From 2fb00589439a4efb3906d4e681e7ed815dcd180a Mon Sep 17 00:00:00 2001
From: "Sean B. Palmer" <http://inamidst.com/sbp/>
Date: Sat, 23 Feb 2008 12:16:43 +0000
Subject: [PATCH] Lots of fixes, changes, and new goodies.

---
 bot.py               | 31 ++++++++++++++++++++-----------
 irc.py               | 24 +++++++++++++++++++-----
 modules/clock.py     |  4 ++--
 modules/etymology.py |  2 +-
 modules/head.py      |  4 ++--
 modules/reload.py    | 10 ++++++----
 modules/search.py    | 17 ++++++++++++-----
 modules/tell.py      |  4 ++++
 modules/translate.py | 14 +++++++++-----
 modules/weather.py   |  2 +-
 modules/wikipedia.py |  2 ++
 opt/freenode.py      | 16 ++++++++++++++--
 phenny               |  4 ++--
 13 files changed, 94 insertions(+), 40 deletions(-)

diff --git a/bot.py b/bot.py
index d84a635..4c27b7d 100755
--- a/bot.py
+++ b/bot.py
@@ -7,7 +7,7 @@ Licensed under the Eiffel Forum License 2.
 http://inamidst.com/phenny/
 """
 
-import sys, os, re, time, threading, optparse
+import sys, os, re, threading, imp
 import irc
 
 home = os.getcwd()
@@ -31,20 +31,29 @@ class Phenny(irc.Bot):
    def setup(self): 
       self.variables = {}
 
+      filenames = []
       if not hasattr(self.config, 'enable'): 
-         load = [('modules', filename[:-3]) 
-                 for filename in os.listdir(os.path.join(home, 'modules'))
-                 if filename.endswith('.py') and
-                 not filename.startswith('_') and 
-                 not filename[:-3] in self.config.disable]
-      else: load = [('modules', e) for e in self.config.enable]
+         for fn in os.listdir(os.path.join(home, 'modules')): 
+            if fn.endswith('.py') and not fn.startswith('_'): 
+               filenames.append(os.path.join(home, 'modules', fn))
+      else: 
+         for fn in self.config.enable: 
+            filenames.append(os.path.join(home, 'modules', fn + '.py'))
+      # @@ exclude
 
-      if hasattr(self.config, 'opt'): 
-         load += [('opt', o) for o in self.config.opt]
+      if hasattr(self.config, 'extra'): 
+         for fn in self.config.extra: 
+            if os.path.isfile(fn): 
+               filenames.append(fn)
+            elif os.path.isdir(fn): 
+               for n in os.listdir(fn): 
+                  if n.endswith('.py') and not n.startswith('_'): 
+                     filenames.append(os.path.join(fn, n))
 
       modules = []
-      for package, name in load: 
-         try: module = getattr(__import__(package + '.' + name), name)
+      for filename in filenames: 
+         name = os.path.basename(filename)[:-3]
+         try: module = imp.load_source(name, filename)
          except Exception, e: 
             print >> sys.stderr, "Error loading %s: %s" % (name, e)
          else: 
diff --git a/irc.py b/irc.py
index 5916396..89114af 100755
--- a/irc.py
+++ b/irc.py
@@ -41,11 +41,21 @@ class Bot(asynchat.async_chat):
       import threading
       self.sending = threading.RLock()
 
-   def write(self, args, text=None): 
+   def __write(self, args, text=None): 
+      # print '%r %r %r' % (self, args, text)
       if text is not None: 
          self.push(' '.join(args) + ' :' + text + '\r\n')
       else: self.push(' '.join(args) + '\r\n')
 
+   def write(self, args, text=None): 
+      # This is a safe version of __write
+      try: 
+         args = [arg.encode('utf-8') for arg in args]
+         if text is not None: 
+            text = text.encode('utf-8')
+         self.__write(args, text)
+      except Exception, e: pass
+
    def run(self, host, port=6667): 
       self.initiate_connect(host, port)
 
