Posted by Stephen Auerbach on April 21, 2008 at 13:20:58:
In Reply to: Re: New string operators coming posted by Keith Smith on April 19, 2008 at 22:10:20:
I'm guessing you're thinking of tools to search for records. Despite the fact that I was so thrilled with Barbara's CMTHELP tool, the guys at the Vegas meeting were underwhelmed. Their complaint was reasonable. Searching based on the starting letter(s) of the name field is kinda primitive.
These new tools SOUNDEX and CONTAINS are a step ahead of the starting letters, but both require that the application program read through the entire file.
So I did some timings:
I have a customer with a customer file containing 181,000 records. My timing program did nothing except read through the file. No logic, no testing, just read.
181,000 records --- 17 seconds.
Then I copied the file to a sequential file and timed the read. No key tree reads.
181,000 records --- 12 seconds. Faster but not amazingly faster.
So Jim: How much faster would the read be if Comet read an entire AU? What is an Allocation Unit using NTFS? My searching says 4K. How about a low level read that scooped in the data portion of the file. If we're searching order does not matter; speed really does. People get nervous when it takes a computer more than 2 or 3 seconds to respond.
My memory says Gary Cooper's Report Generator did just that and BB has not disputed my failing memory.
Each file can be a maximum of 1MB in length Uploaded files will be purged from the server on a regular basis.