers.
Thanks, will keep that in mind for my next dumb idea :-)