From 771494658a89070ef40db6c3f9921d20227dd5eb Mon Sep 17 00:00:00 2001
From: Driedsandwich <214981711+Driedsandwich@users.noreply.github.com>
Date: Fri, 14 Aug 2026 08:45:54 +0900
Subject: [PATCH] Canonicalize exported source URLs

---
 CHANGELOG.md                  | 1 +
 README.md                     | 2 +-
 docs/01_requirements.md       | 2 +-
 docs/03_status.md             | 3 ++-
 src/extractor/pageMetadata.js | 5 ++++-
 test/pageMetadata.test.mjs    | 4 ++--
 6 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index ee123e3..bdea4b2 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -15,6 +15,7 @@ All notable product changes will be recorded in this file.
   `textContent` fallback.
 - Added page title provenance to Markdown output.
 - Kept page titles inside a Markdown-inert untrusted-content fence.
+- Canonicalized supported source URLs without query parameters or fragments.
 - Added structural blockquote boundaries for untrusted body and comment
   previews.
 - Added regression checks for hidden content, Markdown structure, background
diff --git a/README.md b/README.md
index 11d1a3f..43ffe13 100644
--- a/README.md
+++ b/README.md
@@ -16,7 +16,7 @@ a complete GitHub conversation exporter.
 
 - recognizes GitHub Issue pages
 - recognizes GitHub Pull Request conversation pages
-- extracts the visible title, source URL, body preview, and visible comment snippets
+- extracts the visible title, canonical source URL without query/fragment data, body preview, and visible comment snippets
 - creates a Markdown preview with the title, explicit limitations, and a review-before-sharing warning
 - keeps the title inside a Markdown-inert fenced block and body/comment previews inside an explicit untrusted-content blockquote boundary
 - copies the Markdown to the clipboard
diff --git a/docs/01_requirements.md b/docs/01_requirements.md
index f3660c3..9502069 100644
--- a/docs/01_requirements.md
+++ b/docs/01_requirements.md
@@ -23,7 +23,7 @@ Pull Request support is limited to the Conversation page.
 - repository owner and name
 - Issue or Pull Request number
 - page type
-- source URL
+- canonical source URL without query parameters or fragments
 - export timestamp
 - exporter name
 
diff --git a/docs/03_status.md b/docs/03_status.md
index 11473ee..cb920e0 100644
--- a/docs/03_status.md
+++ b/docs/03_status.md
@@ -11,7 +11,8 @@ Maintainer-controlled public OSS preview.
 - Chrome Manifest V3 popup
 - GitHub Issue and Pull Request Conversation URL classification
 - self-contained visible-page metadata injection
-- visible title, source URL, body preview, and visible comment snippets
+- visible title, canonical source URL without query/fragment data, body
+  preview, and visible comment snippets
 - rendered-visibility checks that reject hidden, inert, transparent, or
   non-rendered body/comment elements
 - explicit body selectors that do not promote the first comment into a
diff --git a/src/extractor/pageMetadata.js b/src/extractor/pageMetadata.js
index f9fd01d..29b908b 100644
--- a/src/extractor/pageMetadata.js
+++ b/src/extractor/pageMetadata.js
@@ -52,6 +52,9 @@ export function extractVisiblePageMetadata() {
   const heading = normalizeText(document.querySelector("h1")?.innerText);
   const path = document.location.pathname;
   const pageKind = classifyPageKind(path);
+  const canonicalUrl = pageKind
+    ? `https://github.com${path.replace(/\/$/, "")}`
+    : "";
 
   let visibleContentPreview = "";
   let visibleContentStatus = "not_applicable";
@@ -108,7 +111,7 @@ export function extractVisiblePageMetadata() {
   return {
     title,
     heading,
-    url: document.location.href,
+    url: canonicalUrl,
     visibleContentPreview,
     visibleContentStatus,
     visibleComments,
diff --git a/test/pageMetadata.test.mjs b/test/pageMetadata.test.mjs
index 6d2a638..0c281fa 100644
--- a/test/pageMetadata.test.mjs
+++ b/test/pageMetadata.test.mjs
@@ -63,7 +63,7 @@ test("runs as a self-contained injected function for an Issue page", () => {
     kind: "issue",
     title: "Example issue · GitHub",
     heading: "Example issue",
-    url: "https://github.com/octo-org/example/issues/123",
+    url: "https://github.com/octo-org/example/issues/123?notification_referrer_id=synthetic#issuecomment-synthetic",
     body: "Visible issue body",
     comments: ["First visible comment"]
   }));
@@ -80,7 +80,7 @@ test("runs as a self-contained injected function for a Pull Request page", () =>
     kind: "pull request",
     title: "Example pull request · GitHub",
     heading: "Example pull request",
-    url: "https://github.com/octo-org/example/pull/456",
+    url: "https://github.com/octo-org/example/pull/456/?synthetic=1#discussion_rsynthetic",
     body: "Visible pull request body",
     comments: ["Review conversation comment"]
   }));