@@ -103,6 +113,10 @@ class Bot(asynchat.async_chat):
          try: text = text.encode('utf-8')
          except UnicodeEncodeError, e: 
             text = e.__class__ + ': ' + str(e)
+      if isinstance(recipient, unicode): 
+         try: recipient = recipient.encode('utf-8')
+         except UnicodeEncodeError, e: 
+            return
 
       # No messages within the last 3 seconds? Go ahead!
       # Otherwise, wait so it's been at least 0.8 seconds + penalty
@@ -115,14 +129,14 @@ class Bot(asynchat.async_chat):
                time.sleep(wait - elapsed)
 
       # Loop detection
-      messages = [m[1] for m in self.stack[-5:]]
-      if messages.count(text) >= 3: 
+      messages = [m[1] for m in self.stack[-8:]]
+      if messages.count(text) >= 5: 
          text = '...'
-         if messages.count('...') >= 1: 
+         if messages.count('...') >= 3: 
             self.sending.release()
             return
 
-      self.write(('PRIVMSG', recipient), text)
+      self.__write(('PRIVMSG', recipient), text)
       self.stack.append((time.time(), text))
       self.stack = self.stack[-10:]
 
diff --git a/modules/clock.py b/modules/clock.py
index 210f8fb..9e38d7b 100755
--- a/modules/clock.py
+++ b/modules/clock.py
@@ -187,7 +187,7 @@ TimeZones.update(TZ1)
 @deprecated
 def f_time(self, origin, match, args): 
    """.t [ <timezone> ] - Returns the current time"""
-   tz = match.group(1) or 'GMT'
+   tz = match.group(2) or 'GMT'
 
    # Personal time zones, because they're rad
    if hasattr(self.config, 'timezones'): 
@@ -196,7 +196,7 @@ def f_time(self, origin, match, args):
 
    if People.has_key(tz): 
       tz = People[tz]
-   elif (not match.group(1)) and People.has_key(origin.nick): 
+   elif (not match.group(2)) and People.has_key(origin.nick): 
       tz = People[origin.nick]
 
    TZ = tz.upper()
diff --git a/modules/etymology.py b/modules/etymology.py
index ecdbb7b..52ae1bc 100755
--- a/modules/etymology.py
+++ b/modules/etymology.py
@@ -76,7 +76,7 @@ def etymology(word):
 def f_etymology(self, origin, match, args): 
    word = match.group(2)
 
-   try: result = etymology(word)
+   try: result = etymology(word.encode('utf-8'))
    except IOError: 
       msg = "Can't connect to etymonline.com (%s)" % (etyuri % word)
       self.msg(origin.sender, msg)
diff --git a/modules/head.py b/modules/head.py
index 4b75cb4..193286a 100755
--- a/modules/head.py
+++ b/modules/head.py
@@ -7,7 +7,7 @@ Licensed under the Eiffel Forum License 2.
 http://inamidst.com/phenny/
 """
 
-import re, urllib
+import re, urllib, urlparse
 from htmlentitydefs import name2codepoint
 import web
 from tools import deprecated
@@ -66,7 +66,7 @@ def f_title(self, origin, match, args):
             status = str(info[1])
             info = info[0]
          if status.startswith('3'): 
-            uri = info['Location']
+            uri = urlparse.urljoin(uri, info['Location'])
          else: break
 
          redirects += 1
diff --git a/modules/reload.py b/modules/reload.py
index 257eaf7..7a4c76f 100755
--- a/modules/reload.py
+++ b/modules/reload.py
@@ -13,8 +13,10 @@ def f_reload(phenny, input):
    """Reloads a module, for use by admins only.""" 
    if not input.admin: return
 
-   name = match.group(2)
-   module = getattr(__import__('modules.' + name), name)
+   name = input.group(2)
+   try: module = getattr(__import__('modules.' + name), name)
+   except ImportError: 
+      module = getattr(__import__('opt.' + name), name)
    reload(module)
 
    if hasattr(module, '__file__'): 
@@ -23,8 +25,8 @@ def f_reload(phenny, input):
       modified = time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime(mtime))
    else: modified = 'unknown'
 
