Post weekend update

Well bulk of phase 2 update is done. It actually went pretty smoothly I think. The things that I thought would take hours took minutes but unfortunately the things that I thought would take minutes tooks hours. Anyway, what is now in existence looks and seems to work OK. Main index page is now drawing from a 1 hour old cached copy to reduce load on server (index page is heavy because it displays loads of database stats). Mod_rewrite of URLS went relatively smoothly. Ive just sorted out a couple of bugs (some weird escape characters in URLs and a trailing space problem with some names) but its working well, and the 301 redirect from the old URLs is working well according to the logs too. Unfortunately, and this never entered my mind as I was planning the directory and URL structure, but I’m now left with a duplication problem with some venues. Consider 2 Tesco supermarkets that are both in Dunfermline City Centre. Fairly simple and probably fairly common situation in many towns across the British Isles. Well my directory structure won’t distinguish between them at the moment and will both assign them an URL of bigreddirectory.com/tesco-dunfermline. Clearly this needs to be sorted. The site will distinguish between the two stores if it detects that the stores are in different AREAS of Dunfermline. I don’t know any areas in Dunfermline but say for London, Chelsea and Holborn for example. I think the only way forward is to give the stores a number. So bigreddirectory.com/tesco_1-dunfermline. I don’t think this is too bad. I think that’s the neatest way to pump out the URL and probably the easiest to remember as well for anyone trying to memorise URLs on the site. I don’t think tesco-127_aberdeen_street-dunfermline pulls it off.

What the site does do is take the full name of the company at the address so in reality /tesco-dunfermline is actually tesco_stores_ltd-dunfermline and most tesco stores actually seem to take the type of supermarket e.g. supermarket/metro/tesco extra in the name so I think the amount of duplication is actually probably rather low. That’s the next largish task on the list of things to do.

Also need to write a new search script for the top right hand corner search bar. I did away with the mootools javascript search script. It was excellent and looked great but I can’t justify loading 150kb of javascript each page when I’m trying to serve pages to the end user as fast I possibly can. Maybe there was a lot of redundant code in there that could have been cut but I don’t have the time to go through it so I binned the whole thing.