Created
April 20, 2012 20:42
-
-
Save akdh/2431767 to your computer and use it in GitHub Desktop.
Generates contexts
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
require 'iconv' | |
def content_to_xml(context) | |
"<context number=\"#{context[:number]}\"> | |
<city>#{context[:city]}</city> | |
<state>#{context[:state]}</state> | |
<lat>#{context[:lat]}</lat> | |
<long>#{context[:long]}</long> | |
<day>#{context[:day]}</day> | |
<time>#{context[:time]}</time> | |
<season>#{context[:season]}</season> | |
</context>" | |
end | |
def weighted_sample(lst) | |
w = Random.new.rand(1..lst.last[:pop_acc]) | |
lst.find do |item| | |
w <= item[:pop_acc] && item[:population] != 0 | |
end | |
end | |
if ARGV.length < 1 | |
puts "usage: FILE [count] [population_min]" | |
exit | |
end | |
# file comes from http://download.geonames.org/export/dump/ | |
filename = ARGV[0] | |
count = [ARGV[1].to_i, 1].max | |
pop_min = ARGV[2] || 100000 | |
content = IO.read(filename) | |
contnet = Iconv.conv('UTF8//UNKNOWN', 'UTF-8', content) | |
content = content.lines.map { |line| line.split("\t") } | |
content = content.each { |cols| cols[14] = cols[14].to_i } | |
content = content.find_all { |cols| cols[6] == 'P' && cols[14] > pop_min } | |
cities = content.map do |cols| | |
{ | |
:city => cols[1], | |
:state => cols[10], | |
:population => cols[14], | |
:lat => cols[4], | |
:long => cols[5], | |
:contexts => [] | |
} | |
end | |
days = %w(weekday weekend) | |
seasons = %w(spring summer fall winter) | |
times = %w(morning afternoon evening) | |
contexts = days.map do |day| | |
seasons.map do |season| | |
times.map do |time| | |
{ :day => day, :season => season, :time => time } | |
end | |
end | |
end.flatten | |
cities.inject(0) do |pop_sum, city| | |
result = pop_sum + city[:population] * contexts.length | |
city[:pop_acc] = result | |
result | |
end | |
(1..count).each do |number| | |
city = weighted_sample(cities) | |
city[:contexts] << (contexts - city[:contexts]).sample | |
if (contexts - city[:contexts]).empty? | |
index = cities.index(city) | |
cities[index..-1].each do |item| | |
item[:pop_acc] -= city[:population] | |
end | |
city[:population] = 0 | |
end | |
end | |
contexts = cities.map do |city| | |
city[:contexts].map do |context| | |
puts context | |
{ | |
:city => city[:city], | |
:state => city[:state], | |
:lat => city[:lat], | |
:long => city[:long], | |
:day => context[:day], | |
:time => context[:time], | |
:season => context[:season] | |
} | |
end | |
end | |
contexts = contexts.flatten.shuffle | |
contexts.each_with_index do |context, number| | |
context[:number] = number + 1 | |
end | |
contexts.each do |context| | |
puts content_to_xml(context) | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment