|
|
Posted by Steve on 01/18/08 20:58
"The Natural Philosopher" <a@b.c> wrote in message
news:1200672543.35803.0@iris.uk.clara.net...
> I've done this.
>
> Itseasierto do one big query, sort by whatever, iterate through the array
> and write a new header everytime the name changes
>
> e.g. select surname,... from mytable, order by name gets the lot in teh
> write order, and then a simple loop with something like posting a varablee
> for the current first letter, and if it DIOES'T match the initial of the
> name you are about to print, executing a bit of code that prints a new
> header, aand set it to the current first name.
>
> PHP may be slow, but not as slow as the overhead on an SQL call with the
> attendant file system shuffling.
phil...i've never, ever seen *any* scripting language that can out-perform a
database. sorry, i don't buy that. that one (whatever is meant by that)sql
query can run cached and be optimized by almost any db. whatever system
shuffling is to be done, it certainly will be mechanically LESS than hitting
the db n times to get the same result...especially considering that php is
involved between calls rather than just one process.
make a believer of me, though, and i will happily change my mind. :)
Navigation:
[Reply to this message]
|