From 9f0a61a3f026019d8705c5666fdeab630480fab3 Mon Sep 17 00:00:00 2001
From: Yarmo Mackenbach <yarmo@yarmo.eu>
Date: Sat, 10 Sep 2022 12:14:05 +0200
Subject: [PATCH] Add util to get URLs from string

---
 src/utils.js       | 35 +++++++++++++++++++++++++++++++++++
 test/utils.test.js | 21 +++++++++++++++++++++
 2 files changed, 56 insertions(+)

diff --git a/src/utils.js b/src/utils.js
index 56fb7c0..9b945f0 100644
--- a/src/utils.js
+++ b/src/utils.js
@@ -63,5 +63,40 @@ const generateClaim = (fingerprint, format) => {
   }
 }
 
+/**
+ * Get the URIs from a string and return them as an array
+ * @param {string} text         - The text that may contain URIs
+ * @returns {Array.string}
+ */
+const getUriFromString = (text) => {
+  const re = /((([A-Za-z0-9]+:(?:\/\/)?)(?:[-;:&=+$,\w]+@)?[A-Za-z0-9.-]+|(?:www\.|[-;:&=+$,\w]+@)[A-Za-z0-9.-]+)((?:\/[+~%/.\w\-_]*)?\??(?:[-+=&;%@.\w_]*)#?(?:[.!/\\\w]*))?)/gi
+  const res = text.match(re)
+
+  const urls = []
+
+  if (!res) {
+    return []
+  }
+
+  res.forEach(url => {
+    // Remove bad trailing characters
+    let hasBadTrailingChars = true
+
+    while (hasBadTrailingChars) {
+      const lastChar = url.charAt(url.length - 1)
+      if ('?!.'.indexOf(lastChar) === -1) {
+        hasBadTrailingChars = false
+        continue
+      }
+      url = url.substring(0, url.length - 1)
+    }
+
+    urls.push(url)
+  })
+
+  return urls
+}
+
 exports.generateProxyURL = generateProxyURL
 exports.generateClaim = generateClaim
+exports.getUriFromString = getUriFromString
diff --git a/test/utils.test.js b/test/utils.test.js
index 9c52103..616f10c 100644
--- a/test/utils.test.js
+++ b/test/utils.test.js
@@ -18,6 +18,14 @@ const expect = chai.expect
 
 const doipjs = require('../src')
 
+const textWithUrls = `This is text with URLs like https://domain.tld. Ow, a trailing dot.
+What about (https://between.parentheses)? What about [https://between.brackets]?
+What about https://in.question? What about https://in.exclamation!
+And openpgp4fpr:123123, nonsense:123123`
+const urlsFromText = ["https://domain.tld", "https://between.parentheses",
+  "https://between.brackets", "https://in.question", "https://in.exclamation",
+  "openpgp4fpr:123123", "nonsense:123123"]
+
 describe('utils.generateClaim', () => {
   it('should be a function (2 arguments)', () => {
     expect(doipjs.utils.generateClaim).to.be.a('function')
@@ -57,3 +65,16 @@ describe('utils.generateProxyURL', () => {
     ).to.equal('https://localhost/api/2/get/dns?domain=domain.org')
   })
 })
+
+describe('utils.getUriFromString', () => {
+  it('should be a function (1 arguments)', () => {
+    expect(doipjs.utils.getUriFromString).to.be.a('function')
+    expect(doipjs.utils.getUriFromString).to.have.length(1)
+  })
+  it('should extract URLs from text', () => {
+    expect(
+      doipjs.utils.getUriFromString(textWithUrls)
+    ).to.have.length(urlsFromText.length)
+  })
+  // TODO Properly check each URL
+})