Qasim, Yes, after seeing the performance from straight cf code I knew I had two ways of going, Java and its methods or DTS. Ultimately it was a choice based on familiarity (DTS) and since this routine is a one time (two times at most) thing just getting it done was what was important.
For the sake of satisfying my own curiosity, I think I'll do as suggested in a subsequent reply, which is to create two versions of the same routine, one cfml and on cfscript and compare the generated class files' code. Kevin > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Qasim Rasheed > Sent: Sunday, October 03, 2004 8:54 PM > To: [EMAIL PROTECTED] > Subject: Re: [CFCDev] OT: cfscript vs cfml > > > Kevin, > > Aside from the argument of cfml and cfscript, have you > thought about using Java StringBuffer class for string > manipulation. It is definitely way faster than regular string > manipulation. > > > > On Sun, 3 Oct 2004 10:54:31 +0200, Kevin J. Miller > <[EMAIL PROTECTED]> wrote: > > CFMX 6.1: I have a routine for parsing some log files that is > > heinously slow (string parsing being arguably never one of cf's > > strengths if memory serves). To make a long story short I did the > > routine almost entirely in cfscript, it just *felt* more cfscript-y > > and so I did it this way (it also only took me an hour to > write, vs a > > couple days in an alternate method - see final note). > > > > After showing it to another developer he asked why I did it in > > cfscript vs. cfml. I answered as before, 'it just felt more like a > > cfscriptish routine', followed by 'and so what, if cf is > all compiled > > to servlets why should it matter if something is done cfscript vs > > cfml?'. His reply was that he had read/heard that while cfml is > > compiled to java, cfscript is left intact and is interpreted by 'cf > > tags' at runtime. Can anyone confirm this? While its > certainly known > > that there are speed differences between some cfml/cfscript > > implementations of the same thing, I hadn't given it much thought > > after moving to MX (see www.developer.be 's benchmarking > articles for > > more). > > > > BTW, I'm redoing the routine as a sql server DTS, so while I don't > > really need any suggestions on resolving my speed issue > (it's a one or > > two time thing at most), I *am* curious if the statement regarding > > compilation of cfscript, or lack thereof, is accurate. > > > > TIA > > Kevin > > > > ---------------------------------------------------------- > > You are subscribed to cfcdev. To unsubscribe, send an email to > > [EMAIL PROTECTED] with the words 'unsubscribe cfcdev' in the > > message of the email. > > > > CFCDev is run by CFCZone (www.cfczone.org) and supported > > by Mindtool, Corporation (www.mindtool.com). > > > > An archive of the CFCDev list is available at > > www.mail-archive.com/[EMAIL PROTECTED] > > > ---------------------------------------------------------- > You are subscribed to cfcdev. To unsubscribe, send an email > to [EMAIL PROTECTED] with the words 'unsubscribe cfcdev' > in the message of the email. > > CFCDev is run by CFCZone (www.cfczone.org) and supported > by Mindtool, Corporation (www.mindtool.com). > > An archive of the CFCDev list is available at > www.mail-archive.com/[EMAIL PROTECTED] > ---------------------------------------------------------- You are subscribed to cfcdev. To unsubscribe, send an email to [EMAIL PROTECTED] with the words 'unsubscribe cfcdev' in the message of the email. CFCDev is run by CFCZone (www.cfczone.org) and supported by Mindtool, Corporation (www.mindtool.com). An archive of the CFCDev list is available at www.mail-archive.com/[EMAIL PROTECTED]
