From 7ba9cc0ecdca695fec3cb90c98dd383e0edb2ba9 Mon Sep 17 00:00:00 2001
From: Michael Williamson <mike@zwobble.org>
Date: Wed, 11 Feb 2015 20:22:54 +0000
Subject: Read anchor from hyperlinks

---
 mammoth/documents.py             |  8 ++++++--
 mammoth/docx/document_xml.py     | 11 +++++++----
 tests/docx/document_xml_tests.py | 13 +++++++++++--
 3 files changed, 24 insertions(+), 8 deletions(-)

diff --git a/mammoth/documents.py b/mammoth/documents.py
index 1adcc0d..9028b03 100644
--- a/mammoth/documents.py
+++ b/mammoth/documents.py
@@ -8,7 +8,7 @@ Run = dodge.data_class("Run", [
     "is_underline", "vertical_alignment",
 ])
 Text = dodge.data_class("Text", ["value"])
-Hyperlink = dodge.data_class("Hyperlink", ["href", "children"])
+Hyperlink = dodge.data_class("Hyperlink", ["href", "anchor", "children"])
 Table = dodge.data_class("Table", ["children"])
 TableRow = dodge.data_class("TableRow", ["children"])
 TableCell = dodge.data_class("TableCell", ["children"])
@@ -55,7 +55,11 @@ def tab():
 
 
 image = Image
-hyperlink = Hyperlink
+
+def hyperlink(children, href=None, anchor=None):
+    return Hyperlink(href, anchor, children)
+    
+
 table = Table
 table_row = TableRow
 table_cell = TableCell
diff --git a/mammoth/docx/document_xml.py b/mammoth/docx/document_xml.py
index 2004948..56298f3 100644
--- a/mammoth/docx/document_xml.py
+++ b/mammoth/docx/document_xml.py
@@ -176,12 +176,15 @@ def _create_reader(numbering, content_types, relationships, styles, note_element
     @handler("w:hyperlink")
     def hyperlink(element):
         relationship_id = element.attributes.get("r:id")
+        anchor = element.attributes.get("w:anchor")
         children_result = _read_xml_elements(element.children)
-        if relationship_id is None:
-            return children_result
-        else:
+        if relationship_id is not None:
             href = relationships[relationship_id].target
-            return children_result.map(lambda children: documents.hyperlink(href, children))
+            return children_result.map(lambda children: documents.hyperlink(href=href, children=children))
+        elif anchor is not None:
+            return children_result.map(lambda children: documents.hyperlink(anchor=anchor, children=children))
+        else:
+            return children_result
     
     @handler("w:br")
     def br(element):
diff --git a/tests/docx/document_xml_tests.py b/tests/docx/document_xml_tests.py
index ffa5fbe..12fcef2 100644
--- a/tests/docx/document_xml_tests.py
+++ b/tests/docx/document_xml_tests.py
@@ -218,12 +218,21 @@ class ReadXmlElementTests(object):
         run_element = xml_element("w:r")
         element = xml_element("w:hyperlink", {"r:id": "r42"}, [run_element])
         assert_equal(
-            documents.hyperlink("http://example.com", [documents.run([])]),
+            documents.hyperlink(href="http://example.com", children=[documents.run([])]),
             _read_and_get_document_xml_element(element, relationships=relationships)
         )
         
     @istest
-    def hyperlink_is_ignored_if_it_does_not_have_a_relationship_id(self):
+    def hyperlink_is_read_if_it_has_an_anchor_attribute(self):
+        run_element = xml_element("w:r")
+        element = xml_element("w:hyperlink", {"w:anchor": "start"}, [run_element])
+        assert_equal(
+            documents.hyperlink(anchor="start", children=[documents.run([])]),
+            _read_and_get_document_xml_element(element)
+        )
+        
+    @istest
+    def hyperlink_is_ignored_if_it_does_not_have_a_relationship_id_nor_anchor(self):
         run_element = xml_element("w:r")
         element = xml_element("w:hyperlink", {}, [run_element])
         assert_equal(
-- 
cgit v1.2.3