-   self.register(vars(module))
-   self.bind_commands()
+   phenny.register(vars(module))
+   phenny.bind_commands()
 
    phenny.reply('%r (version: %s)' % (module, modified))
 f_reload.name = 'reload'
diff --git a/modules/search.py b/modules/search.py
index 9ad1a04..06241d4 100755
--- a/modules/search.py
+++ b/modules/search.py
@@ -30,10 +30,14 @@ def search(query, n=1):
 
 def result(query): 
    results = search(query)
-   return results['results'][0]['url']
+   if results['results']: 
+      return results['results'][0]['url']
+   return None
 
 def count(query): 
    results = search(query)
+   if not results['results']: 
+      return '0'
    return results['estimatedCount']
 
 def formatnumber(n): 
@@ -44,8 +48,11 @@ def formatnumber(n):
    return ''.join(parts)
 
 def g(phenny, input): 
-   uri = result(input.group(2))
-   phenny.reply(uri)
+   query = input.group(2)
+   uri = result(query)
+   if uri: 
+      phenny.reply(uri)
+   else: phenny.reply("No results found for '%s'." % query)
 g.commands = ['g']
 g.priority = 'high'
 
@@ -60,7 +67,7 @@ r_query = re.compile(
    r'\+?"[^"\\]*(?:\\.[^"\\]*)*"|\[[^]\\]*(?:\\.[^]\\]*)*\]|\S+'
 )
 
-def compare(phenny, input): 
+def gcs(phenny, input): 
    queries = r_query.findall(input.group(2))
    if len(queries) > 6: 
       return phenny.reply('Sorry, can only compare up to six things.')
@@ -76,7 +83,7 @@ def compare(phenny, input):
    results = [(term, n) for (n, term) in reversed(sorted(results))]
    reply = ', '.join('%s (%s)' % (t, formatnumber(n)) for (t, n) in results)
    phenny.say(reply)
-compare.commands = ['gco', 'comp']
+gcs.commands = ['gcs']
 
 if __name__ == '__main__': 
    print __doc__.strip()
diff --git a/modules/tell.py b/modules/tell.py
index 3b487c8..e4af264 100755
--- a/modules/tell.py
+++ b/modules/tell.py
@@ -59,6 +59,10 @@ def f_remind(phenny, input):
 
    # @@ Multiple comma-separated tellees? Cf. Terje, #swhack, 2006-04-15
    verb, tellee, msg = input.groups()
+   verb = verb.encode('utf-8')
+   tellee = tellee.encode('utf-8')
+   msg = msg.encode('utf-8')
+
    tellee_original = tellee.rstrip(',:;')
    tellee = tellee.lower()
 
diff --git a/modules/translate.py b/modules/translate.py
index ed3589f..7e14b1d 100644
--- a/modules/translate.py
+++ b/modules/translate.py
@@ -67,21 +67,24 @@ def translate(phrase, lang, target='en'):
    return None
 
 def tr(phenny, input): 
+   """Translates a phrase, with an optional language hint."""
    lang, phrase = input.groups()
-
+   phrase = phrase.encode('utf-8')
    if (len(phrase) > 350) and (not phenny.admin(input.nick)): 
       return phenny.reply('Phrase must be under 350 characters.')
 
    language = guess_language(phrase)
    if language is None: 
       return phenny.reply('Unable to guess the language, sorry.')
+   else: language = lang.encode('utf-8')
 
    if language != 'en': 
       translation = translate(phrase, language)
       if translation is not None: 
-         return phenny.reply(u'"%s" (%s)' % (translation, language))
+         translation = translation.decode('utf-8').encode('utf-8')
+         return phenny.reply('"%s" (%s)' % (translation, language))
 
-      error = "I think it's %s, but I can't translate that language."
+      error = "I think it's %s, which I can't translate."
       return phenny.reply(error % language.title())
 
