Merge branch '5629-add_one_trackers_list' into 'main' (149fc182) · Commits · e / os / tracker-list

.gitlab-ci.yml

+1 −1

Original line number	Diff line number	Diff line
		image: "node:16.15.0-slim"
		image: "node:18.19-slim"

		stages:
		- deploy

README.md

+19 −11

Original line number	Diff line number	Diff line
		# tracker-list

		List of trackers, used by PrivacyCentral app to detect and filter trackers. The list is a merge of Exodus and Adaway trackers lists, with custom /e/ rules.
		List of trackers, used by PrivacyCentral app to detect and filter trackers. The list is a merge of Exodus, WhoTracksMe and StevenBlack trackers lists, with custom /e/ rules.
		The list is located in list/e_trackers.json

		## Auto upgrade list WIP
		The list is automatically upgraded each day (if modified) with a CI script.

		List can be manualy updated for now :
		## Manualy upgrade list

		List can be manualy updated :

		1. build a new list :
		```
		$ node build_list.js

		Starting list update
		adaway OK
		exodus OK
		enrich exodus OK
		Trackers list finished
		DONE
		Fetch and parse Whotracksme OK - 2404 trackers
		Fetch and parse Exodus OK - 260 trackers
		updateWithTrackerList - baseList 3784 with otherList 260
		updateWithTrackerList - remaining Others: 0
		updateWithTrackerList now total of baseList 3784
		updateWithTrackerList - baseList 3784 with otherList 2404
		updateWithTrackerList - remaining Others: 0
		updateWithTrackerList now total of baseList 3784
		1 hostnames lists with each : 141886 hostnames
		hostnames total : 141886
		0 trackers deprecated, to remove
		Trackers list finished, 3784 trackers, 21224 hostnames
		No changes in trackers, abort updating.

		```
		2. Chech validity of the list :

build_list.js

+138 −141

Original line number	Diff line number	Diff line
		const outputFile = "list/e_trackers.json"

		const adawayUri = "https://adaway.org/hosts.txt"
		const exodusUri = "https://reports.exodus-privacy.eu.org/api/trackers"
		const EXIT_CODE_NO_CHANGES = 5

		const whitelistedDomains = [
		"login.microsoftonline.com",
		]

		const https = require('https')
		const fs = require('fs')
		const url = require('url')

		const EXIT_CODE_NO_CHANGES = 5

		function fetch(url) {
		return new Promise((resolve, reject) => {
		https.get(url, (response) => {
		let data = ""
		response.on('data', (chunk) => { data += chunk })
		response.on('end', () => resolve(data))
		}).on('error', reject)
		})
		}
		const outputFile = "./list/e_trackers.json"

		const fs = require('fs')

		let adawayText = null
		let exodusJson = null
		const exodus = require('./src/exodus_trackers.js')
		const whotracksme = require('./src/whotracksme_trackers.js')
		const hostsfiles = require('./src/hostsfiles_hostnames.js')

		let hostnames = []

		let trackers = []
		const eTrackers = require("./" + outputFile).trackers.map(it => {
		it.dirty = true
		return it
		}).sort((a, b) => b.id.localeCompare(a.id))

		console.log("Starting list update")
		function remove(list, item) {
		list.splice(list.indexOf(item), 1)
		}

		fetch(adawayUri)
		.then(text => {
		adawayText = text
		hostnames = parseHostToHostnames(text)
		console.log("adaway OK")
		}).then(toto => {
		return fetch(exodusUri)
		})
		.then(response => { return JSON.parse(response) })
		.then(json => {
		exoduxJson = json
		console.log("exodus OK")
		})
		.then(toto => {
		trackers = exodusToETrackers(exoduxJson)
		function updateWithTrackerList(baseList, otherList) {
		const mergeTrackers = (base, other) => {
		base.hostnames = Array.from(new Set(base.hostnames.concat(other.hostnames)))
		base.name = other.name
		base.dirty = undefined
		base.link = (base.link) ? base.link : other.link
		base.exodusId = (base.exodusId) ? base.exodusId : other.exodusId
		}

