ordat
ordat copied to clipboard
Chicago Data: CTA Train Tracker
This library is for accessing Chicago related data via Python. I'd prefer to focus on web APIs but I'm gonna go where the data is.
## CTA TraintrackerTo start, I'm implementing Train Tracker API. Below is a cursory usage example.
### To start up, just instantiate an instance of the Train class to initialize the connection to the API.>>> import ordat.cta as cta
>>> train = cta.Train(key='YOURKEYHERE')
### Get Arrivals by mapid
>>> train.arrivals(mapid=40380)
[Blue Line/Clark/Lake run 139 to O'Hare at 03:00:00, Blue Line/Clark/Lake run 225 to Forest Park at 03:06:52]
### Advanced Mode: Find All Scheduled/Active Runs
>>> # get arrivals for every station in the system
>>> all_arrivals = reduce(lambda a1,a2: a1+a2, [station.arrivals() for station in cta.Station.all])
>>> # get the run numbers of every active train in the system at the current time
>>> run_nums = set([arrival.run_number for arrival in all_arrivals])
>>> len(run_nums)
16
This was run late at night, so it found far less than normal!
### Find Current Train Locationsimport ordat.cta as cta
train = cta.Train(key='YOURKEYHERE')
tr = cta.panopticon.Tracker()
for rn,lat,lon in tr.step() :
print rn, lat, lon
This can take a bit to pick up the locations, so run the last two lines a couple times and wait in between.
### Station Search, Station Arrivals, Stop Arrivals>>> # Find a station, and then check its arrivals
>>> cta.Station.find('Clark/Lake')[0].arrivals()
[Blue Line/Clark/Lake run 139 to O'Hare at 03:00:00, Blue Line/Clark/Lake run 225 to Forest Park at 03:05:53]
>>> # Check what lines a station services
>>> cta.Station.find('Clark/Lake')[0].lines
[Pink Line, Blue Line, Purple Express Line, Orange Line, Brown Line, Green Line]
>>> cl_stops = cta.Station.find('Clark/Lake')[0].stops
>>> cl_stops
[E-bound Stop Clark/Lake (Inner Loop) at Station Clark/Lake, W-bound Stop Clark/Lake (Outer Loop) at Station Clark/Lake, S-bound Stop Clark/Lake (Forest Pk-bound) at Station Clark/Lake, N-bound Stop Clark/Lake (O'Hare-bound) at Station Clark/Lake]
>>> cl_stops[-1].arrivals()
[Blue Line/Clark/Lake run 139 to O'Hare at 03:00:00]
I intend to make the data more well formed and parsed, validated etc later. It implements a thin little caching layer to conserve API usage for requests for the same information placed in quick succession.
This is a quick wrapper around http://www.transitchicago.com/assets/1/developer_center/cta_Train_Tracker_API_documentation_v1_2.pdf
To add:
- Finding closest stations by lat/long, line, name, etc
- inbuilt code for acquiring data about stop-to-stop transit timings
- ability to understand delays, scheduled stops, faults in the system
- Unit testing for all of above
- Relate responses and requests on the arrivals API to my object hierarchy
- Color codes on lines
- Directionality of lines
- Determine 'next' stops for each stop (platform)
- git://github.com/martinblech/xmltodict.git
- Everything in requirements.txt
PYTHONPATH=../xmltodict:~/pyenvs/cta/lib/python2.7/site-packages/ nosetests -vv tests
The above is just an example, I'm not sure why virtualenv isn't picking up the requests module just by sourcing activate.