From 84d3a487a2654e8afa6a09c175bfd2e28ed8c5b9 Mon Sep 17 00:00:00 2001 From: Mark Sapiro Date: Mon, 22 Dec 2008 18:17:35 -0800 Subject: 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. --- Mailman/Archiver/HyperArch.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'Mailman') 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*$') -- cgit v1.2.3