Use linkify-it

This commit is contained in:
Pavel Djundik 2018-04-26 19:03:33 +03:00
parent c7c2587079
commit 58ec2768ec
5 changed files with 172 additions and 51 deletions

View File

@ -1,13 +1,12 @@
"use strict";
const URI = require("urijs");
const linkify = require("linkify-it")()
.tlds(require("tlds"))
.tlds("onion", true);
// Known schemes to detect in a text. If a text contains `foo...bar://foo.com`,
// the parsed scheme should be `foo...bar` but if it contains
// `foo...http://foo.com`, we assume the scheme to extract will be `http`.
// Known schemes to detect in text
const commonSchemes = [
"http", "https",
"ftp", "sftp",
"sftp",
"smb", "file",
"irc", "ircs",
"svn", "git",
@ -15,54 +14,22 @@ const commonSchemes = [
"svn+ssh", "ssh",
];
for (const schema of commonSchemes) {
linkify.add(schema + ":", "http:");
}
function findLinks(text) {
const result = [];
const matches = linkify.match(text);
// URI.withinString() identifies URIs within text, e.g. to translate them to
// <a>-Tags.
// See https://medialize.github.io/URI.js/docs.html#static-withinString
// In our case, we store each URI encountered in a result array.
try {
URI.withinString(text, function(url, start, end) {
let parsedScheme;
try {
// Extract the scheme of the URL detected, if there is one
parsedScheme = URI(url).scheme().toLowerCase();
} catch (e) {
// URI may throw an exception for malformed urls,
// as to why withinString finds these in the first place is a mystery
return;
}
// Check if the scheme of the detected URL matches a common one above.
// In a URL like `foo..http://example.com`, the scheme would be `foo..http`,
// so we need to clean up the end of the scheme and filter out the rest.
const matchedScheme = commonSchemes.find((scheme) => parsedScheme.endsWith(scheme));
// A known scheme was found, extract the unknown part from the URL
if (matchedScheme) {
const prefix = parsedScheme.length - matchedScheme.length;
start += prefix;
url = url.slice(prefix);
}
// The URL matched but does not start with a scheme (`www.foo.com`), add it
if (!parsedScheme.length) {
url = "http://" + url;
}
result.push({
start: start,
end: end,
link: url,
});
});
} catch (e) {
// withinString is wrapped in a try/catch due to https://github.com/medialize/URI.js/issues/359
if (!matches) {
return [];
}
return result;
return matches.map((url) => ({
start: url.index,
end: url.lastIndex,
link: url.url,
}));
}
module.exports = findLinks;

View File

@ -46,6 +46,7 @@
"express": "4.16.3",
"fs-extra": "6.0.1",
"irc-framework": "2.11.0",
"linkify-it": "2.0.3",
"lodash": "4.17.10",
"mime-types": "2.1.18",
"moment": "2.22.1",
@ -58,6 +59,7 @@
"spdy": "3.4.7",
"sqlite3": "4.0.0",
"thelounge-ldapjs-non-maintained-fork": "1.0.2",
"tlds": "1.203.1",
"ua-parser-js": "0.7.18",
"urijs": "1.19.1",
"uuid": "3.2.1",

