Freelance web app developer with a focus on data vis and science communication.
Public
Edited
Oct 19, 2023
Munging UN Demographics Data
file = FileAttachment("medium.csv").csv()
fileLow = FileAttachment("low.csv").csv()
fileHigh = FileAttachment("high.csv").csv()
PRIMARY_KEY = "Region, subregion, country or area *"
SECONDARY_KEY="Year"
VARIANT_KEY = "Variant"
LOCATION_CODE_KEY = "Location code"
FULL_YEARS = [2022, 2060, 2100]
VARIANTS = ["Medium", "Low", "High"]
regionKeys = file.reduce((acc, row) => {
const code = parseInt(row[LOCATION_CODE_KEY], 10);
const key = row[PRIMARY_KEY];
if (acc.includes(code) || !row.Year) {
return acc;
}
return acc.concat(code);
}, []);
secondaryKeys = file.reduce((acc, row) => {
const key = parseInt(row[SECONDARY_KEY], 10);
if (acc.includes(key) || isNaN(key)) {
return acc;
}
return acc.concat(key);
}, []);
rows = [...file, ...fileLow, ...fileHigh].map(row => {
const region = row[PRIMARY_KEY];
const variant = row[VARIANT_KEY];
const year = parseInt(row[SECONDARY_KEY]);
const regionKey = parseInt(row[LOCATION_CODE_KEY], 10);
if (isNaN(year)) {
return null
}
const pop = Object.keys(row).map(key => {
if (isNaN(parseInt(key, 10))) {
return null;
}
return parseInt(row[key].split(/\D+/).join(''), 10);
}).filter(val => val !== null);
return { region, variant, year, pop, regionKey }
}).filter(Boolean);
regions = regionKeys.map((key) => {
const row = rows.find(row => row.regionKey === key);
if (!row || !row.pop.reduce((acc, value) => acc + value, 0)) {
return null;
}
const name = row.region;
return {key, name};
}).filter(Boolean);
records = VARIANTS.reduce((acc, variant) => {
const records = regions.map(({ key: regionKey }) => {
const selected = rows.filter(row => row.regionKey === regionKey && row.variant === variant);
const regionName = selected[0].region;
const structure = secondaryKeys.map(year => {
const row = selected.find(row => row.year === year);
const total = row.pop.reduce((acc, value) => acc + value, 0);
return {
year,
amounts: row.pop,
fractions: row.pop.map(value => (value / total) || 0),
total,
dependencyRatio: popToDependencyRatio(row.pop)
}
});
return { variant, regionName: regionName, regionKey: regionKey, structure };
});
return [...acc, ...records];
}, []);
function popToDependencyRatio(pop) {
const dependent = [...pop.slice(0, 16), ...pop.slice(65)].reduce((total, item) => total + item, 0);
const working = pop.slice(16, 65).reduce((total, item) => total + item, 0);
const ratio = dependent / working;
return ratio && !isNaN(ratio) && ratio !== Infinity ? ratio : null;
}
records.filter(record => !record.structure[2].structure[99] === null);
rows.filter(row => row.regionKey === 900 && ['Low', 'High'].includes(row.variant)).map(row => ({
year: row.year,
variant: row.variant,
pop: row.pop.reduce((total, item) => total + item, 0),
}))
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.