From 253c812fb310733d724f5a765a77d1963c45bba8 Mon Sep 17 00:00:00 2001
From: Ben Kurtovic <ben.kurtovic@verizon.net>
Date: Fri, 27 Jul 2012 19:21:38 -0400
Subject: [PATCH] Implement __strip__ API for Wikicode.

---
 mwparserfromhell/nodes/__init__.py    |  3 ++-
 mwparserfromhell/nodes/heading.py     |  3 +++
 mwparserfromhell/nodes/html_entity.py |  5 +++++
 mwparserfromhell/nodes/tag.py         |  5 +++++
 mwparserfromhell/nodes/text.py        |  3 +++
 mwparserfromhell/wikicode.py          | 16 +++-------------
 6 files changed, 21 insertions(+), 14 deletions(-)

diff --git a/mwparserfromhell/nodes/__init__.py b/mwparserfromhell/nodes/__init__.py
index 2d5d1c2..c03785b 100644
--- a/mwparserfromhell/nodes/__init__.py
+++ b/mwparserfromhell/nodes/__init__.py
@@ -25,7 +25,8 @@ from mwparserfromhell.string_mixin import StringMixIn
 __all__ = ["Node"]
 
 class Node(StringMixIn):
-    pass
+    def __strip__(self, normalize=True, collapse=True):
+        return None
 
 from mwparserfromhell.nodes import extras
 from mwparserfromhell.nodes.text import Text
diff --git a/mwparserfromhell/nodes/heading.py b/mwparserfromhell/nodes/heading.py
index 71181e6..67851d5 100644
--- a/mwparserfromhell/nodes/heading.py
+++ b/mwparserfromhell/nodes/heading.py
@@ -32,6 +32,9 @@ class Heading(Node):
     def __unicode__(self):
         return ("=" * self.level) + self.title + ("=" * self.level)
 
+    def __strip__(self, normalize=True, collapse=True):
+        return self.title
+
     @property
     def title(self):
         return self._title
diff --git a/mwparserfromhell/nodes/html_entity.py b/mwparserfromhell/nodes/html_entity.py
index 8276d7e..aede31f 100644
--- a/mwparserfromhell/nodes/html_entity.py
+++ b/mwparserfromhell/nodes/html_entity.py
@@ -53,6 +53,11 @@ class HTMLEntity(Node):
             return u"&#x{0};".format(self.value)
         return u"&#{0};".format(self.value)
 
+    def __strip__(self, normalize=True, collapse=True):
+        if normalize:
+            return self.normalize()
+        return self
+
     def _unichr(self, value):
         """Implement the builtin unichr() with support for non-BMP code points.
 
diff --git a/mwparserfromhell/nodes/tag.py b/mwparserfromhell/nodes/tag.py
index f395fe4..f0a7748 100644
--- a/mwparserfromhell/nodes/tag.py
+++ b/mwparserfromhell/nodes/tag.py
@@ -100,6 +100,11 @@ class Tag(Node):
             result += "</" + unicode(self.tag) + " " * self.close_padding + ">"
         return result
 
+    def __strip__(self, normalize=True, collapse=True):
+        if self.type in self.TAGS_VISIBLE:
+            return self.contents.strip_code(normalize, collapse)
+        return None
+
     def translate(self):
         translations {
             self.TAG_ITALIC: ("''", "''"),
diff --git a/mwparserfromhell/nodes/text.py b/mwparserfromhell/nodes/text.py
index afdc235..d6b929c 100644
--- a/mwparserfromhell/nodes/text.py
+++ b/mwparserfromhell/nodes/text.py
@@ -31,6 +31,9 @@ class Text(Node):
     def __unicode__(self):
         return unicode(self.value)
 
+    def __strip__(self, normalize=True, collapse=True):
+        return self
+
     @property
     def value(self):
         return self._value
diff --git a/mwparserfromhell/wikicode.py b/mwparserfromhell/wikicode.py
index 044d635..ae05879 100644
--- a/mwparserfromhell/wikicode.py
+++ b/mwparserfromhell/wikicode.py
@@ -253,20 +253,10 @@ class Wikicode(StringMixIn):
     def strip_code(self, normalize=True, collapse=True):
         nodes = []
         for node in self.nodes:
-            if isinstance(node, Heading):
-                nodes.append(child.title)
-            elif isinstance(node, HTMLEntity):
-                if normalize:
-                    nodes.append(node.normalize())
-                else:
-                    nodes.append(node)
-            elif isinstance(node, Tag):
-                if node.type in node.TAGS_VISIBLE:
-                    nodes.append(node.contents.strip_code(normalize, collapse))
-            elif isinstance(node, Text):
-                nodes.append(node)
+            stripped = node.__strip__(normalize)
+            if stripped:
+                nodes.append(unicode(stripped))
 
-        nodes = map(unicode, nodes)
         if collapse:
             stripped = u"".join(nodes).strip("\n")
             while "\n\n\n" in stripped: