Created
October 19, 2022 16:08
-
-
Save seanghay/daff995b7f66dafe2ec9a39dd64079f5 to your computer and use it in GitHub Desktop.
Download Postal Code Info in Cambodia
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
import axios from 'axios' | |
import { load } from 'cheerio' | |
import fs from 'fs/promises' | |
import fse from 'fs-extra' | |
import Interpreter from 'js-interpreter' | |
// download provinces list | |
async function downloadProvinceList() { | |
const { data: html } = await axios.get('https://www.cambodiapost.post/page/postal-codes', { responseType: "text" }); | |
const $ = load(html); | |
const items = $("#city option").map(function () { | |
return { | |
id: +$(this).attr('value'), | |
name: $(this).text() | |
} | |
}).get(); | |
$('script').each(function() { | |
const text = $(this).text(); | |
if (/window\.windowvar\s=\swindow\.windowvar/.test(text)) { | |
const intr = new Interpreter(`${text}; windowvar.data`); | |
intr.run(); | |
const data = intr.pseudoToNative(intr.value); | |
fse.writeFileSync('data.json', JSON.stringify(data, null, 2), 'utf-8'); | |
} | |
}) | |
await fs.writeFile('provinces.json', JSON.stringify(items), 'utf-8'); | |
} | |
export async function parsePostalCollection(provinceId, districtNo) { | |
const { data: html } = await axios.get(`https://www.cambodiapost.post/page/postal-codes?city=${provinceId}&district=${districtNo}`, { responseType: 'text' }) | |
const $ = load(html) | |
const items = $('tbody.tbl-content tr').map(function () { | |
const km = $(this).find('th.kh-name').text() | |
const [en, code] = $(this).find('td').get() | |
return { | |
km, | |
en: $(en).text(), | |
code: $(code).text() | |
} | |
}).get() | |
return items; | |
} | |
export async function downloadPostalInfo() { | |
const provinces = JSON.parse(await fs.readFile('./provinces.json', 'utf-8')); | |
const districtsOfProvince = JSON.parse(await fs.readFile('./data.json', 'utf-8')); | |
const data = []; | |
for (const province of provinces) { | |
const p = { | |
...province, | |
districts: [], | |
} | |
data.push(p); | |
const districts = districtsOfProvince[province.id]; | |
for (const district of districts) { | |
const codes = await parsePostalCollection(province.id, district.no) | |
p.districts.push({ | |
...district, | |
codes, | |
}) | |
console.log(codes) | |
await fse.writeJSON('./out.json', data) | |
} | |
} | |
} | |
await downloadProvinceList() | |
await downloadPostalInfo() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment