Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix psl parsing #52

Merged
merged 4 commits into from
Jul 29, 2024
Merged
Show file tree
Hide file tree
Changes from 3 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 10 additions & 5 deletions src/trackers/helpers/url.js
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ class ParsedURL extends URL {
}

get domainInfo() {
// extend domainInfo to use PSL
// extend domainInfo to use PSL
if (!this._domainInfo) {
this._domainInfo = parse(this.hostname, {
extractHostname: false,
Expand All @@ -25,7 +25,12 @@ class ParsedURL extends URL {
if (!this._domainInfo.isPrivate && pslExtras && pslExtras.privatePSL) {
// get list of possible suffix matches, we can have multiple matches for a single request
// a.example.com and b.a.example.com for a request from 123.b.a.example.com
const suffixMatches = pslExtras.privatePSL.filter(suffix => this._domainInfo.hostname.match(suffix))
// check that suffix is preceded by a dot or is at the beginning of the hostname
const suffixMatches = pslExtras.privatePSL.filter(suffix => {
const escapedSuffix = suffix.replace('.', '\\.')
const regex = new RegExp(`(^|\\.)${escapedSuffix}$`)
return regex.test(this._domainInfo.hostname)
})

// reformat domainInfo to make this request look like a private domain
if (suffixMatches && suffixMatches.length) {
Expand All @@ -34,13 +39,13 @@ class ParsedURL extends URL {
const suffix = suffixMatches.reduce((l,s) => {
return l.length >= s.length ? l : s
})

// Array of subdomain after removing suffix from hostname
const splitSubdomain = this._domainInfo.hostname.replace(suffix, '').replace(/\.$/,'').split('.')
const splitSubdomain = this._domainInfo.hostname.replace(new RegExp(`\\.?${suffix}$`), '').split('.')
const domainWithoutSuffix = splitSubdomain.pop()

this._domainInfo.publicSuffix = suffix
this._domainInfo.domain = `${domainWithoutSuffix}.${suffix}`
this._domainInfo.domain = domainWithoutSuffix ? `${domainWithoutSuffix}.${suffix}` : suffix
this._domainInfo.domainWithoutSuffix = domainWithoutSuffix
this._domainInfo.subdomain = splitSubdomain.join('.')
this._domainInfo.isPrivate = true
Expand Down
140 changes: 140 additions & 0 deletions test/custom-psl-paring.test.js
Original file line number Diff line number Diff line change
@@ -0,0 +1,140 @@
const assert = require('assert');
const URL = require('../src/trackers/helpers/url.js');

const testCases = [
{
input: "media-rockstargames-com.akamaized.net",
expectedOutput: {
publicSuffix: "akamaized.net",
domain: "media-rockstargames-com.akamaized.net",
domainWithoutSuffix: "media-rockstargames-com",
subdomain: "",
isPrivate: true
}
},
{
input: "sub.media-rockstargames-com.akamaized.net",
expectedOutput: {
publicSuffix: "akamaized.net",
domain: "media-rockstargames-com.akamaized.net",
domainWithoutSuffix: "media-rockstargames-com",
subdomain: "sub",
isPrivate: true
}
},
{
input: "example.akamaihd.net",
expectedOutput: {
publicSuffix: "akamaihd.net",
domain: "example.akamaihd.net",
domainWithoutSuffix: "example",
subdomain: "",
isPrivate: true
}
},
{
input: "a.b.c.d.e.example.akamaihd.net",
expectedOutput: {
publicSuffix: "akamaihd.net",
domain: "example.akamaihd.net",
domainWithoutSuffix: "example",
subdomain: "a.b.c.d.e",
isPrivate: true
}
},
{
input: "example.com",
expectedOutput: {
publicSuffix: "com",
domain: "example.com",
domainWithoutSuffix: "example",
subdomain: "",
isPrivate: false
}
},
{
input: "sub.test.edgekey-staging.net",
expectedOutput: {
publicSuffix: "test.edgekey-staging.net",
domain: "sub.test.edgekey-staging.net",
domainWithoutSuffix: "sub",
subdomain: "",
isPrivate: true
}
},
{
input: "bucket.s3.us-west-2.amazonaws.com",
expectedOutput: {
publicSuffix: "s3.us-west-2.amazonaws.com",
domain: "bucket.s3.us-west-2.amazonaws.com",
domainWithoutSuffix: "bucket",
subdomain: "",
isPrivate: true
}
},
{
input: "example.ssl.global.fastly.net",
expectedOutput: {
publicSuffix: "ssl.global.fastly.net",
domain: "example.ssl.global.fastly.net",
domainWithoutSuffix: "example",
subdomain: "",
isPrivate: true
}
},
{
input: "example.x.incapdns.net",
expectedOutput: {
publicSuffix: "x.incapdns.net",
domain: "example.x.incapdns.net",
domainWithoutSuffix: "example",
subdomain: "",
isPrivate: true
}
},
{
input: "example.trafficmanager.net",
expectedOutput: {
publicSuffix: "trafficmanager.net",
domain: "example.trafficmanager.net",
domainWithoutSuffix: "example",
subdomain: "",
isPrivate: true
}
},
{
input: "akamaized.net",
expectedOutput: {
publicSuffix: "akamaized.net",
domain: "akamaized.net",
domainWithoutSuffix: "",
subdomain: "",
isPrivate: true
}
},
{
input: "example.com.akamaized.net",
expectedOutput: {
publicSuffix: "com.akamaized.net",
domain: "example.com.akamaized.net",
domainWithoutSuffix: "example",
subdomain: "",
isPrivate: true
}
}
];

describe('PSL domain parsing', () => {
describe('parse domain with custom PSL', () => {
testCases.forEach(({ input, expectedOutput }) => {
it(`should correctly parse ${input}`, () => {
const result = new URL('https://' + input).domainInfo
assert.strictEqual(result.publicSuffix, expectedOutput.publicSuffix)
assert.strictEqual(result.domain, expectedOutput.domain)
assert.strictEqual(result.domainWithoutSuffix, expectedOutput.domainWithoutSuffix)
assert.strictEqual(result.subdomain, expectedOutput.subdomain)
assert.strictEqual(result.isPrivate, expectedOutput.isPrivate)
});
});
});
});
Loading