		hostnames = applyOutbrainRule(trackers, hostnames)
		hostnames = enrichExodusWithAdaway(trackers, hostnames)
		console.log("enrich exodus OK")
		hostnames = createdLevel2Trackers(trackers, hostnames)
		console.log("define adaway trackers OK")
		removeWhitelistedHostnames(trackers)
		console.log("remove whitelisted hostnames OK")
		console.log("Trackers list finished")
		})
		.then(toto => {
		if (listHasNoChanges()) {
		process.exitCode = EXIT_CODE_NO_CHANGES
		throw "No changes in trackers, abort updating."
		console.log("updateWithTrackerList - baseList " + baseList.length + " with otherList " + otherList.length)
		baseList.forEach(tracker => {
		const sameId = otherList.find(candidate => candidate.id == tracker.id)
		if (sameId) {
		mergeTrackers(tracker, sameId)
		remove(otherList, sameId)
		} else {
		otherList.some(candidate => {
		if (tracker.hostnames.some(host => candidate.hostnames.some(h2 => h2 == host))) {
		mergeTrackers(tracker, candidate)
		remove(otherList, candidate)
		return true
		} else {
		return false
		}
		})
		.then(toto => {
		return printTrackers()
		})
		.then(trackersJson => {
		return saveTrackersFile(trackersJson)
		})
		.then(toto => {
		console.log("DONE")
		})
		.catch(err => {
		if (process.exitCode === undefined) {
		process.exitCode = 1
		}
		console.log(err)
		})

		console.log("updateWithTrackerList - remaining Others: " + otherList.length)

		function parseHostToHostnames(hostText) {
		return hostText
		.split("\n")
		.filter(line => { return !line.startsWith("#") && line != ""})
		.map(line => line.split(" ")[1])
		.filter(it => it != "")
		Array.prototype.push.apply(baseList, otherList)
		console.log("updateWithTrackerList now total of baseList " + baseList.length)
		}

		function exodusToETrackers(exodusTrackers) {
		return Object.values(exodusTrackers.trackers)
		.filter(tracker => tracker.network_signature != "")
		.map(tracker => {
		const id = "exodus_" + tracker.id
		const hostnames = tracker.network_signature
		.replace(/\\/g, "")
		.split("\|")
		return {
		id: id,
		hostnames: hostnames,
		name: tracker.name,
		exodusId: tracker.id
		function extractSubhostnamesToTracker(hostname, tracker, hostnameSet) {
		const toDelete = new Set()
		hostnameSet.forEach(subhost => {
		if (subhost.endsWith(hostname)) {
		if (!tracker.hostnames.includes(subhost)) {
		tracker.hostnames.push(subhost)
		}
		tracker.dirty = undefined
		toDelete.add(subhost)
		}
		})
		toDelete.forEach(it => hostnameSet.delete(it))
		}


		function applyOutbrainRule(trackersList, adawayList) {
		function applyOutbrainRule(trackersList, hostnameSet) {
		const tracker = trackersList.find(it => it.name === "OutBrain")
		if (tracker != null) {
		return extractSubhostnamesToTracker('outbrainimg.com', tracker, adawayList)
		extractSubhostnamesToTracker('outbrainimg.com', tracker, hostnameSet)
		}
		return adawayList
		}

		function extractSubhostnamesToTracker(hostname, tracker, adawayList) {
		let hostnamesSet = new Set(tracker.hostnames)
		adawayList.filter(it => it.endsWith(hostname))
		.forEach(subhost => hostnamesSet.add(subhost))
		tracker.hostnames = Array.from(hostnamesSet)
		return adawayList.filter(it => !it.endsWith(hostname))
		}

		function enrichExodusWithAdaway(trackersList, adawayList) {
		trackersList.forEach(tracker => {
		function enrichTrackersWithHostnames(trackerList, hostnameSet) {
		trackerList.forEach(tracker => {
		tracker.hostnames.forEach(hostname => {
		adawayList = extractSubhostnamesToTracker(hostname, tracker, adawayList)
		extractSubhostnamesToTracker(hostname, tracker, hostnameSet)
		})
		})
		return adawayList
		}

		function createdLevel2Trackers(trackersList, adawayList) {

		while(adawayList.length > 0) {
		let hostname = adawayList[0]
		//return adawayList.reduce((acc, hostname) => {

		let lv2 = toLevel2Domain(hostname)
		let tracker = {
		id: "adaway_" + lv2,
		name: lv2,
		hostnames: []
		}
		trackersList.push(tracker)
		adawayList = extractSubhostnamesToTracker(lv2, tracker, adawayList)
		}
		return adawayList
		}

