From 954a02126eb21ca230319cd8ff62adf04d962de9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Adeodato=20Sim=C3=B3?= <dato@users.noreply.github.com>
Date: Thu, 2 Nov 2023 21:59:34 -0300
Subject: [PATCH] format_links: parse punctuation inside brackets

Also, consolidate all punctuation tests into a single table-driven one.
---
 bookwyrm/tests/views/test_status.py | 37 ++++++++++++++---------------
 bookwyrm/views/status.py            |  6 +++++
 2 files changed, 24 insertions(+), 19 deletions(-)

diff --git a/bookwyrm/tests/views/test_status.py b/bookwyrm/tests/views/test_status.py
index 67e1f6d76..424698130 100644
--- a/bookwyrm/tests/views/test_status.py
+++ b/bookwyrm/tests/views/test_status.py
@@ -421,25 +421,24 @@ http://www.fish.com/"""
         )
 
     def test_format_links_punctuation(self, *_):
-        """don’t take trailing punctuation into account pls"""
-        url = "http://www.fish.com/"
-        self.assertEqual(
-            views.status.format_links(f"{url}."),
-            f'<a href="{url}">www.fish.com/</a>.',
-        )
-        self.assertEqual(
-            views.status.format_links(f"{url}!?!"),
-            f'<a href="{url}">www.fish.com/</a>!?!',
-        )
-
-    def test_format_links_punctuation_parens(self, *_):
-        """ignore trailing punctuation and brackets combined"""
-        # Period at the end, wrapped in brackets.
-        url = "http://www.fish.com"
-        self.assertEqual(
-            views.status.format_links(f"({url})."),
-            f'(<a href="{url}">www.fish.com</a>).',
-        )
+        """test many combinations of brackets, URLs, and punctuation"""
+        url = "https://bookwyrm.social"
+        html = f'<a href="{url}">bookwyrm.social</a>'
+        test_table = [
+            ("punct", f"text and {url}.", f"text and {html}."),
+            ("multi_punct", f"text, then {url}?...", f"text, then {html}?..."),
+            ("bracket_punct", f"here ({url}).", f"here ({html})."),
+            ("punct_bracket", f"there [{url}?]", f"there [{html}?]"),
+            ("punct_bracket_punct", f"not here? ({url}!).", f"not here? ({html}!)."),
+            (
+                "multi_punct_bracket",
+                f"not there ({url}...);",
+                f"not there ({html}...);",
+            ),
+        ]
+        for desc, text, output in test_table:
+            with self.subTest(desc=desc):
+                self.assertEqual(views.status.format_links(text), output)
 
     def test_format_links_special_chars(self, *_):
         """find and format urls into a tags"""
diff --git a/bookwyrm/views/status.py b/bookwyrm/views/status.py
index 4c1d049df..8dab11a27 100644
--- a/bookwyrm/views/status.py
+++ b/bookwyrm/views/status.py
@@ -333,6 +333,12 @@ def _unwrap(text):
             # Split out wrapping chars.
             suffix = text[-1] + suffix
             prefix, text = text[:1], text[1:-1]
+            break  # Nested wrappers not supported atm.
+
+    if punct.search(text):
+        # Move inner punctuation to suffix segment.
+        text, inner_punct, _ = punct.split(text)
+        suffix = inner_punct + suffix
 
     return prefix, text, suffix