3 files changed, 166 insertions, 0 deletions
diff --git a/tests/docx/__init__.py b/tests/docx/__init__.py
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/tests/docx/__init__.py
diff --git a/tests/docx/docx_tests.py b/tests/docx/docx_tests.py
new file mode 100644
index 0000000..9e9efe2
--- /dev/null
+++ b/tests/docx/docx_tests.py
@@ -0,0 +1,97 @@
+from nose.tools import istest, assert_equal
+
+from mammoth import docx, documents
+from mammoth.docx.xmlparser import element as xml_element, text as xml_text
+from ..testing import test_path
+
+@istest
+class ReadTests(object):
+    @istest
+    def can_read_document_with_single_paragraph_with_single_run_of_text(self):
+        with open(test_path("single-paragraph.docx")) as fileobj:
+            result = docx.read(fileobj=fileobj)
+            expected_document = documents.Document([
+                documents.paragraph([
+                    documents.Run([
+                        documents.Text("Walking on imported air")
+                    ])
+                ])
+            ])
+            assert_equal(expected_document, result.value)
+
+
+@istest
+class ReadXmlElementTests(object):
+    @istest
+    def text_from_text_element_is_read(self):
+        element = _text_element("Hello!")
+        assert_equal(documents.Text("Hello!"), docx.read_xml_element(element))
+    
+    @istest
+    def can_read_text_within_run(self):
+        element = _run_element_with_text("Hello!")
+        assert_equal(
+            documents.Run([documents.Text("Hello!")]),
+            docx.read_xml_element(element)
+        )
+    
+    @istest
+    def can_read_text_within_paragraph(self):
+        element = _paragraph_element_with_text("Hello!")
+        assert_equal(
+            documents.paragraph([documents.Run([documents.Text("Hello!")])]),
+            docx.read_xml_element(element)
+        )
+    
+    @istest
+    def can_read_text_within_document(self):
+        element = _document_element_with_text("Hello!")
+        assert_equal(
+            documents.Document([documents.paragraph([documents.Run([documents.Text("Hello!")])])]),
+            docx.read_xml_element(element)
+        )
+        
+    @istest
+    def paragraph_has_no_style_if_it_has_no_properties(self):
+        element = xml_element("w:p")
+        assert_equal(None, docx.read_xml_element(element).style_name)
+        
+    @istest
+    def paragraph_has_style_name_read_from_paragraph_properties_if_present(self):
+        style_xml = xml_element("w:pStyle", {"w:val": "Heading1"})
+        properties_xml = xml_element("w:pPr", {}, [style_xml])
+        paragraph_xml = xml_element("w:p", {}, [properties_xml])
+        paragraph = docx.read_xml_element(paragraph_xml)
+        assert_equal("Heading1", paragraph.style_name)
+    
+    @istest
+    def unrecognised_elements_are_ignored(self):
+        element = xml_element("w:huh", {}, [])
+        assert_equal(None, docx.read_xml_element(element))
+    
+    @istest
+    def unrecognised_children_are_ignored(self):
+        element = xml_element("w:r", {}, [_text_element("Hello!"), xml_element("w:huh", {}, [])])
+        assert_equal(
+            documents.Run([documents.Text("Hello!")]),
+            docx.read_xml_element(element)
+        )
+
+
+def _document_element_with_text(text):
+    return xml_element("w:document", {}, [
+        xml_element("w:body", {}, [_paragraph_element_with_text(text)])
+    ])
+
+
+def _paragraph_element_with_text(text):
+    return xml_element("w:p", {}, [_run_element_with_text(text)])
+
+
+def _run_element_with_text(text):
+    return xml_element("w:r", {}, [_text_element(text)])
+
+
+def _text_element(value):
+    return xml_element("w:t", {}, [xml_text(value)])
+    
diff --git a/tests/docx/xmlparser_tests.py b/tests/docx/xmlparser_tests.py
new file mode 100644
index 0000000..5e33778
--- /dev/null
+++ b/tests/docx/xmlparser_tests.py
@@ -0,0 +1,69 @@
+import io
+
+from nose.tools import istest, assert_equal
+
+from mammoth.docx.xmlparser import parse_xml, element as xml_element, text as xml_text
+
+
+@istest
+def can_parse_self_closing_element():
+    xml = _parse_xml_string(b"<body/>")
+    assert_equal(xml_element("body", {}, []), xml)
+
+
+@istest
+def can_parse_empty_element_with_separate_closing_tag():
+    xml = _parse_xml_string(b"<body></body>")
+    assert_equal(xml_element("body", {}, []), xml)
+
+
+@istest
+def can_parse_attributes_of_tag():
+    xml = _parse_xml_string(b"<body name='bob'></body>")
+    assert_equal(xml_element("body", {"name": "bob"}, []), xml)
+
+
+@istest
+def can_parse_text_element():
+    xml = _parse_xml_string(b"<body>Hello!</body>")
+    assert_equal(xml_element("body", {}, [xml_text("Hello!")]), xml)
+
+
+@istest
+def can_parse_text_element_before_new_tag():
+    xml = _parse_xml_string(b"<body>Hello!<br/></body>")
+    assert_equal(xml_element("body", {}, [xml_text("Hello!"), xml_element("br", {}, [])]), xml)
+
+
+@istest
+def can_parse_element_with_children():
+    xml = _parse_xml_string(b"<body><a/><b/></body>")
+    assert_equal([xml_element("a", {}, []), xml_element("b", {}, [])], xml.children)
+
+
+@istest
+def unmapped_namespaces_uris_are_included_in_braces_as_prefix():
+    xml = _parse_xml_string(b'<w:body xmlns:w="word"/>')
+    assert_equal("{word}body", xml.name)
+
+
+@istest
+def mapped_namespaces_uris_are_translated_using_namespace_map():
+    xml = _parse_xml_string(b'<w:body xmlns:w="word"/>', [("x", "word")])
+    assert_equal("x:body", xml.name)
+
+
+@istest
+def namespace_of_attributes_is_mapped_to_prefix():
+    xml = _parse_xml_string(b'<w:body xmlns:w="word" w:val="Hello!"/>', [("x", "word")])
+    assert_equal("Hello!", xml.attributes["x:val"])
+
+
+@istest
+def whitespace_between_xml_declaration_and_root_tag_is_ignored():
+    xml = _parse_xml_string(b'<?xml version="1.0" ?>\n<body/>')
+    assert_equal("body", xml.name)
+
+
+def _parse_xml_string(string, namespace_mapping=None):
+    return parse_xml(io.BytesIO(string), namespace_mapping)