From 3e153a525c54bc7b9285c9a8f8b7834cfa9d3d1a Mon Sep 17 00:00:00 2001
From: Ohad Lutzky <ohad@lutzky.net>
Date: Sat, 14 Dec 2024 08:19:25 +0000
Subject: [PATCH] Fix trailing periods for naked URLs (#1175)

Fixes https://github.com/silverbulletmd/silverbullet/issues/1160
---
 common/markdown_parser/parser.test.ts | 23 +++++++++++++++++++++++
 common/markdown_parser/parser.ts      |  2 +-
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/common/markdown_parser/parser.test.ts b/common/markdown_parser/parser.test.ts
index f763460e..0c11be0b 100644
--- a/common/markdown_parser/parser.test.ts
+++ b/common/markdown_parser/parser.test.ts
@@ -216,3 +216,26 @@ Deno.test("Test hashtag helper functions", () => {
   // should behave like this for all characters in tagRegex
   assertEquals(renderHashtag("exclamation!"), "#<exclamation!>");
 });
+
+const nakedURLSample = `
+http://abc.com is a URL
+Also http://no-trailing-period.com. That's a URL. It ends with m, not '.'.
+http://no-trailing-comma.com, that same a URL, ends with m (and not ',').
+http://trailing-slash.com/. That ends with '/' (still not '.').
+http://abc.com?e=2.71,pi=3.14 is a URL too.
+http://abc.com?e=2.71. That is a URL, which ends with 1 (and not '.').
+`;
+
+Deno.test("Test NakedURL parser", () => {
+  const tree = parseMarkdown(nakedURLSample);
+  const urls = collectNodesOfType(tree, "NakedURL");
+
+  assertEquals(urls.map((x) => x.children![0].text), [
+    "http://abc.com",
+    "http://no-trailing-period.com",
+    "http://no-trailing-comma.com",
+    "http://trailing-slash.com/",
+    "http://abc.com?e=2.71,pi=3.14",
+    "http://abc.com?e=2.71",
+  ]);
+});
diff --git a/common/markdown_parser/parser.ts b/common/markdown_parser/parser.ts
index d7c5e327..ce9f5c9c 100644
--- a/common/markdown_parser/parser.ts
+++ b/common/markdown_parser/parser.ts
@@ -577,7 +577,7 @@ const NakedURL = regexParser(
   {
     firstCharCode: 104, // h
     regex:
-      /^https?:\/\/[-a-zA-Z0-9@:%._\+~#=,;]{1,256}([-a-zA-Z0-9()@:%_\+.,;~#?&=\/]*)/,
+      /(^https?:\/\/([-a-zA-Z0-9@:%_\+~#=]|(?:[.](?!(\s|$)))){1,256})(([-a-zA-Z0-9(@:%_\+~#?&=\/]|(?:[.,:;)](?!(\s|$))))*)/,
     nodeType: "NakedURL",
     className: "sb-naked-url",
     tag: NakedURLTag,