core: configure schemes to trigger URL Callbacks on

sopel/config/core_section.py

@@ -87,6 +87,12 @@ class CoreSection(StaticSection):
 
     May not apply, depending on ``auth_method``."""
 
+    auto_url_schemes = ListAttribute(
+        'auto_url_schemes',
+        strip=True,
+        default=['http', 'https', 'ftp'])
+    """List of URL schemes that will trigger URL callbacks."""
+
     bind_host = ValidatedAttribute('bind_host')
     """Bind the connection to a specific IP"""
 

sopel/coretasks.py

@@ -800,16 +800,17 @@ def track_topic(bot, trigger):
     bot.channels[channel].topic = trigger.args[-1]
 
 
-@sopel.module.rule(r'(?u).*(https?://\S+).*')
+@sopel.module.rule(r'(?u).*(.+://\S+).*')
 @sopel.module.unblockable
 def handle_url_callbacks(bot, trigger):
     """Dispatch callbacks on URLs
 
     For each URL found in the trigger, trigger the URL callback registered by
     the ``@url`` decorator.
     """
+    schemes = bot.config.core.auto_url_schemes
     # find URLs in the trigger
-    for url in sopel.web.search_urls(trigger):
+    for url in sopel.web.search_urls(trigger, schemes=schemes):
         # find callbacks for said URL
         for function, match in bot.search_url_callbacks(url):
             # trigger callback defined by the `@url` decorator

sopel/web.py

@@ -252,10 +252,13 @@ def trim_url(url):
     return url
 
 
-def search_urls(text, exclusion_char=None, clean=False):
-    re_url = r'((?:http|https|ftp)(?::\/\/\S+))'
+def search_urls(text, exclusion_char=None, clean=False, schemes=None):
+    schemes = schemes or ['http', 'https', 'ftp']
+    schemes_patterns = '|'.join(re.escape(scheme) for scheme in schemes)
+    re_url = r'((?:%s)(?::\/\/\S+))' % schemes_patterns
     if exclusion_char is not None:
-        re_url = r'((?<!%s)(?:http|https|ftp)(?::\/\/\S+))' % (exclusion_char)
+        re_url = r'((?<!%s)(?:%s)(?::\/\/\S+))' % (
+            exclusion_char, schemes_patterns)
 
     r = re.compile(re_url, re.IGNORECASE | re.UNICODE)
 

test/test_web.py

@@ -96,6 +96,30 @@ def test_search_urls_exclusion_char_only_once():
     assert 'http://b.com' in urls
 
 
+def test_search_urls_default_schemes():
+    urls = list(search_urls('http://a.com ftp://b.com https://c.com'))
+    assert len(urls) == 3, 'Must find all three URLs'
+    assert 'http://a.com' in urls
+    assert 'ftp://b.com' in urls
+    assert 'https://c.com' in urls
+
+
+@pytest.mark.parametrize('scheme', ['http', 'https', 'ftp', 'steam'])
+def test_search_urls_defined_schemes(scheme):
+    expected = {
+        'http': 'http://a.com',
+        'https': 'https://c.com',
+        'ftp': 'ftp://b.com',
+        'steam': 'steam://portal2',
+    }.get(scheme)
+
+    urls = list(
+        search_urls('http://a.com ftp://b.com https://c.com steam://portal2',
+        schemes=[scheme]))
+    assert len(urls) == 1, 'Only %s URLs must be found' % scheme
+    assert expected in urls
+
+
 TRAILING_CHARS = list('.,?!\'":;')
 ENCLOSING_PAIRS = [('(', ')'), ('[', ']'), ('{', '}'), ('<', '>')]