    # Otherwise, it's English, so mangle it for fun
@@ -93,10 +96,11 @@ def tr(phenny, input):
       return phenny.reply(u'"%s" (en-unmangled)' % phrase)
    return phenny.reply("I think it's English already.")
    # @@ or 'Why but that be English, sire.'
-tr.doc = ('phenny: "<phrase>"? or phenny: <lang> "<phrase>"?', 
-   'Translate <phrase>, optionally forcing the <lang> interpretation.')
 tr.rule = ('$nick', ur'(?:([a-z]{2}) +)?["“](.+?)["”]\? *$')
+tr.example = '$nickname: "mon chien"? or $nickname: fr "mon chien"?'
 tr.priority = 'low'
 
+# @@ mangle
+
 if __name__ == '__main__': 
    print __doc__.strip()
diff --git a/modules/weather.py b/modules/weather.py
index 9e03bf4..a8fdf4f 100755
--- a/modules/weather.py
+++ b/modules/weather.py
@@ -26,7 +26,7 @@ def json(text):
    raise ValueError('Input must be serialised JSON.')
 
 def location(name): 
-   name = urllib.quote(name)
+   name = urllib.quote(name.encode('utf-8'))
    uri = 'http://ws.geonames.org/searchJSON?q=%s&maxRows=1' % name
    for i in xrange(10): 
       u = urllib.urlopen(uri)
diff --git a/modules/wikipedia.py b/modules/wikipedia.py
index 893ecab..0a0a415 100644
--- a/modules/wikipedia.py
+++ b/modules/wikipedia.py
@@ -123,6 +123,8 @@ def wikipedia(term, last=False):
 
 def wik(phenny, input): 
    origterm = input.groups()[1]
+   origterm = origterm.encode('utf-8')
+
    term = urllib.unquote(origterm)
    if not term: 
       return phenny.say(origin.sender, 'Maybe you meant ".wik Zen"?')
diff --git a/opt/freenode.py b/opt/freenode.py
index 50e6b5d..b87a5e2 100644
--- a/opt/freenode.py
+++ b/opt/freenode.py
@@ -12,10 +12,22 @@ def replaced(phenny, input):
    response = {
       'cp': '.cp has been replaced by .u', 
       'pc': '.pc has been replaced by .u', 
-      'unicode': '.unicode has been replaced by .u'
+      'unicode': '.unicode has been replaced by .u', 
+      'compare': '.compare has been replaced by .gcs (googlecounts)', 
+      'map': 'the .map command has been removed; ask sbp for details', 
+      'acronym': 'the .acronym command has been removed; ask sbp for details', 
+      'img': 'the .img command has been removed; ask sbp for details', 
+      'v': '.v has been replaced by .val', 
+      'validate': '.validate has been replaced by .validate', 
+      'thesaurus': ".thesaurus hasn't been ported to my new codebase yet", 
+      'rate': ".rate hasn't been ported to my new codebase yet, sorry!", 
+      'rates': ".rates hasn't been ported to my new codebase yet, sorry!"
    }[command]
    phenny.reply(response)
-replaced.commands = ['cp', 'pc', 'unicode']
+replaced.commands = [
+   'cp', 'pc', 'unicode', 'compare', 'map', 'acronym', 'img', 
+   'v', 'validate', 'thesaurus', 'rate', 'rates'
+]
 replaced.priority = 'low'
 
 if __name__ == '__main__': 
diff --git a/phenny b/phenny
index ca52b6d..f54cc6c 100755
--- a/phenny
+++ b/phenny
@@ -29,7 +29,7 @@ def create_default_config(fn):
    # These are people who will be able to use admin.py's functions...
    admins = [owner, 'someoneyoutrust']
    # But admin.py is disabled by default, as follows:
-   disable = ['admin']
+   exclude = ['admin']
 
    # If you want to enumerate a list of modules rather than disabling
    # some, use "enable = ['example']", which takes precedent over disable
@@ -37,7 +37,7 @@ def create_default_config(fn):
    # enable = []
 
    # Modules to load from the opt directory
-   opt = []
+   extra = []
 
    # EOF
    """)