Mark J Cox, mark@awe.com  
   
mark :: blog

<< prev [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 ] next >>



So I got sent a copy of the file and it turns out only two UK banks actually have OFX servers, Nationwide and Woolwich. Since I'm with the Nationwide I thought I'd have a go. Added the Nationwide entry to the fipartnr.ini file and bingo, it all worked perfectly. Until Money decided to check for updates and found that fipartnr.ini had changed. Anyway as long as I ignore the "Nationwide has changed it's online services" messages then all is okay.

Have you seen things like yodlee.com? I've been thinking of writing a OFX server with screen scraping so that you could connect to it and it would connect to all those other banks I use that I can screen scrape the info out: Lloyds TSB, Egg, Cahoot, Citibank, various pension providers.

[By messing around with URLs at the various sites and using the mnn extraction tool I've found the Microsoft test bank server and some other fun things!]

Updated my OFX import utility to screen-scrape American Express cards too



So it turns out the guy who wrote libxml and libxslt works for Red Hat too as I was in a phone conference today with him. I've also been looking at the Perl module that links in libxslt but it's having teething problems with glibc 2.2.2.

Finished off my Guide to ApacheCon 2001 and my ApacheCon photos



My last week here so decided to go up to the mountains with Bryce who for some reason decided to take his laptop. After my recent fun with cars and Durham Police I let Phil do all the driving, which I don't think he appreciated.

I found a new xsl parser, libxslt and libxml, part of the gnome project, seems to be quicker than Xalan and has the advantage of being in C and not C++ (I'm so used to release engineering for 30 platforms where even a working C compiler is sometimes a stretch). Spent a good 6 hours on Saturday failing to get it working right with Perl the way I want it to, seems to be a strange memory problem that goes away everytime I turn on debugging.

I was going to buy a Earthmate GPS receiver for my laptop until I read that it doesn't output standard NMEA format data so it won't work with Autoroute. I'm still looking for a GPS solution so the next time I'm coming back from Scotland and get stuck for 4 hours on the M6 without a map I can find an alternative route.



Finally committed the raw XML for the apache.org in the news page. I've been meaning to get that all updated for months, but never had the time.

Random other fun I can't talk about yet.



Worked out my schedule for the next weeks, I fly to England next week for 4 days, then to Raleigh for 4 days, then Reston Virginia for 4, then Raleigh for 3, then Santa Clara for ApacheCon for 5, then Raleigh for a week, then back to the UK. All economy too :-(

Finally got a spare few hours so finished updating the XML database of news stories about Apache. Used it to update the apache.org in the news page.

FInished "The Longest Journey". Best interactive fiction I've ever played.

My DELL laptop wasn't suspending to disk, found out it was because the suspend partition had to be in the first 8Gb of the drive. Took that time to re-install Linux (it had never been installed right) and decided to use the latest Red Hat rawhide. I was pretty amazed when it detected and worked with my Wavelan card and random PCMCIA modem without having to download a thing. These Red Hat guys are great *grin*



Back from the UK armed with fizzy sweets (US translation: sour candy) which seem to be popular in the office. When we got bought by Red Hat last year I gave a short presentation on Stronghold to all the developers. There was a large collective groan when they found out bits of it were Perl based and that all our developers knew Perl and that we didn't like Python. Actually, I shouldn't have said I didn't like Python until I'd actually read about it because after spending the flight learning it I think I'd prefer it for most of my scripting needs. So I'm back in the North Carolina snow for a little while, although I don't think Tracy back in Glasgow is too pleased at that thought.

Meanwhile I converted my photo album scripts to output XML (index.xml) with stylesheets (index.xsl) as the old system wasn't very flexible



Popped down to "Best Buy" at the weekend and came away with a new adventure game "Longest Journey" and a pair of speakers so I can finally listen to music in the appartment. Spent the rest of the weekend playing the game (it's the best adventure I've played in a long time), writing reports, and thinking of fun things to do with XML-RPC and Python.

My screen-scraping script to convert EGG credit card statements into OFX for Microsoft Money broke and attempts to get EGG to actually output their information in a format other than HTML has failed. I'll switch my credit card to the first provider that offers a SSL enabled XML-RPC interface for interacting with my account and downloading statements and transaction lists. Was thinking of cool things I could have done to the Bradford Robotic Telescope using XML-RPC or SOAP; it could replace all those proprietary communication protocols and systems I wrote and allow others to embed the functionality of the telescope into their own applications. But it looks like the project has died, they've not noticed my Y2K bug for example.



I can't believe I wrote the first issue of Apache Week five years ago today. Happy Birthday! It's interesting looking through the past issues to see when we expected a 2.0 beta release; one year ago we were expecting it "inside a month". We're currently expecting it "inside a month".



Another online credit card that gives you a statement but no way of downloading it. How difficult is it to add that function? Anyway until they do here is a program to convert the Cahoot HTML pages into either QIF or OFX format

To use the script:
  • Log onto the Cahoot web site, and select the statement you want to convert, display the printer friendly version
  • In your browser save the current page to a file. You sometimes need to do "view source" and save that version. You'll end up with a file, lets say cahoot.html
  • Make sure you have a perl interpreter on your system. If you're using Linux or other Unix you'll most likely have perl installed. If you're using Windows you'll need to get Perl from someone like ActiveState and figure out the command line stuff
  • Run
    perl 2ofx.pl cahoot.html > cahoot.qif
  • Use "Import" in Microsoft Money to load in cahoot.qif



Another online credit card that gives you a statement but no way of downloading it. If you have a card in the USA you can use Microsoft Money to interactively download your statement. If you live in the UK you can't. Anyway until they do here is a program to convert their HTML pages into either QIF or OFX format for import into Money

<< prev [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 ] next >>

       


Hi! I'm Mark Cox. This blog gives my thoughts and opinions on my security work, open source, fedora, home automation, and other topics.

pics from my twitter:


popular tags: [all], apache, apachecon, apacheweek, cve, cvss, fedora, financial, geocaching, ha, metrics, microsoft, nashville, north carolina, red hat summit, redhat, security, trips


Subscribe to RSS feed