[cfaussie] Re: QofQ

[Joel Cass]

Hi Rod,   If it's just one field and you are matching dates (ie. date1 = date2), a referential struct may help, ie   <cfset struct[date]=arrayNew(1)> ... <cfset arrayAppend(struct[date],structOfcolumns)>   Or you may be able to use looping or cfoutput grouping, ie.   <cfoutput query="myquery" group="datefield">     #datefield#<hr />     <cfoutput>         #recordDetails#<br>     </cfoutput>     <hr /> </cfoutput>   or   <cfloop query="myquery">     <cfif datefield IS thisDate>         ...     </cfif>     <cfif dateField IS thatDate>         ...     </cfif> </cfloop>   Not really sure about the purpose you need but there would be a solution without QoQ, just not sure how difficult it would be to implement.   QoQ is pretty inefficient - I'm not saying you shouldn't use it, I just don't like QoQ all that much. It's a personal thing, really. Perhaps someone with more experience in QoQ may be able to help..   Joel -----Original Message----- From: cfaussie@googlegroups.com [mailto:[EMAIL PROTECTED]On Behalf Of Rod Higgins Sent: Wednesday, 28 June 2006 10:39 AM To: cfaussie@googlegroups.com Subject: [cfaussie] Re: QofQ Hi Joel,   The reason Im using QofQ firstly is because Im lazy and secondly because Im using a lot of date/time matching from the source recordset. The source recordset has about 160 odd records so it's not that large. What I am finding though is QofQ is not really that efficient. I was wondering what other ppl's thoughts are on this approach. I was under the impression this approach would be faster then hitting the db 50 odd times pulling out recordsets containing one or two records each. Maybe I will have to convert the orginial recordset into arrays, and then pull out the data that way instead.   Any thoughts?   ----- Original Message ----- From: Joel Cass To: cfaussie@googlegroups.com Sent: Wednesday, June 28, 2006 10:27 AM Subject: [cfaussie] Re: QofQ As someone who does not use QoQ that often, here are some other ways you can do this:   - If you want to look up a certain value, create a referential struct based on the column you wish to lookup, containing the vital information - If you want to create little queries, you could try query functions (querynew, queryAddColumn etc) - You could try converting the query into an array, CFC, function or other structure that retrieves data in a more relevant way - Use query caching and the startrow / endrow features of cfloop - Do lots of little queries with the database instead of continually messing around with one big query stored in memory   CPU and memory are different things. Someone with a CS degree should clear that up for you. Storing data in memory doesn't mean that you avoid processing the data when you start using it in QoQ.   I don't know why people use QoQ anyway. It seems limited compared to other ways of processing data. That's just a personal opinion. I know it's probably extremely useful in very specific cases.   Joel   -----Original Message----- From: cfaussie@googlegroups.com [mailto:[EMAIL PROTECTED]On Behalf Of Rod Higgins Sent: Wednesday, 28 June 2006 9:50 AM To: cfaussie@googlegroups.com Subject: [cfaussie] QofQ When running a page that uses excessive QofQ from a single query the CPU goes bezerk. I was under the impression that QofQ was simply accessing a recordset from memory. I have a couple of questions if anyone has some ideas, workarounds for this. Why does the cfstat utility record QofQ's as DB hits/sec? Also what methods have ppl used to improve performance on pages that use a lot of QofQ - where talking about 50 odd QofQ calls to a single recordset.   TIA --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "cfaussie" group. To post to this group, send email to cfaussie@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/cfaussie -~----------~----~----~----~------~----~------~--~---