Extended formatting library

plugins/google.py

@@ -1,6 +1,7 @@
 import random
-from util import hook
+from util import hook, http
-from util import http
+
+from util.formatting import truncate_words
 
 
 def api_get(kind, query):
@@ -22,7 +23,7 @@ def googleimage(inp):
                       parsed['responseStatus'], ''))
     if not parsed['responseData']['results']:
         return 'no images found'
-    return random.choice(parsed['responseData']['results'][:10])\
+    return random.choice(parsed['responseData']['results'][:10]) \
                         ['unescapedUrl']
 
 
@@ -53,7 +54,6 @@ def google(inp):
 
     out = ' '.join(out.split())
 
-    if len(out) > 300:
+    out = truncate_words(out, 300)
-        out = out[:out.rfind(' ')] + '...'
 
     return out

plugins/util/formatting.py

@@ -1,8 +1,48 @@
-""" formatting.py - handy functions for formatting text """
+""" formatting.py - handy functions for formatting text
+    this file contains code from the following URL:
+    <http://code.djangoproject.com/svn/django/trunk/django/utils/text.py>
+"""
+import re
 
 
 def capitalize_first(line):
-    """ capitalises the first letter of words
+    """
-        (keeps other letters intact)
+    capitalises the first letter of words
+    (keeps other letters intact)
     """
     return ' '.join([s[0].upper() + s[1:] for s in line.split(' ')])
+
+
+def truncate_words(s, num):
+    "Truncates a string after a certain number of words."
+    length = int(num)
+    words = s.split()
+    if len(words) > length:
+        words = words[:length]
+        if not words[-1].endswith('...'):
+            words.append('...')
+    return ' '.join(words)
+
+# Expression to match some_token and some_token="with spaces" (and similarly
+# for single-quoted strings).
+split_re = re.compile(r"""((?:[^\s'"]*(?:(?:"(?:[^"\\]|\\.)*" | '(?:[""" \
+                      r"""^'\\]|\\.)*')[^\s'"]*)+) | \S+)""", re.VERBOSE)
+
+
+def smart_split(text):
+    r"""
+    Generator that splits a string by spaces, leaving quoted phrases together.
+    Supports both single and double quotes, and supports escaping quotes with
+    backslashes. In the output, strings will keep their initial and trailing
+    quote marks and escaped quotes will remain escaped (the results can then
+    be further processed with unescape_string_literal()).
+
+    >>> list(smart_split(r'This is "a person\'s" test.'))
+    [u'This', u'is', u'"a person\\\'s"', u'test.']
+    >>> list(smart_split(r"Another 'person\'s' test."))
+    [u'Another', u"'person\\'s'", u'test.']
+    >>> list(smart_split(r'A "\"funky\" style" test.'))
+    [u'A', u'"\\"funky\\" style"', u'test.']
+    """
+    for bit in split_re.finditer(text):
+        yield bit.group(0)

plugins/wikipedia.py

@@ -4,6 +4,7 @@ Scaevolus 2009'''
 import re
 
 from util import hook, http
+from util.formatting import truncate_words
 
 
 api_prefix = "http://en.wikipedia.org/w/api.php"
@@ -43,8 +44,7 @@ def wiki(inp):
         desc = title + desc
 
     desc = re.sub('\s+', ' ', desc).strip()  # remove excess spaces
-
+    
-    if len(desc) > 300:
+    desc = truncate_words(desc, 300)
-        desc = desc[:300] + '...'
 
     return '%s -- %s' % (desc, http.quote(url, ':/'))

plugins/wolframalpha.py

@@ -5,6 +5,7 @@ from util import hook, http
 
 from urllib2 import HTTPError
 from util.web import bitly, ShortenError
+from util.formatting import truncate_words
 
 
 @hook.command('wa')
@@ -15,14 +16,14 @@ def wolframalpha(inp, bot=None):
     api_key = bot.config.get("api_keys", {}).get("wolframalpha", None)
     bitly_user = bot.config.get("api_keys", {}).get("bitly_user", None)
     bitly_key = bot.config.get("api_keys", {}).get("bitly_api", None)
-    
+
     if not api_key:
         return "error: missing api key"
 
     url = 'http://api.wolframalpha.com/v2/query?format=plaintext'
 
     result = http.get_xml(url, input=inp, appid=api_key)
-    
+
     # get the URL for a user to view this query in a browser
     query_url = "http://www.wolframalpha.com/input/?i=" + \
                 urllib.quote(inp.encode('utf-8'))
@@ -46,7 +47,7 @@ def wolframalpha(inp, bot=None):
         if results:
             pod_texts.append(title + ': ' + ','.join(results))
 
-    ret = ' | '.join(pod_texts)
+    ret = ' - '.join(pod_texts)
 
     if not pod_texts:
         return 'No results.'
@@ -58,10 +59,8 @@ def wolframalpha(inp, bot=None):
 
     ret = re.sub(r'\\:([0-9a-z]{4})', unicode_sub, ret)
 
-    if len(ret) > 410:
+    ret = truncate_words(ret, 410)
-        ret = ret[:ret.rfind(' ', 0, 410)]
+    
-        ret = re.sub(r'\W+$', '', ret) + '...'
-
     if not ret:
         return 'No results.'