Markov: track the context of said lines

a context is a meta-classification ('banter, 'secrets', whatever)
based on targets (channels or nicknames). when a line is being
learned from a known target, the chains are placed in that context.

this is for allowing one brain to have multiple personalities, in
a sense, for large networks or cases where there may be a more
sanitized set of channels and a couple channels where everyone lets
it rip. a later enhancement would have sentence creation choose from
context-less chains (and contexts matching the current target), but
i need to go back to the drawing board on that one a bit.

ramble ramble ramble

modules/Markov.py

@@ -87,6 +87,32 @@ class Markov(Module):
                 db.rollback()
                 print("sqlite error: " + str(e))
                 raise
+        if (version < 2):
+            db = self.get_db()
+            try:
+                db.execute('''
+                    ALTER TABLE markov_chain
+                        ADD COLUMN context TEXT DEFAULT NULL''')
+                db.execute('''
+                    CREATE TABLE markov_context (
+                        id INTEGER PRIMARY KEY AUTOINCREMENT,
+                        context TEXT NOT NULL
+                    )''')
+                db.execute('''
+                    CREATE TABLE markov_target_to_context_map (
+                        id INTEGER PRIMARY KEY AUTOINCREMENT,
+                        target TEXT NOT NULL,
+                        context_id INTEGER NOT NULL,
+                        FOREIGN KEY(context_id) REFERENCES markov_context(id)
+                    )''')
+                db.execute('UPDATE drbotzo_modules SET version = ? WHERE module = ?',
+                    (2, self.__class__.__name__))
+                db.commit()
+                version = 2
+            except sqlite3.Error as e:
+                db.rollback()
+                print('sqlite error: ' + str(e))
+                raise
 
     def register_handlers(self):
         """Handle pubmsg/privmsg, to learn and/or reply to IRC events."""
@@ -108,12 +134,13 @@ class Markov(Module):
         what = ''.join(event.arguments()[0])
         my_nick = connection.get_nickname()
         what = re.sub('^' + my_nick + '[:,]\s+', '', what)
+        target = event.target()
 
         # don't learn from commands
         if self.trainre.search(what) or self.learnre.search(what) or self.replyre.search(what):
             return
 
-        self._learn_line(what)
+        self._learn_line(what, target)
 
     def do(self, connection, event, nick, userhost, what, admin_unlocked):
         """Handle commands and inputs."""
@@ -154,10 +181,11 @@ class Markov(Module):
     def markov_learn(self, connection, event, nick, userhost, what, admin_unlocked):
         """Learn one line, as provided to the command."""
 
+        target = event.target()
         match = self.learnre.search(what)
         if match:
             line = match.group(1)
-            self._learn_line(line)
+            self._learn_line(line, target)
 
             # return what was learned, for weird chaining purposes
             return line
@@ -181,13 +209,18 @@ class Markov(Module):
             else:
                 return self._reply(min_size=min_size, max_size=max_size)
 
-    def _learn_line(self, line):
+    def _learn_line(self, line, target=None):
         """Create Markov chains from the provided line."""
 
         # set up the head of the chain
         k1 = self.start1
         k2 = self.start2
 
+        # see if there's a context for this
+        context = None
+        if target:
+            context = self._get_context_for_target(target)
+
         words = line.split()
         if len(words) <= 0:
             return line
@@ -197,12 +230,20 @@ class Markov(Module):
         try:
             db = self.get_db()
             cur = db.cursor()
-            statement = 'INSERT INTO markov_chain (k1, k2, v) VALUES (?, ?, ?)'
+            if context:
+                statement = 'INSERT INTO markov_chain (k1, k2, v, context) VALUES (?, ?, ?, ?)'
 
-            for word in words:
-                cur.execute(statement, (k1.decode('utf-8', 'replace').lower(), k2.decode('utf-8', 'replace').lower(), word.decode('utf-8', 'replace').lower()))
-                k1, k2 = k2, word
-            cur.execute(statement, (k1.decode('utf-8', 'replace').lower(), k2.decode('utf-8', 'replace').lower(), self.stop))
+                for word in words:
+                    cur.execute(statement, (k1.decode('utf-8', 'replace').lower(), k2.decode('utf-8', 'replace').lower(), word.decode('utf-8', 'replace').lower(), context))
+                    k1, k2 = k2, word
+                cur.execute(statement, (k1.decode('utf-8', 'replace').lower(), k2.decode('utf-8', 'replace').lower(), self.stop, context))
+            else:
+                statement = 'INSERT INTO markov_chain (k1, k2, v) VALUES (?, ?, ?)'
+
+                for word in words:
+                    cur.execute(statement, (k1.decode('utf-8', 'replace').lower(), k2.decode('utf-8', 'replace').lower(), word.decode('utf-8', 'replace').lower()))
+                    k1, k2 = k2, word
+                cur.execute(statement, (k1.decode('utf-8', 'replace').lower(), k2.decode('utf-8', 'replace').lower(), self.stop))
 
             db.commit()
         except sqlite3.Error as e:
@@ -336,5 +377,24 @@ class Markov(Module):
             print('sqlite error: ' + str(e))
             raise
 
+    def _get_context_for_target(self, target):
+        """Get the context for a channel/nick, if defined."""
+
+        try:
+            db = self.get_db()
+            query = '''
+                SELECT context_id FROM markov_target_to_context_map
+                    WHERE target = ?
+                '''
+            cursor = db.execute(query, (target,))
+            result = cursor.fetchone()
+            if result:
+                return result['context_id']
+            else:
+                return None
+        except sqlite3.Error as e:
+            print('sqlite error: ' + str(e))
+            raise
+
 # vi:tabstop=4:expandtab:autoindent
 # kate: indent-mode python;indent-width 4;replace-tabs on;