Archives June 2006
FormEncode is nice (Validators are so much nicer than coding everything oneself)
Written by
on
in
Snaking.
Dipping my toes in the waters of FormEncode today. Basically decided I needed a break from financial aspects, so spent some time roughing out a simple list and add view for Inventory objects (in this particular case, phone numbers/DIDs). Most of the work was:
Figuring out that All and Any values are applied in reverse ...
Bogging down a bit on the charge-setup subsystem (Uncharted areas for me to some degree...)
Written by
on
in
Knight Errant.
One of the major wish-list items for the new billing system is the ability to set up differential charging schemes using a tabular setup, so that admins can set up a default table of charges and then sets of overrides to those charges based on a rule match (e.g. you have a given token, or ...
Addresses and constrained addresses (Long-term wish-list items...)
Written by
on
in
Knight Errant.
When you run a cable company, it makes sense to have the ability to explicitly and exhaustively define the streets and street-ranges in the town so that you can never have an invalid address entered for service. I've just finished the basic implementation of that for the better billing project, along with the other major ...
These are the continuing voyages... (More work on BetterBilling...)
Written by
on
in
Knight Errant.
Continued work on BetterBilling this evening. Particularly started working on creating stand-ins for the eventual user-driven package-setup functions. Eventually a user should be able to compose a new package (billable thing) out of technical effects (services provided by the backend) and configurations (user choices that control the operation technical effects). For now I'm just manually ...
Test suites are fun! (Cancelled meetings and working on almost-spec...)
Written by
on
in
Knight Errant.
Other than a few hours on the (largely finished) old project (so far), I've been concentrating today on TurboGears and particularly moving the BetterBilling project ahead. I've been running into (minor) SQLObject bugs here and there, probably because inheritance isn't a commonly used feature and I'm using it extensively.
I've finally got to the point ...
Select from an inherited table in TurboGears (Since it took me so long to figure this out...)
Written by
on
in
Snaking.
Okay, key tip for debugging TurboGears SQLObject problems, add ?debug=1 to the dev.cfg URL for your database. Once you have that and you try to run a query for an inherited class you'll see the problem, but so that you don't have to set it up, here's the description of the problem and the solution: ...
Fits and starts (Little things...)
Written by
on
in
Snaking.
Did some more work on TurboGears. Found another couple of bugs in the SQLObject SQL generation code. Now failing with the inheritance stuff, seems that the documented ability to do Child.select( Parent.field==something) doesn't throw any errors but also doesn't seem to return the results. Anyway, nothing huge, just little things I'll have to track down ...
Offerred Water
Written by
on
in
Vindaloo.
It's not that I'm not thirsty
I'm parched to dusty stone
But the timing seems all wrong
For I've just resolved to quit
The search for cups of water
In hopes forgetting hard thirst
Will let me grow to be a man
Who deserves a cup of water
3 hours of procrastination on a 5 minute fix (Well, at least I got my laundry folded...)
Written by
on
in
Snaking.
The bug in SQLObject was just a missing '+' character, it was doing:
sql = ';\n'+join_sql
instead of:
sql += ';\n'+join_sql
so the SQL for the table was getting dropped when there was "join" sql for the final content. Somewhat of a letdown, I was thinking it would be hours of plunging through deep-dark code ...
Where'd that three hours go? (So much for buying my interest with smoothies...)
Written by
on
in
Snaking.
It seems that the smoothies were not sufficiently enticing to get me to actually sit down and work on the SQLObject problem. Shane's coming by in an hour to get his router re-provisioned, so not going to have a big block of time for work tonight. Maybe I should just call the day a write-off ...
Daily archives
- June 1, 2006
- June 2, 2006
- June 3, 2006
- June 4, 2006
- June 5, 2006
- June 6, 2006
- June 7, 2006
- June 8, 2006
- June 9, 2006
- June 10, 2006
- June 11, 2006
- June 12, 2006
- June 13, 2006
- June 14, 2006
- June 15, 2006
- June 16, 2006
- June 17, 2006
- June 18, 2006
- June 19, 2006
- June 20, 2006
- June 22, 2006
- June 23, 2006
- June 24, 2006
- June 25, 2006
- June 26, 2006
- June 27, 2006
- June 29, 2006
- June 30, 2006
Previous month
Next month
Archives
Mike C. Fletcher, the VRPlumber, is a philosopher of design turned computer programmer based in Toronto, Canada. He works primarily in the Python programming language, and maintains a large number of Open Source software packages. You can reach him at: mcfletch@vrplumber.com.