fix: improve ietf doc string urlization. Fixes #3828. (#3831)

* fix: improve ietf doc string urlization. Fixes #3828.

* fix: dont urlize names immediately after - or /. Avoids urizing in the middle of urls.

* fix: cover some additional corners when urlifying document names.

* chore: remove unnecessary code complication

ietf/doc/templatetags/ietf_filters.py

@@ -185,6 +185,12 @@ def rfceditor_info_url(rfcnum : str):
     """Link to the RFC editor info page for an RFC"""
     return urljoin(settings.RFC_EDITOR_INFO_BASE_URL, f'rfc{rfcnum}')
 
+def link_non_charter_doc_match(match):
+    if len(match[3])==2 and match[3].isdigit():
+        return f'<a href="/doc/{match[2][:-1]}/{match[3]}/">{match[0]}</a>'
+    else:
+        return f'<a href="/doc/{match[2]}{match[3]}/">{match[0]}</a>'
+
 @register.filter(name='urlize_ietf_docs', is_safe=True, needs_autoescape=True)
 def urlize_ietf_docs(string, autoescape=None):
     """
@@ -192,8 +198,31 @@ def urlize_ietf_docs(string, autoescape=None):
     """
     if autoescape and not isinstance(string, SafeData):
         string = escape(string)
+    exp1 = r"\b(charter-(?:[\d\w\.+]+-)*)(\d\d-\d\d)(\.txt)?\b"
+    exp2 = r"\b(charter-(?:[\d\w\.+]+-)*)(\d\d)(\.txt)?\b"
+    if re.search(exp1, string):
+        string = re.sub(
+            exp1,
+            lambda x: f'<a href="/doc/{x[1][:-1]}/{x[2]}/">{x[0]}</a>',
+            string,
+            flags=re.IGNORECASE | re.ASCII,
+        )
+    elif re.search(exp2, string): 
+        string = re.sub(
+            exp2,
+            lambda x: f'<a href="/doc/{x[1][:-1]}/{x[2]}/">{x[0]}</a>',
+            string,
+            flags=re.IGNORECASE | re.ASCII,
+        )
     string = re.sub(
-        r"\b((RFC|BCP|STD|FYI|(?:draft-|bofreq-|conflict-review-|status-change-|charter-)[-\d\w.+]+)\s*0*(\d+))\b",
+        r"\b(?<![/-])(((?:draft-|bofreq-|conflict-review-|status-change-)(?:[\d\w\.+]+-)*)([\d\w\.+]+?)(\.txt)?)\b",
+        link_non_charter_doc_match,
+        string,
+        flags=re.IGNORECASE | re.ASCII,
+    )
+    string = re.sub(
+        # r"\b((RFC|BCP|STD|FYI|(?:draft-|bofreq-|conflict-review-|status-change-|charter-)[-\d\w.+]+)\s*0*(\d+))\b",
+        r"\b(?<!-)((RFC|BCP|STD|FYI)\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,

ietf/doc/templatetags/tests_ietf_filters.py

@@ -1,6 +1,9 @@
+# Copyright The IETF Trust 2022, All Rights Reserved
+
 from ietf.doc.templatetags.ietf_filters import urlize_ietf_docs
 from ietf.utils.test_utils import TestCase
 
+import debug # pyflakes: ignore
 # TODO: most other filters need test cases, too
 
 
@@ -18,10 +21,48 @@ class IetfFiltersTests(TestCase):
             ("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",
+                "draft-ietf-rfc9999-bis-01.txt",
-                '<a href="/doc/draft-ietf-rfc9999-bis-01/">draft-ietf-rfc9999-bis-01</a>',
+                '<a href="/doc/draft-ietf-rfc9999-bis/01/">draft-ietf-rfc9999-bis-01.txt</a>',
             ),
+            (
+                "foo RFC 9999 draft-ietf-rfc9999-bis-01 bar",
+                'foo <a href="/doc/rfc9999/">RFC 9999</a> <a href="/doc/draft-ietf-rfc9999-bis/01/">draft-ietf-rfc9999-bis-01</a> bar',
+            ),
+            (
+                "New version available: <b>draft-bryan-sipping-p2p-03.txt</b>",
+                'New version available: <b><a href="/doc/draft-bryan-sipping-p2p/03/">draft-bryan-sipping-p2p-03.txt</a></b>',
+            ),
+            (
+                "New version available: <b>charter-ietf-6man-04.txt</b>",
+                'New version available: <b><a href="/doc/charter-ietf-6man/04/">charter-ietf-6man-04.txt</a></b>'
+            ),
+            (
+                "New version available: <b>charter-ietf-6man-03-07.txt</b>",
+                'New version available: <b><a href="/doc/charter-ietf-6man/03-07/">charter-ietf-6man-03-07.txt</a></b>'
+            ),
+            (
+                "repository https://github.com/tlswg/draft-ietf-tls-ticketrequest",
+                'repository https://github.com/tlswg/draft-ietf-tls-ticketrequest'
+            ),
+            (
+                "draft-madanapalli-nd-over-802.16-problems",
+                '<a href="/doc/draft-madanapalli-nd-over-802.16-problems/">draft-madanapalli-nd-over-802.16-problems</a>' 
+            ),
+            (
+                 "draft-madanapalli-nd-over-802.16-problems-02.txt",
+                '<a href="/doc/draft-madanapalli-nd-over-802.16-problems/02/">draft-madanapalli-nd-over-802.16-problems-02.txt</a>'                
+            )
         ]
+        
+        # Some edge cases scraped from existing old draft names
+        for name in [
+            # "draft-odell-8+8", # This fails since + matches the right side of \b
+            # "draft-durand-gse+", # same failure
+            "draft-kim-xcast+-few-2-few",
+            #"draft-ietf-pem-ansix9.17",  # Fails because of not being greedy with . before txt
+        ]:
+            cases.append((name,f'<a href="/doc/{name}/">{name}</a>'))
 
         for input, output in cases:
+            #debug.show("(urlize_ietf_docs(input),output)")
             self.assertEqual(urlize_ietf_docs(input), output)