Add conversion of paragraphs using styles

4 files changed, 82 insertions, 26 deletions

diff --git a/mammoth/__init__.py b/mammoth/__init__.py
index df032b2..1fca3d6 100644
--- a/mammoth/__init__.py
+++ b/mammoth/__init__.py
@@ -4,5 +4,5 @@ from . import docx, conversion
 
 def convert_to_html(fileobj):
     document = docx.read(fileobj).value
-    html = conversion.convert_document_element_to_html(document)
+    html = conversion.convert_document_element_to_html(document).value
     return Result(html, [])
diff --git a/mammoth/conversion.py b/mammoth/conversion.py
index 0d6dd6d..4a4cf99 100644
--- a/mammoth/conversion.py
+++ b/mammoth/conversion.py
@@ -1,41 +1,59 @@
-from .results import Result
-from . import documents
+from . import documents, results, html_paths
+from .html_generation import HtmlGenerator, satisfy_html_path
 
 
-def convert_document_element_to_html(element):
-    return _converters[type(element)](element)
-    
+def convert_document_element_to_html(element, styles=None):
+    if styles is None:
+        styles = []
+    html_generator = HtmlGenerator()
+    converter = DocumentConverter(styles)
+    converter.convert_element_to_html(element, html_generator)
+    html_generator.end_all()
+    return results.success(html_generator.html_string())
 
-_converters = {}
 
+class DocumentConverter(object):
+    def __init__(self, styles):
+        self._styles = styles
+        self._converters = {
+            documents.Document: self._convert_document,
+            documents.Paragraph: self._convert_paragraph,
+            documents.Run: self._convert_run,
+            documents.Text: self._convert_text,
+        }
 
-def _converter(element_type):
-    def add(func):
-        _converters[element_type] = func
-        
-    return add
 
+    def convert_element_to_html(self, element, html_generator):
+        self._converters[type(element)](element, html_generator)
 
-@_converter(documents.Document)
-def convert_document(document):
-    return _convert_elements_to_html(document.children)
 
+    def _convert_document(self, document, html_generator):
+        self._convert_elements_to_html(document.children, html_generator)
 
-@_converter(documents.Paragraph)
-def convert_paragraph(paragraph):
-    return "<p>{0}</p>".format(_convert_elements_to_html(paragraph.children))
 
+    def _convert_paragraph(self, paragraph, html_generator):
+        html_path = self._find_html_path_for_paragraph(paragraph)
+        satisfy_html_path(html_generator, html_path)
+        self._convert_elements_to_html(paragraph.children, html_generator)
 
-@_converter(documents.Run)
-def convert_run(run):
-    return _convert_elements_to_html(run.children)
 
+    def _convert_run(self, run, html_generator):
+        self._convert_elements_to_html(run.children, html_generator)
 
-@_converter(documents.Text)
-def convert_text(text):
-    return text.value
 
+    def _convert_text(self, text, html_generator):
+        html_generator.text(text.value)
 
-def _convert_elements_to_html(elements):
-    return "".join(map(convert_document_element_to_html, elements))
 
+    def _convert_elements_to_html(self, elements, html_generator):
+        for element in elements:
+            self.convert_element_to_html(element, html_generator)
+
+
+    def _find_html_path_for_paragraph(self, paragraph):
+        for style in self._styles:
+            document_matcher = style.document_matcher
+            if document_matcher.element_type == "paragraph" and document_matcher.style_name == paragraph.style_name:
+                return style.html_path
+                
+        return html_paths.path([html_paths.element("p")])
diff --git a/mammoth/document_matchers.py b/mammoth/document_matchers.py
index 5cfe92d..e7cd12c 100644
--- a/mammoth/document_matchers.py
+++ b/mammoth/document_matchers.py
@@ -6,6 +6,7 @@ def paragraph(style_name=None, numbering=None):
 
 
 ParagraphMatcher = collections.namedtuple("ParagraphMatcher", ["style_name", "numbering"])
+ParagraphMatcher.element_type = "paragraph"
 
 
 def run(style_name=None):
diff --git a/tests/conversion_tests.py b/tests/conversion_tests.py
new file mode 100644
index 0000000..ab90eb0
--- /dev/null
+++ b/tests/conversion_tests.py
@@ -0,0 +1,37 @@
+from nose.tools import istest, assert_equal
+
+from mammoth import documents, style_reader
+from mammoth.conversion import convert_document_element_to_html
+
+
+@istest
+def plain_paragraph_is_converted_to_plain_paragraph():
+    result = convert_document_element_to_html(
+        documents.paragraph(children=[_run_with_text("Hello")])
+    )
+    assert_equal('<p>Hello</p>', result.value)
+
+
+@istest
+def empty_paragraphs_are_ignored():
+    result = convert_document_element_to_html(
+        documents.paragraph(children=[_run_with_text("")])
+    )
+    assert_equal('', result.value)
+
+
+@istest
+def paragraphs_are_converted_by_satisfying_matching_paths():
+    result = convert_document_element_to_html(
+        documents.paragraph(style_name="TipsParagraph", children=[
+            _run_with_text("Tip")
+        ]),
+        styles=[
+            style_reader.read_style("p.TipsParagraph => p.tip")
+        ]
+    )
+    assert_equal('<p class="tip">Tip</p>', result.value)
+
+
+def _run_with_text(text):
+    return documents.run(children=[documents.text(text)])