		function toLevel2Domain(hostname) {
		try {
		let parts = hostname.split("\.")
		return parts.slice(parts.length - 2, parts.length).join(".")
		} catch {
		console.log(hostname)
		}
		function keepOnlyHostnamesInHostnamesSet(trackerList, hostnameSet) {
		console.log("hostnameSet.size : " + hostnameSet.size)
		const deleted = new Set()
		trackerList.forEach(tracker => {
		tracker.hostnames
		.filter(it => it.split(".").length == 2 && !hostnameSet.has(it))
		.forEach(it => {
		remove(tracker.hostnames, it)
		deleted.add(it)
		})
		})
		console.log("Deleted hostnames not in hostnames list : " + deleted.size)
		}

		function removeWhitelistedHostnames(trackersList) {
		@@ -173,19 +114,75 @@ function removeWhitelistedHostnames(trackersList) {
		function listHasNoChanges() {
		// Compare actual list with the builded one (without the createdAt field!)
		const oldTrackers = require("./" + outputFile).trackers
		return JSON.stringify(trackers) == JSON.stringify(oldTrackers)
		return JSON.stringify(eTrackers) == JSON.stringify(oldTrackers)
		}


		function printTrackers() {
		return JSON.stringify({
		trackers: trackers,
		trackers: eTrackers,
		createdAt: new Date().toISOString()
		})
		}



		function saveTrackersFile(trackersJson) {
		fs.writeFileSync(outputFile, trackersJson)
		}


		Promise.all([
		exodus.getTrackers(),
		whotracksme.getTrackers()
		]).then(trackerLists => {
		trackerLists.forEach(otherList => updateWithTrackerList(eTrackers, otherList))
		return eTrackers
		}).then(eTrackers => hostsfiles.getHostnames()
		).then(hostnameSet => {
		const hostnameSetToConsume = new Set(hostnameSet)
		applyOutbrainRule(eTrackers, hostnameSetToConsume)
		enrichTrackersWithHostnames(eTrackers, hostnameSetToConsume)

		// Remove 10% of the hostnames,
		// but should avoid unaccessible sites like adob.com or snapchat.com
		keepOnlyHostnamesInHostnamesSet(eTrackers, hostnameSet)
		removeWhitelistedHostnames(eTrackers)
		return eTrackers
		}).then(eTrackers => {
		const deprecatedTrackers = eTrackers.filter(it => it.dirty)
		console.log(deprecatedTrackers.length + " trackers deprecated, to remove")
		if (deprecatedTrackers.length > 0) {
		console.log(deprecatedTrackers.map(it => it.name))
		console.log("before clean up: " + eTrackers.length)
		eTrackers.filter(it => it.dirty).forEach(it => {
		remove(eTrackers, it)
		})
		console.log("after clean up: " + eTrackers.length)
		}

		const trackersCount = eTrackers.length
		const hostnamesCount = eTrackers.reduce((acc, tracker) => acc + tracker.hostnames.length, 0)
		console.log("Trackers list finished, " + trackersCount + " trackers, " + hostnamesCount + " hostnames")
		return eTrackers
		})
		.then(trackers => {
		if (listHasNoChanges()) {
		process.exitCode = EXIT_CODE_NO_CHANGES
		throw "No changes in trackers, abort updating."
		}
		})
		.then(toto => {
		return printTrackers()
		})
		.then(trackersJson => {
		return saveTrackersFile(trackersJson)
		})
		.then(toto => {
		console.log("DONE")
		})
		.catch(err => {
		if (process.exitCode === undefined) {
		process.exitCode = 1
		}
		console.log(err)
		})

list/e_trackers.json

+1 −1

File changed.

Preview size limit exceeded, changes collapsed.

src/exodus_trackers.js

0 → 100644

+30 −0

Original line number	Diff line number	Diff line
		const exodusUri = "https://reports.exodus-privacy.eu.org/api/trackers"

		const fetcher = require("./fetch_promise.js")

		function exodusToETrackers(exodusTrackers) {
		return Object.values(exodusTrackers.trackers)
		.filter(tracker => tracker.network_signature != "")
		.map(tracker => {
		const id = "exodus_" + tracker.id
		const hostnames = tracker.network_signature
		.replace(/\\/g, "")
		.split("\|")
		return {
		id: id,
		hostnames: hostnames,
		name: tracker.name,
		exodusId: tracker.id,
		link: "https://reports.exodus-privacy.eu.org/trackers/" + tracker.id
		}
		})
		}

		exports.getTrackers = () => fetcher.fetch(exodusUri)
		.then(response => {
		const exodusData = JSON.parse(response)
		const exodusTrackers = exodusToETrackers(exodusData)
		console.log("Fetch and parse Exodus OK - " + exodusTrackers.length + " trackers")
		return exodusTrackers
		})