From 79c7081181cef24e7b1cc44ee4738f9052e35465 Mon Sep 17 00:00:00 2001
From: Lars Eggert <lars@eggert.org>
Date: Thu, 14 Apr 2022 20:33:47 +0200
Subject: [PATCH] * fix: Correctly urlize bis documents

The old code would first substitute the "rfx<nnnn>" part, and then again
on the full draft name. This now only ever applies one substitute,
and does longer ones first.
---
 ietf/doc/templatetags/ietf_filters.py       | 17 ++++++-------
 ietf/doc/templatetags/tests_ietf_filters.py | 27 +++++++++++++++++++++
 2 files changed, 34 insertions(+), 10 deletions(-)
 create mode 100644 ietf/doc/templatetags/tests_ietf_filters.py

diff --git a/ietf/doc/templatetags/ietf_filters.py b/ietf/doc/templatetags/ietf_filters.py
index 58d487160..43f6717b5 100644
--- a/ietf/doc/templatetags/ietf_filters.py
+++ b/ietf/doc/templatetags/ietf_filters.py
@@ -192,15 +192,12 @@ def urlize_ietf_docs(string, autoescape=None):
     """
     if autoescape and not isinstance(string, SafeData):
         string = escape(string)
-    string = re.sub(r"(RFC\s*?)0{0,3}(\d+)", "<a href=\"/doc/rfc\\2/\">\\1\\2</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(BCP\s*?)0{0,3}(\d+)", "<a href=\"/doc/bcp\\2/\">\\1\\2</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(STD\s*?)0{0,3}(\d+)", "<a href=\"/doc/std\\2/\">\\1\\2</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(FYI\s*?)0{0,3}(\d+)", "<a href=\"/doc/fyi\\2/\">\\1\\2</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(draft-[-0-9a-zA-Z._+]+)", "<a href=\"/doc/\\1/\">\\1</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(bofreq-[-0-9a-zA-Z._+]+)", "<a href=\"/doc/\\1/\">\\1</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(conflict-review-[-0-9a-zA-Z._+]+)", "<a href=\"/doc/\\1/\">\\1</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(status-change-[-0-9a-zA-Z._+]+)", "<a href=\"/doc/\\1/\">\\1</a>", string, flags=re.IGNORECASE)
-    string = re.sub(r"(charter-[-0-9a-zA-Z._+]+)", "<a href=\"/doc/\\1/\">\\1</a>", string, flags=re.IGNORECASE)
+    string = re.sub(
+        r"\b((RFC|BCP|STD|FYI|(?:draft-|bofreq-|conflict-review-|status-change-|charter-)[-\d\w.+]+)\s*0*(\d+))\b",
+        lambda x: f'<a href="/doc/{x[2].strip().lower()}{x[3]}/">{x[1]}</a>',
+        string,
+        flags=re.IGNORECASE | re.ASCII,
+    )
     return mark_safe(string)
 urlize_ietf_docs = stringfilter(urlize_ietf_docs)
 
@@ -739,4 +736,4 @@ def absurl(viewname, **kwargs):
 
     Uses settings.IDTRACKER_BASE_URL as the base.
     """
-    return urljoin(settings.IDTRACKER_BASE_URL, urlreverse(viewname, kwargs=kwargs))
\ No newline at end of file
+    return urljoin(settings.IDTRACKER_BASE_URL, urlreverse(viewname, kwargs=kwargs))
diff --git a/ietf/doc/templatetags/tests_ietf_filters.py b/ietf/doc/templatetags/tests_ietf_filters.py
new file mode 100644
index 000000000..3db85f5f7
--- /dev/null
+++ b/ietf/doc/templatetags/tests_ietf_filters.py
@@ -0,0 +1,27 @@
+from ietf.doc.templatetags.ietf_filters import urlize_ietf_docs
+from ietf.utils.test_utils import TestCase
+
+# TODO: most other filters need test cases, too
+
+
+class IetfFiltersTests(TestCase):
+    def test_urlize_ietf_docs(self):
+        cases = [
+            ("no change", "no change"),
+            ("bcp1", '<a href="/doc/bcp1/">bcp1</a>'),
+            ("Std 003", '<a href="/doc/std3/">Std 003</a>'),
+            (
+                "FYI02 changes Std 003",
+                '<a href="/doc/fyi2/">FYI02</a> changes <a href="/doc/std3/">Std 003</a>',
+            ),
+            ("rfc2119", '<a href="/doc/rfc2119/">rfc2119</a>'),
+            ("Rfc 02119", '<a href="/doc/rfc2119/">Rfc 02119</a>'),
+            ("draft-abc-123", '<a href="/doc/draft-abc-123/">draft-abc-123</a>'),
+            (
+                "draft-ietf-rfc9999-bis-01",
+                '<a href="/doc/draft-ietf-rfc9999-bis-01/">draft-ietf-rfc9999-bis-01</a>',
+            ),
+        ]
+
+        for input, output in cases:
+            self.assertEqual(urlize_ietf_docs(input), output)