Dockerize ridership map computations

ridership_map/Dockerfile

@@ -0,0 +1,34 @@
+FROM debian:sid
+
+MAINTAINER Tristan Crockett <tristan.h.crockett@gmail.com>
+
+RUN groupadd -r transit && useradd -r -g transit transit
+
+RUN \
+  apt-get update && \
+  apt-get install -y wget unzip git
+RUN mkdir /app
+RUN mkdir /app/data && chown transit:transit /app/data
+VOLUME /app/data
+WORKDIR /app
+RUN wget http://gtfs.s3.amazonaws.com/chicago-transit-authority_20121108_0426.zip
+RUN unzip chicago-transit-authority_20121108_0426.zip
+
+RUN apt-get install -y postgresql-client
+RUN apt-get install -y python2.7
+RUN apt-get install -y python-psycopg2
+RUN apt-get install -y python-geojson
+
+RUN git clone https://github.com/thcrock/gtfs_SQL_IMPORTER.git
+
+COPY import.sh ./
+COPY create_foia_table.sql ./
+COPY create_segment_ridership.sql ./
+COPY join_riders_using_gtfs.py ./
+COPY create_geojson.py ./
+COPY run.sh ./
+
+RUN chmod +x import.sh
+RUN chmod +x run.sh
+RUN wget https://raw.githubusercontent.com/chihacknight/BetterTransit/master/ridership-data/CTA%20original%20bus%20ridership%20data%20for%20October%202012.csv
+RUN sed -i -e "1d" 'CTA original bus ridership data for October 2012.csv'

ridership_map/README.md

@@ -0,0 +1,11 @@
+The ridership map's input data computation process is now automated using Docker. The steps to run are:
+
+- Install Docker ( https://docs.docker.com/engine/installation/ )
+- Install Docker-Compose ( https://docs.docker.com/compose/install/ )
+- `cd ridership_map`
+- `mkdir data`
+- `chmod 777 data`
+- `docker-compose build` (the path to the docker-compose binary may be different than this, depending on how you installed)
+- `docker-compose up`
+
+Both of the last two steps will take a while, as many dependencies will be downloaded and installed during the 'build' command, and a lot of computation will happen during the 'up' command. You should see progress scroll by in your terminal window. Eventually, it should finish and you will have a new .geojson file in your data directory that the map will use.

ridership_map/create_foia_table.sql

@@ -0,0 +1,17 @@
+create table if not exists foia_data (
+	stop_id varchar,
+	on_street varchar,
+	cross_street varchar,
+	route varchar,
+	direction varchar,
+	boardings float,
+	alightings float,
+	month_beginning varchar,
+	daytype varchar,
+	lat varchar,
+	lon varchar
+);
+create index foia_direction on foia_data(direction);
+create index foia_route on foia_data(route);
+create index foia_stop on foia_data(stop_id);
+COPY foia_data from '/app/CTA original bus ridership data for October 2012.csv';

ridership_map/create_geojson.py

@@ -27,7 +27,7 @@ def get_data(cursor):
 	return list(reversed(cursor.fetchall()))
 
 if __name__ == '__main__':
-	conn = psycopg2.connect(database='transit', user='transit')
+	conn = psycopg2.connect(database='transit', user='postgres', host='db')
 	cursor = conn.cursor()
 	results = get_data(cursor)
 	features = []

ridership_map/create_segment_ridership.sql

@@ -0,0 +1,10 @@
+create table if not exists segment_ridership(
+	begin_lon varchar,
+	begin_lat varchar,
+	end_lon varchar,
+	end_lat varchar,
+	route varchar,
+	direction varchar,
+	riders integer,
+	primary key (begin_lon, begin_lat, end_lon, end_lat, route, direction)
+);

ridership_map/docker-compose.yml

@@ -0,0 +1,13 @@
+db:
+        image: postgres:9.4
+        ports:
+                - "5432"
+compute:
+        build: .
+        stdin_open: true
+        tty: true
+        entrypoint: /app/run.sh
+        volumes:
+                - ./data:/app/data
+        links: 
+                - db

ridership_map/import.sh

@@ -0,0 +1,8 @@
+#!/bin/bash
+cd gtfs_SQL_IMPORTER/src
+echo 'CREATE DATABASE transit' | psql -h db -U postgres 
+cat gtfs_tables.sql \
+  <(python2.7 import_gtfs_to_sql.py /app) \
+  gtfs_tables_makeindexes.sql \
+  vacuumer.sql \
+| psql -h db -U postgres transit

ridership_map/join_riders_using_gtfs.py

@@ -3,7 +3,7 @@
 import geojson
 import argparse
 
-conn = psycopg2.connect(database='transit', user='transit')
+conn = psycopg2.connect(database='transit', user='postgres', host='db')
 cursor = conn.cursor()
 
 

ridership_map/run.sh

@@ -0,0 +1,8 @@
+#!/bin/sh
+
+./import.sh
+cat create_foia_table.sql | psql -h db -U postgres transit
+cat CTA\ original\ bus\ ridership\ data\ for\ October\ 2012.csv | psql -h db -U postgres -c 'COPY foia_data from stdin csv' transit
+cat create_segment_ridership.sql | psql -h db -U postgres transit
+python2.7 join_riders_using_gtfs.py
+python2.7 create_geojson.py