Support footnotes

4 files changed, 29 insertions, 1 deletions

diff --git a/mammoth/__init__.py b/mammoth/__init__.py
index e70e8ca..b0ffa12 100644
--- a/mammoth/__init__.py
+++ b/mammoth/__init__.py
@@ -3,12 +3,13 @@ from . import docx, conversion, options, documents, images, underline
 __all__ = ["convert_to_html", "extract_raw_text", "images", "underline"]
 
 
-def convert_to_html(fileobj, transform_document=None, **kwargs):
+def convert_to_html(fileobj, transform_document=None, generate_uniquifier=None, **kwargs):
     if transform_document is None:
         transform_document = lambda x: x
     return docx.read(fileobj).map(transform_document).bind(lambda document: 
         conversion.convert_document_element_to_html(
             document,
+            generate_uniquifier=generate_uniquifier,
             **options.read_options(kwargs)
         )
     )
diff --git a/mammoth/docx/__init__.py b/mammoth/docx/__init__.py
index df3071d..2c3e780 100644
--- a/mammoth/docx/__init__.py
+++ b/mammoth/docx/__init__.py
@@ -7,6 +7,7 @@ from .content_types_xml import read_content_types_xml_element
 from .relationships_xml import read_relationships_xml_element
 from .numbering_xml import read_numbering_xml_element, Numbering
 from .styles_xml import read_styles_xml_element
+from .footnotes_xml import read_footnotes_xml_element
 
 
 _namespaces = [
@@ -37,6 +38,12 @@ def read(fileobj):
     with _open_entry(zip_file, "word/styles.xml") as styles_fileobj:
         styles = read_styles_xml_element(_parse_docx_xml(styles_fileobj))
     
+    if _has_entry(zip_file, "word/footnotes.xml"):
+        with _open_entry(zip_file, "word/footnotes.xml") as footnotes_fileobj:
+            footnote_elements = read_footnotes_xml_element(_parse_docx_xml(footnotes_fileobj))
+    else:
+        footnote_elements = []
+    
     with _open_entry(zip_file, "word/document.xml") as document_fileobj:
         document_xml = _parse_docx_xml(document_fileobj)
         return read_document_xml_element(
@@ -46,6 +53,7 @@ def read(fileobj):
             docx_file=zip_file,
             numbering=numbering,
             styles=styles,
+            footnote_elements=footnote_elements,
         )
 
 
diff --git a/tests/mammoth_tests.py b/tests/mammoth_tests.py
index 7ef8049..f744711 100644
--- a/tests/mammoth_tests.py
+++ b/tests/mammoth_tests.py
@@ -1,3 +1,5 @@
+# coding=utf-8
+
 from nose.tools import istest, assert_equal
 
 from .testing import test_path
@@ -52,6 +54,23 @@ def word_tables_are_converted_to_html_tables():
         assert_equal([], result.messages)
         assert_equal(expected_html, result.value)
 
+
+@istest
+def footnotes_are_appended_to_text():
+    # TODO: don't duplicate footnotes with multiple references
+    expected_html = (u'<p>Ouch' +
+        u'<sup><a href="#footnote-42-1" id="footnote-ref-42-1">[1]</a></sup>.' +
+        u'<sup><a href="#footnote-42-2" id="footnote-ref-42-2">[2]</a></sup></p>' +
+        u'<ol><li id="footnote-42-1"><p> A tachyon walks into a bar. <a href="#footnote-ref-42-1">↑</a></p></li>' +
+        u'<li id="footnote-42-2"><p> Fin. <a href="#footnote-ref-42-2">↑</a></p></li></ol>')
+    
+    with open(test_path("footnotes.docx"), "rb") as fileobj:
+        result = mammoth.convert_to_html(fileobj=fileobj, generate_uniquifier=lambda: 42)
+        # TODO: get rid of warnings
+        #~ assert_equal([], result.messages)
+        assert_equal(expected_html, result.value)
+
+
 @istest
 def transform_document_is_applied_to_document_before_conversion():
     def transform_document(document):
diff --git a/tests/test-data/footnotes.docx b/tests/test-data/footnotes.docx
new file mode 100755
index 0000000..3228e74
--- /dev/null
+++ b/tests/test-data/footnotes.docx