Merge pull request #118 from laws-africa/unicode

support unicode chars

bluebell/__init__.py

@@ -1 +1 @@
-__version__ = "2.2.0"
+__version__ = "2.3.0"

bluebell/xml.py

@@ -245,7 +245,7 @@ def xml_from_dict(self, tree, is_root=False):
     def xml_from_tree(self, tree):
         """ Transform an entire parse tree to XML.
         """
-        return etree.fromstring(etree.tostring(self.item_to_xml(tree), encoding='unicode'))
+        return etree.fromstring(etree.tostring(self.item_to_xml(tree), encoding='utf-8'))
 
     def item_to_xml(self, item):
         return getattr(self, f'item_to_xml_{item["type"]}')(item)
@@ -510,7 +510,7 @@ def add_meta(self, xml):
         if not self.frbr_uri:
             raise ValueError("An frbr_uri must be provided when generating top-level documents.")
 
-        meta = etree.fromstring(etree.tostring(self.make_meta(self.frbr_uri, True)))
+        meta = etree.fromstring(etree.tostring(self.make_meta(self.frbr_uri, True), encoding='utf-8'))
         list(xml)[0].insert(0, meta)
         return xml
 

pyproject.toml

@@ -22,7 +22,7 @@ classifiers = [
     "Programming Language :: Python :: 3",
 ]
 dependencies = [
-    "cobalt>=7.0.0",
+    "cobalt>=8.0.0",
     "lxml>=3.4.1",
 ]
 dynamic = ["version"]

tests/roundtrip/act.txt

@@ -10,6 +10,8 @@ CHAP 1 - Heading
 
   SEC.custom-class{status edited} 1 - Section
 
+    emoji 😀
+
     some **text** at //the start// with **multiple //types// of** markup
 
     \***\*Notice**s\*\*

tests/test_round_trip.py

@@ -31,7 +31,7 @@ def roundtrip(self, prefix, root):
         self.assertMultiLineEqual(input, actual)
 
         # ensure it validates
-        doc = StructuredDocument.for_document_type(root)(etree.tostring(xml, encoding='unicode'))
+        doc = StructuredDocument.for_document_type(root)(etree.tostring(xml, encoding='utf-8'))
         assert_validates(doc, strict=False)
 
     def roundtrip_xml(self, file_in, root='statement'):