Who mapped where, when, and with whom?
The setup directory contains detailed information on installation for Linux and Mac OS X. The dependencies are many and somewhat unruly, but we promise it's worth it.
Once dependencies are all installed, running bundle install from the root of this repository should handle the rest. Note, if you bundle install, you most likely need to add bundle exec before all of the commands.
Additionally, there is a virtual machine by Tom Gertin (@d3nexter) which has the dependencies installed: https://github.com/d3netxer/osmbox.
An analysis window is a YAML document which defines file locations, database names, and the questions that you'd like to ask of the framework. A basic analysis window looks like this:
title: 'Chicago 2015'
pbf_file: '/data/osm_files/splitter_exports/north-america/illinois.osm.pbf'
database: chicago-2008-10
start_date: '2008-01-01'
end_date: '2010-12-31'
bbox: '-87.940101,41.643919,-87.523984,42.023022'
write_directory: '/data/www/chicago'
Node Questions:
- number_of_nodes_edited_by_new_mappers
- number_of_nodes_edited_by_experienced_mappers
Changeset Questions:
- changesets_x_day
User Questions:
- total_user_count
Bbox Questions:
- bbox_geojson_geometry
There are a series of rake tasks which run all of the software standalone if you want a command line usage. Otherwise, see section below on API usage.
All of the cutting and importing is bundled into one rake task:
rake new ~/my_osm_questions/my_analysis_window.yml
The individual processes that rake new runs can be run separately as:
rake cut # Write appropriate configuration file and cut ...
rake import:pbf # Import PBF File (Nodes, Ways, Relations)
rake import:changesets # Import Changesets
rake import:users # Import Users
rake import:notes # Import Notes
Once the data is imported to Mongo, you can begin to run the questions.
rake questions:all ~/my_osm_questions/my_analysis_window.yml
This task runs all of these questions tasks, which can also be called individually:
rake questions:bbox # Run BBox Questions
rake questions:changesets # Run Changeset Questions
rake questions:multi_users # Run Multi-User Questions
rake questions:nodes # Run Node Questions
rake questions:notes # Run Notes Questions
rake questions:users # Run User Questions
rake questions:ways # Run Way Questions
All of the questions were written out as JSON to the write_directory in the configuration file.
$ls /data/www/chicago/json
number_of_nodes_edited_by_experienced_mappers.json
number_of_nodes_edited_by_new_mappers.json
changesets_x_day.json
bbox_geojson_geometry.json
total_user_count.json
$cat data/www/chicago/json/total_user_count.json
{"Total User Count":238}
4. Use OSMDOWN to display results
An EPIC-OSM Oriented markdown language and compiler for generating static, interactive HTML docs with d3 and Leaflet (© OpenStreetMap Contributors):
After including the ruby library, you can call questions directly:
o = OSMHistory.new(analysis_window: 'sample-awconfig.yml')
aw = o.analysis_window
aw.bounding_box.geojson_geometry.to_json
# => {"type":"Polygon","coordinates" [[[-105.3424072,39.946595], ...
aw.new_contributors.count
# => 85
q = QuestionAsker.new(analysis_window: aw)
q.run('number_of_experienced_users')
# => Running Function: number_of_experienced_users...Up to QM...Success
The final command created the JSON file: number_of_experienced_users.json in the write_directory/json as specified in the sample-awconfig.yml.
Distributed Under the MIT License
© 2015 The Regents of the University of Colorado