View File

@ -104,6 +104,144 @@ describe("findLinks", () => {
expect(actual).to.deep.equal(expected);
});
it("should find domains without www. but valid tld", () => {
const input = "google.com google.lv google.museum";
const expected = [{
link: "http://google.com",
start: 0,
end: 10,
}, {
link: "http://google.lv",
start: 11,
end: 20,
}, {
link: "http://google.museum",
start: 21,
end: 34,
}];
const actual = findLinks(input);
expect(actual).to.deep.equal(expected);
});
it("should find .onion domains", () => {
const input = "facebookcorewwwi.onion/test?url";
const expected = [{
link: "http://facebookcorewwwi.onion/test?url",
start: 0,
end: 31,
}];
const actual = findLinks(input);
expect(actual).to.deep.equal(expected);
});
it("should not consider invalid TLDs as domains", () => {
const input = "google.wtfgugl google.xx www.google.wtfgugl www.google.xx";
const expected = [];
const actual = findLinks(input);
expect(actual).to.deep.equal(expected);
});
it("should consider invalid TLDs as domains if protocol is specified", () => {
const input = "http://google.wtfgugl http://google.xx http://www.google.wtfgugl http://www.google.xx";
const expected = [{
link: "http://google.wtfgugl",
start: 0,
end: 21,
}, {
link: "http://google.xx",
start: 22,
end: 38,
}, {
link: "http://www.google.wtfgugl",
start: 39,
end: 64,
}, {
link: "http://www.google.xx",
start: 65,
end: 85,
}];
const actual = findLinks(input);
expect(actual).to.deep.equal(expected);
});
it("should correctly stop at punctuation", () => { // Issue #2351
const input =
"https://en.wikipedia.org/wiki/Dig! " +
"https://en.wikipedia.org/wiki/Dig? " +
"https://en.wikipedia.org/wiki/Dig. " +
"https://www.google.com* " +
"https://www.google.com/test* " +
"https://www.google.com@ " +
"https://www.google.com/test@ " +
"https://www.google.com! ";
const expected = [{
link: "https://en.wikipedia.org/wiki/Dig",
start: 0,
end: 33,
}, {
link: "https://en.wikipedia.org/wiki/Dig",
start: 35,
end: 68,
}, {
link: "https://en.wikipedia.org/wiki/Dig",
start: 70,
end: 103,
}, {
link: "https://www.google.com",
start: 105,
end: 127,
}, {
link: "https://www.google.com/test*",
start: 129,
end: 157,
}, {
link: "https://www.google.com",
start: 158,
end: 180,
}, {
link: "https://www.google.com/test@",
start: 182,
end: 210,
}, {
link: "https://www.google.com",
start: 211,
end: 233,
}];
const actual = findLinks(input);
expect(actual).to.deep.equal(expected);
});
it("should correctly stop at apostrophe", () => {
const input = "https://www.google.com's www.google.com's google.com's"; // Issue #1302
const expected = [{
link: "https://www.google.com",
start: 0,
end: 22,
}, {
link: "http://www.google.com",
start: 25,
end: 39,
}, {
link: "http://google.com",
start: 42,
end: 52,
}];
const actual = findLinks(input);
expect(actual).to.deep.equal(expected);
});
it("does not find invalid urls", () => {
const input = "www.example.com ssh://-oProxyCommand=whois"; // Issue #1412
const expected = [{

View File

@ -7,7 +7,7 @@ describe("parse Handlebars helper", () => {
it("should not introduce xss", () => {
const testCases = [{
input: "<img onerror='location.href=\"//youtube.com\"'>",
expected: "&lt;img onerror&#x3D;&#x27;location.href&#x3D;&quot;//youtube.com&quot;&#x27;&gt;",
expected: "&lt;img onerror&#x3D;&#x27;location.href&#x3D;&quot;<a href=\"//youtube.com\" target=\"_blank\" rel=\"noopener\">//youtube.com</a>&quot;&#x27;&gt;",
}, {
input: '#&">bug',
expected: '<span class="inline-channel" role="button" tabindex="0" data-chan="#&amp;&quot;&gt;bug">#&amp;&quot;&gt;bug</span>',

View File

@ -4310,6 +4310,12 @@ levn@^0.3.0, levn@~0.3.0:
prelude-ls "~1.1.2"
type-check "~0.3.2"
linkify-it@2.0.3:
version "2.0.3"
resolved "https://registry.yarnpkg.com/linkify-it/-/linkify-it-2.0.3.tgz#d94a4648f9b1c179d64fa97291268bdb6ce9434f"
dependencies:
uc.micro "^1.0.1"
listr-silent-renderer@^1.1.1:
version "1.1.1"
resolved "https://registry.yarnpkg.com/listr-silent-renderer/-/listr-silent-renderer-1.1.1.tgz#924b5a3757153770bf1a8e3fbf74b8bbf3f9242e"
@ -7446,6 +7452,10 @@ timers-browserify@^2.0.4:
dependencies:
setimmediate "^1.0.4"
tlds@1.203.1:
version "1.203.1"
resolved "https://registry.yarnpkg.com/tlds/-/tlds-1.203.1.tgz#4dc9b02f53de3315bc98b80665e13de3edfc1dfc"
tmp@^0.0.33:
version "0.0.33"
resolved "https://registry.yarnpkg.com/tmp/-/tmp-0.0.33.tgz#6d34335889768d21b2bcda0aa277ced3b1bfadf9"
@ -7559,6 +7569,10 @@ ua-parser-js@0.7.18:
version "0.7.18"
resolved "https://registry.yarnpkg.com/ua-parser-js/-/ua-parser-js-0.7.18.tgz#a7bfd92f56edfb117083b69e31d2aa8882d4b1ed"
uc.micro@^1.0.1:
version "1.0.5"
resolved "https://registry.yarnpkg.com/uc.micro/-/uc.micro-1.0.5.tgz#0c65f15f815aa08b560a61ce8b4db7ffc3f45376"
uglify-es@^3.3.4:
version "3.3.9"
resolved "https://registry.yarnpkg.com/uglify-es/-/uglify-es-3.3.9.tgz#0c1c4f0700bed8dbc124cdb304d2592ca203e677"