Mecklenburg County data for the Quality of Life Dashboard.
This project requires NodeJS.
git clone https://github.com/tobinbradley/mecklenburg-quality-of-life-data.git data
cd data
npm installIt isn't meant as a stand-alone repository, but rather to be used with the other Quality of Life projects.
There are several parts to configuring this repo:
- Creating your geography (GeoJSON)
- Creating your data files (CSV)
- Creating your metadata (Markdown)
- Updating configuration files
The existing files are a good guidepost for creating your own, so dig in before you get started.
The first thing you'll need is your geography, in GeoJSON. Your geography:
- Must be named
geography.geojson.jsonand placed in the root folder. - Must be WGS84 (EPSG:4326).
- Must contain an
idproperty that's a string and a unique identifier for each polygon that you'll use for your data files.
For serving and rendering GeoJSON, smaller is better, but watch out for topologically unaware simplification tools, as they'll leave ugly slivers in your data. You could use v.generalize in QGIS (or GRASS directly), or you could go the shapefile->topojson->geojson route.
Data files are simple CSV's in the format:
id,y_2000,y_2010
jim,23,432
suzy,,100
The file header is id and each year of data is expressed as y_<year>. No data values are empty, like suzy,,100. CSV files are processed into JSON by the consuming applications. Each metric file will be named with a metric number found in /config/data.js.
The type of data will decide the files required:
sum: The data is summed when polygons are selected. This will require ar<metric>.csvfile.mean: The data is averaged when polygons are selected. This will require an<metric>.csvfile.weighted: A weighted average is calculated when polygons are selected. This requires the raw data inr<metric>.csvand a denominator for weighting/calculations ind<metric>.csv. r/d is each individual polygon value.
After creating your data files, run the test suite to make sure the basics check out.
npm run test --silentMetadata files in markdown format are located in /meta. Each metadata file is named in format m<metric number>.md with heading tags exactly like this:
## Title of Metric
Median age of poodles
### Why is this important?
Because we like poodles.
### About the Data
I hang out at dog parks and type stuff in my phone. Circa 1986.
### Additional Resources
Dog pound yo.The markdown is processed to HTML by the consuming projects, and as there's a lot of ugly text processing going on, messing with the layout will involve tinkering with code.
There are four configuration files in /config:
data.json: Configuration information for your metrics.map.json: Configuration information for your maps.selectgroups.json: Groups of your geography units that you want to make selectable as a group.site.json: Configuration for the site, like titles etc.
Check the README.md in the config folder for configuration options.
The /config/legacy folder is for an earlier version of the project.
In particular, you will also want to change the Mapbox GL JS /gl-style/style.json pointer in main.js to your own tiles, Mapbox tiles, OSM2VectorTiles tiles, or what have you. The tiles we use don't cover much of the planet beyond our needs. The excellent OSM-Liberty is provided as a drop-in replacement.
Spreadsheet software often likes to capitalize the first letter in the y_XXXX field. That will turn your life into a furious ball of nothing. On a 'nix machine you can fix your files automatically via npm run clean-csv.
Your CSV files should be sorted on the metric ID. On a 'nix machine you can sort them automatically via npm run clean-csv.
Some identifiers like Census tracts can have a hanging zero, like 541.10. If you are manipulating your data files in spreadsheet software, make sure your ID column is being treated as a string and not a number or it will get dropped and your life will be destroyed.
Because we're using those as choppers for layout, adding more of those will screw up your formatting.
Don't edit Markdown in Word. You're welcome.