Changed the pattern used to recognize URLs in messages for the pipermail

archive in order to try to do a better job of making hyperlinks.
Bug #310124.

2 files changed, 10 insertions, 1 deletions

diff --git a/Mailman/Archiver/HyperArch.py b/Mailman/Archiver/HyperArch.py
index 33a389ed..ad51596e 100644
--- a/Mailman/Archiver/HyperArch.py
+++ b/Mailman/Archiver/HyperArch.py
@@ -156,7 +156,12 @@ REpat = re.compile( r"\s*RE\s*(\[\d+\]\s*)?:\s*", re.IGNORECASE)
 emailpat = re.compile(r'([-+,.\w]+@[-+.\w]+)')
 
 #  Argh!  This pattern is buggy, and will choke on URLs with GET parameters.
-urlpat = re.compile(r'(\w+://[^>)\s]+)') # URLs in text
+# MAS: Given that people are not constrained in how they write URIs in plain
+# text, it is not possible to have a single regexp to reliably match them.
+# The regexp below is intended to match straightforward cases.  Even humans
+# can't reliably tell whether various punctuation at the end of a URI is part
+# of the URI or not.
+urlpat = re.compile(r'([a-z]+://.*?)(?:_\s|_$|$|[]})>\'"\s])', re.IGNORECASE)
 
 # Blank lines
 blankpat = re.compile(r'^\s*$')
diff --git a/NEWS b/NEWS
index 010edc54..58a9867f 100644
--- a/NEWS
+++ b/NEWS
@@ -61,6 +61,10 @@ Here is a history of user visible changes to Mailman.
 
     - Corrected a typo in Mailman/Gui/Privacy.py. Bug #309757.
 
+    - Changed the pattern used to recognize URLs in messages for the pipermail
+      archive in order to try to do a better job of making hyperlinks.
+      Bug #310124.
+
 2.1.11 (30-Jun-2008)
 
   New Features