Merge pull request #386 from malthe/issue-385-auto-boolean-attributes

Automatically configure boolean attributes for non-XML documents
malthe · Sep 22, 2023 · 8d411b2 · 8d411b2
2 parents a1fccb7 + 3b3625f
commit 8d411b2
Show file tree

Hide file tree

Showing 8 changed files with 448 additions and 22 deletions.
diff --git a/CHANGES.rst b/CHANGES.rst
@@ -3,7 +3,15 @@ Changes
 
 In next release ...
 
--
+- An XML document provided as a string (i.e. decoded) now correctly
+  has its content encoding parsed.
+
+- Boolean attributes are now automatically configured for templates in
+  non-XML mode, presuming that we're being used to generate HTML.
+
+  This means that the same loading mechanism can be used for both XML-
+  and HTML-based templates.
+
 
 4.1.0 (2023-08-29)
 ------------------

diff --git a/src/chameleon/template.py b/src/chameleon/template.py
@@ -19,10 +19,12 @@
 from .utils import DebuggingOutputStream
 from .utils import Scope
 from .utils import create_formatted_exception
+from .utils import detect_encoding
 from .utils import join
 from .utils import mangle
 from .utils import raise_with_traceback
 from .utils import read_bytes
+from .utils import read_xml_encoding
 from .utils import value_repr
 
 
@@ -87,6 +89,7 @@ class BaseTemplate:
     """
 
     default_encoding = "utf-8"
+    default_content_type = None
 
     # This attribute is strictly informational in this template class
     # and is used in exception formatting. It may be set on
@@ -224,11 +227,15 @@ def write(self, body):
             body, encoding, content_type = read_bytes(
                 body, self.default_encoding
             )
+        elif body.startswith('<?xml'):
+            content_type = 'text/xml'
+            encoding = read_xml_encoding(body.encode("utf-8"))
         else:
-            content_type = body.startswith('<?xml')
-            encoding = None
+            content_type, encoding = detect_encoding(
+                body, self.default_encoding
+            )
 
-        self.content_type = content_type
+        self.content_type = content_type or self.default_content_type
         self.content_encoding = encoding
 
         self.cook(body)
@@ -344,12 +351,7 @@ def read(self):
             data, self.default_encoding
         )
 
-        # In non-XML mode, we support various platform-specific line
-        # endings and convert them to the UNIX newline character
-        if content_type != "text/xml" and '\r' in body:
-            body = body.replace('\r\n', '\n').replace('\r', '\n')
-
-        self.content_type = content_type
+        self.content_type = content_type or self.default_content_type
         self.content_encoding = encoding
 
         return body

diff --git a/src/chameleon/tests/outputs/028.pt b/src/chameleon/tests/outputs/028.pt
@@ -1,5 +1,5 @@
 <div xmlns="http://www.w3.org/1999/xhtml">
-  <option selected="True"></option>
-  <option selected="False"></option>
+  <option selected="selected"></option>
+  <option></option>
   <option></option>
 </div>
diff --git a/src/chameleon/tests/outputs/071.pt b/src/chameleon/tests/outputs/071.pt
@@ -1,7 +1,7 @@
 <html>
   <body>
-    <input type="input" checked="True" />
-    <input type="input" checked="False" />
+    <input type="input" checked="checked" />
+    <input type="input" />
     <input type="input" />
     <input type="input" checked="checked" />
     <input type="input" checked />