I am developing app which needs to query a timetable. The user specifies
where they want to go from (start_station) and to (end_station) and the
time (start_time) they wish to leave.
My models are as follows:
Provdier -> Service -> Run -> Stop <- Station
-> = has many
Provider = Bus Company
Service = eg. Number 10 bus
Run = Column on timetable
Stop = Cell on timetable
Station = Row on timetable
Really provider and service dont come in to the equation and can be
ignored.
Station_A 09:45 11:23 12:01 2:30
Station_B 09:47 11:25 12:05 2:33
Station_C 09:55 11:31 12:15 2:37
Station_D 10:10 11:42 12:23 2:43
Station_E 10:25 11:45 12:30 2:51
So I need to find all runs (columns) where the end_station follows the
start_station.
This is the code I have at the moment, which finds two stations in the
same run, and does not take in to account either the start_time or
end_station coming after start_station. It works but is *very* database
intensive, and that with only a small amount of sample data. This stikes
me as a problem which has a cute solution.
# params[:search][:start_station].class => string
# params[:search][:end_station].class => string
# params[:search][:start_time].class => time
# Do actual search
@matches = Array.new
@start_station = Station.find(:first, :conditions => ''name =
"'' +
params[:search][:start_station] + ''"'', :include =>
:stops)
@end_station = Station.find(:first, :conditions => ''name =
"'' +
params[:search][:end_station] + ''"'')
@start_station.stops.each do |stop|
@joining_stops = stop.run.stops
@joining_stops.each do |stop|
if stop.station.id = @end_station.id
@matches.push(stop.run)
break
end
end
end
Can anyone suggest a better(simpler) way of approching this problem?
--
Posted via http://www.ruby-forum.com/.