Add local concat mode option support

kirsle · kirsle · commit 8d5e14c0831a · 2015-02-11T17:27:38.000-08:00
Adds an option to enable RiveScript documents to override the
concatenation character on a file-scoped basis. Valid examples are:

! local concat = none
! local concat = space
! local concat = newline

The default is 'none'. This affects which character is used when a line
has to be concatenated via the ^ command.
diff --git a/Changes b/Changes
@@ -11,6 +11,8 @@ Revision history for the Python package RiveScript.
 	  especially helpful was to precompile substitution and simple trigger
 	  regexps), taking the time-to-reply for the default brain from ~0.19s down
 	  to ~0.04s
+	- Add support for `! local concat` option to override concatenation mode
+	  (file scoped)
 
 1.06  Nov 25 2014
 	- Change package name from python-rivescript to simply rivescript.
diff --git a/rivescript/__init__.py b/rivescript/__init__.py
@@ -25,7 +25,7 @@
 # SOFTWARE.
 
 # Python 3 compat
-from __future__ import print_function
+from __future__ import print_function, unicode_literals
 
 __author__     = 'Noah Petherbridge'
 __copyright__  = 'Copyright 2013, Noah Petherbridge'
diff --git a/rivescript/__main__.py b/rivescript/__main__.py
@@ -22,7 +22,7 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
 
-from __future__ import absolute_import
+from __future__ import absolute_import, unicode_literals
 
 """RiveScript's __main__.py
 
diff --git a/rivescript/interactive.py b/rivescript/interactive.py
@@ -22,7 +22,7 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
 
-from __future__ import print_function
+from __future__ import print_function, unicode_literals
 
 """interactive.py: RiveScript's built-in interactive mode.
 
diff --git a/rivescript/python.py b/rivescript/python.py
@@ -23,12 +23,12 @@
 # SOFTWARE.
 
 # Python3 compat
-from __future__ import print_function
+from __future__ import print_function, unicode_literals
 
 __docformat__ = 'plaintext'
 
 
-class PyRiveObjects:
+class PyRiveObjects(object):
     """A RiveScript object handler for Python code.
 
 This class provides built-in support for your RiveScript documents to include
diff --git a/rivescript/rivescript.py b/rivescript/rivescript.py
@@ -22,6 +22,7 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
 
+from __future__ import unicode_literals
 import sys
 import os
 import re
@@ -78,9 +79,16 @@ class RE(object):
 RS_ERR_REPLY = "ERR: No Reply Found"
 
 
-class RiveScript:
+class RiveScript(object):
     """A RiveScript interpreter for Python 2 and 3."""
 
+    # Concatenation mode characters.
+    _concat_modes = dict(
+        none="",
+        space=" ",
+        newline="\n",
+    )
+
     ############################################################################
     # Initialization and Utility Methods                                       #
     ############################################################################
@@ -220,6 +228,11 @@ def _parse(self, fname, code):
         concnt  = 0        # Condition counter
         isThat  = ''       # Is a %Previous trigger
 
+        # Local (file scoped) parser options.
+        local_options = dict(
+            concat="none", # Concat mode for ^Continue command
+        )
+
         # Read each line.
         for lp, line in enumerate(code):
             lineno = lineno + 1
@@ -329,7 +342,9 @@ def _parse(self, fname, code):
                     # end of the current line.
                     if cmd != '^' and lookCmd != '%':
                         if lookCmd == '^':
-                            line += lookahead
+                            line += self._concat_modes.get(
+                                local_options["concat"], ""
+                            ) + lookahead
                         else:
                             break
 
@@ -372,7 +387,11 @@ def _parse(self, fname, code):
                     continue
 
                 # Handle the rest of the types.
-                if type == 'global':
+                if type == 'local':
+                    # Local file-scoped parser options.
+                    self._say("\tSet parser option " + var + " = " + value)
+                    local_options[var] = value
+                elif type == 'global':
                     # 'Global' variables
                     self._say("\tSet global " + var + " = " + value)
 
diff --git a/tests/test_rivescript.py b/tests/test_rivescript.py
@@ -559,6 +559,56 @@ def test_topic_inheritence(self):
         self.reply("Say stuff.", RS_ERR_MATCH)
 
 
+class ParserOptionTest(RiveScriptTestCase):
+    """File scoped parser option tests."""
+
+    def test_concat(self):
+        self.new("""
+            // Default concat mode = none
+            + test concat default
+            - Hello
+            ^ world!
+
+            ! local concat = space
+            + test concat space
+            - Hello
+            ^ world!
+
+            ! local concat = none
+            + test concat none
+            - Hello
+            ^ world!
+
+            ! local concat = newline
+            + test concat newline
+            - Hello
+            ^ world!
+
+            // invalid concat setting is equivalent to `none`
+            ! local concat = foobar
+            + test concat foobar
+            - Hello
+            ^ world!
+
+            // the option is file scoped so it can be left at
+            // any setting and won't affect subsequent parses
+            ! local concat = newline
+        """)
+        self.extend("""
+            // concat mode should be restored to the default in a
+            // separate file/stream parse
+            + test concat second file
+            - Hello
+            ^ world!
+        """)
+
+        self.reply("test concat default", "Helloworld!")
+        self.reply("test concat space", "Hello world!")
+        self.reply("test concat none", "Helloworld!")
+        self.reply("test concat newline", "Hello\nworld!")
+        self.reply("test concat second file", "Helloworld!")
+
+
 class UnicodeTest(RiveScriptTestCase):
     """UTF-8 Tests."""
 
