Re: [python-uk] Tell us what you did with python this year
I have had a number of python projects, my happiest one was one that scraped government websites for jobs and then applied for those with semi fitting criteria. I was unemployed at the time with unemployment insurance but my insurers wanted proof that I'd been looking. 24 hours hacking and 1 cron job later , I had a script that scraped a government website for jobs, applied for jobs via email , saved applied for jobs in a Django database (so I couldnt apply twice) and sent a monthly report to my insurer. Bingo, paid holiday for a year. Created a couple of Django simple Django websites. Built a python robot for Google wave, that did the background research for finding references (http://vimeo.com/5772930) for the document you are collaboratively typing , only that my robot did legal research screen scraper robot to spam gumtree with my girlfriends flat that she was trying to advertise, turned out to be 8 times as good as the estate agents she was paying currently playing a bit with NLTK, Twisted and using mechanize to scrape facebook for nefarious reasons, im getting into functional programming recently have the ephinany that if I want to write a functional screen scraper I must view the mechanize.Browser as data! future plans are to use twisted and/or google app engine to massively parallelize my scraper Also got involved with planning the python dojos , but so far have been a no show to every single one *cringe* Greg
I didn't think I did all that much with Python this year other than throwing together a few "one off" scripts to collate and analyze data from several thousand tills. These so-called "one off" scripts have been remarkably useful and they saved one of our teams a lot of time by automating what was previously a rather time consuming, ad hoc process. That being said, our client's demand for stats is insatiable. Last year I did a bit more. We had what appeared to be a performance issue with online authorisation for a POS system. We suspected that the problems were coming from an external system, but couldn't prove it given the data that we had. We used Stackless Python, along with a useful little framework called Concurrence (http://opensource.hyves.org/concurrence/), to write some performance testing utilities to demonstrate that the POS system was indeed fit for purpose and that the cause of the problem was external. --- Rod
participants (2)
-
greg nwosu -
Rod Hyde