print maps of stops coded by stoptime count

transitmapper.py

from gtfs import Schedule
from gtfs.entity import Trip, StopTime, Stop
from sqlalchemy.orm import eagerload
import csv
import sys

from prender import processing

def cons(ary):
  for i in range(len(ary)-1):
    yield ary[i],ary[i+1]
    
def find_freqs(sched):
  counts = {}
    
  n_trips = sched.session.query(Trip).count()

  for i in range(0,n_trips,500):
      print (i/float(n_trips))*100
      for trip in sched.session.query(Trip).options(eagerload("stop_times")).limit(500).offset(i):
          for st1,st2 in cons(sorted(trip.stop_times,key=lambda x:x.stop_sequence)):
              counts[(st1.stop_id,st2.stop_id)] = counts.get( (st1.stop_id,st2.stop_id), 0 ) + 1
              
  fp = open("freqs","w")
  for (st1,st2),count in counts.items():
    fp.write( "%s,%s,%s\n"%(st1,st2,count) )
  fp.close()
  
def map_line_freqs():
    sched = Schedule( "davepeck_20100806_2257.db" )
    
    ll,bb,rr,tt = sched.session.execute("SELECT min(stop_lon),min(stop_lat),max(stop_lon),max(stop_lat) from Stops").fetchone()
    
    
    mr = processing.MapRenderer()
    mr.start(ll,bb,rr,tt,4000) #left,bottom,right,top,width
    mr.background(255,255,255)
    mr.smooth()
    
    stops = dict([(stop.stop_id,stop) for stop in sched.stops])
    
    for st1_id, st2_id, count in csv.reader( open( "freqs" ) ):
        st1 = stops[st1_id]
        st2 = stops[st2_id]
        print st1.stop_lat,st1.stop_lon, st2.stop_lat,st2.stop_lon, count
        mr.strokeWeight( (float(count))*0.000008 )
        mr.line(st1.stop_lon,st1.stop_lat,st2.stop_lon,st2.stop_lat)
        
    mr.saveLocal("map.png")
    mr.stop()
    
def map_stop_freqs():
    sched = Schedule( "trimet.db" )
    
    ll,bb,rr,tt = sched.session.execute("SELECT min(stop_lon),min(stop_lat),max(stop_lon),max(stop_lat) from Stops").fetchone()
    
    mr = processing.MapRenderer()
    mr.start(ll,bb,rr,tt,4000) #left,bottom,right,top,width
    mr.background(255,255,255)
    mr.smooth()
    mr.strokeWeight( 0.0001 )
    
    stops = dict([(stop.stop_id,stop) for stop in sched.stops])
    
    counts = sched.session.execute( "select stop_id, count(*) from stop_times group by stop_id" )
    for stop_id, count in counts:
        print stop_id
        stop = stops[stop_id]
        mr.ellipse( stop.stop_lon, stop.stop_lat, 0.0001*count**0.5, 0.0001*count**0.5 )
        
    mr.saveLocal("trimet_stop_freqs.png")
    mr.stop()
    
def hyper_loader():
    from gtfs.feed import Feed
    
    feed = Feed( "/home/brandon/Desktop/wmata_20100822_0217.zip" )
    stop_times_table = feed.get_table( "stop_times.txt" )
    print stop_times_table
    
    for i, record in enumerate( stop_times_table ):
        if i%500==0: sys.stdout.write("."); sys.stdout.flush()
        pass
    sys.stdout.write("\n")

if __name__=='__main__':
    map_stop_freqs()