Fix strip_urls not stripping domains with dashes (fixes #32)

Before this commit, an url with dashes in the domain was not stripped
correctly by the botogram.utils.strip_urls functions, which caused
issues in the syntax detector (recognizing URLs as Markdown).

This commit fixes the regular expression used to strip URLs, and also
adds some unit tests for that function in order to prevent issues like
this one in the future.

Author: Pietro Albini

botogram/utils.py

@@ -22,7 +22,7 @@
 _username_re = re.compile(r"\@([a-zA-Z0-9_]{5}[a-zA-Z0-9_]*)")
 _command_re = re.compile(r"^\/[a-zA-Z0-9_]+(\@[a-zA-Z0-9_]{5}[a-zA-Z0-9_]*)?$")
 _email_re = re.compile(r"[a-zA-Z0-9_\.\+\-]+\@[a-zA-Z0-9_\.\-]+\.[a-zA-Z]+")
-_url_re = re.compile(r"https?://(-\.)?([^\s/?\.#-]+\.?)+(/[^\s]*)?")
+_url_re = re.compile(r"https?://(-\.)?([^\s/?\.#]+\.?)+(/[^\s]*)?")
 
 # This small piece of global state will track if logbook was configured
 _logger_configured = False

docs/changelog.rst

@@ -19,13 +19,16 @@ botogram 0.2
 * Renamed ``Bot.init_shared_memory`` to ``Bot.prepare_memory``
 * Renamed ``Component.add_shared_memory_initializer`` to
   ``Component.add_memory_preparer``
+* Fix the syntax detector checking URLs with dashes in the domain (`issue 32`_)
 
 The following things are now **deprecated**:
 
 * ``Bot.init_shared_memory``, and it will be removed in botogram 1.0
 * ``Component.add_shared_memory_initializer``, and it will be removed in
   botogram 1.0
 
+.. _issue 32: https://github.com/pietroalbini/botogram/issues/32
+
 .. _changelog-0.1.2:
 
 botogram 0.1.2

tests/test_utils.py

@@ -17,6 +17,32 @@
 """
 
 
+def test_strip_urls():
+    # Standard HTTP url
+    assert botogram.utils.strip_urls("http://www.example.com") == ""
+
+    # Standard HTTPS url
+    assert botogram.utils.strip_urls("https://www.example.com") == ""
+
+    # HTTP url with dashes in the domain (issue #32)
+    assert botogram.utils.strip_urls("http://www.ubuntu-it.org") == ""
+
+    # HTTP url with a path
+    assert botogram.utils.strip_urls("http://example.com/~john/d/a.txt") == ""
+
+    # Standard email address
+    assert botogram.utils.strip_urls("[email protected]") == ""
+
+    # Email address with a comment (+something)
+    assert botogram.utils.strip_urls("[email protected]") == ""
+
+    # Email address with subdomains
+    assert botogram.utils.strip_urls("[email protected]") == ""
+
+    # Email address with dashes in the domain name (issue #32)
+    assert botogram.utils.strip_urls("[email protected]") == ""
+
+
 def test_format_docstr():
     # This docstring needs lots of cleanup...
     res = botogram.utils.format_docstr(STRANGE_DOCSTRING)