Handle hyperlinked wp:anchor and wp:inline elements

Author: mwilliamson

NEWS

@@ -1,3 +1,7 @@
+# 1.12.0
+
+* Handle hyperlinked wp:anchor and wp:inline elements.
+
 # 1.11.0
 
 * Ignore style definitions using a style ID that has already been used.

mammoth/docx/body_xml.py

@@ -516,29 +516,46 @@ def break_(element):
 
 
     def inline(element):
-        properties = element.find_child_or_null("wp:docPr").attributes
+        properties_element = element.find_child_or_null("wp:docPr")
+
+        properties = properties_element.attributes
         if properties.get("descr", "").strip():
             alt_text = properties.get("descr")
         else:
             alt_text = properties.get("title")
+
+        hlink_click_element = properties_element.find_child_or_null("a:hlinkClick")
+        hyperlink_relationship_id = hlink_click_element.attributes.get("r:id")
+        if hyperlink_relationship_id:
+            href = relationships.find_target_by_relationship_id(hyperlink_relationship_id)
+        else:
+            href = None
+
         blips = element.find_children("a:graphic") \
             .find_children("a:graphicData") \
             .find_children("pic:pic") \
             .find_children("pic:blipFill") \
             .find_children("a:blip")
-        return _read_blips(blips, alt_text)
+        return _read_blips(blips, alt_text=alt_text, href=href)
 
-    def _read_blips(blips, alt_text):
-        return _ReadResult.concat(lists.map(lambda blip: _read_blip(blip, alt_text), blips))
+    def _read_blips(blips, alt_text, href):
+        return _ReadResult.concat(lists.map(lambda blip: _read_blip(blip, alt_text=alt_text, href=href), blips))
 
-    def _read_blip(element, alt_text):
+    def _read_blip(element, alt_text, href):
         blip_image = _find_blip_image(element)
 
         if blip_image is None:
             warning = results.warning("Could not find image file for a:blip element")
             return _empty_result_with_message(warning)
+
+        result = _read_image(blip_image, alt_text)
+        if href is None:
+            return result
         else:
-            return _read_image(blip_image, alt_text)
+            return result.map(lambda image_elements: documents.hyperlink(
+                image_elements,
+                href=href,
+            ))
 
     def _read_image(image_file, alt_text):
         image_path, open_image = image_file

tests/docx/body_xml_tests.py

@@ -22,6 +22,7 @@
     is_text,
     is_table,
     is_row,
+    is_image,
 )
 from ..testing import assert_equal
 
@@ -1356,10 +1357,10 @@ class ImageTests(object):
     IMAGE_BYTES = b"Not an image at all!"
     IMAGE_RELATIONSHIP_ID = "rId5"
 
-    def _read_embedded_image(self, element):
+    def _read_embedded_image(self, element, relationships=None):
         relationships = Relationships([
             _image_relationship(self.IMAGE_RELATIONSHIP_ID, "media/hat.png"),
-        ])
+        ] + (relationships or []))
 
         mocks = funk.Mocks()
         docx_file = mocks.mock()
@@ -1555,6 +1556,31 @@ def test_no_elements_created_if_image_cannot_be_found_in_wp_inline(self):
         assert_equal([], result.messages)
         assert_equal(None, result.value)
 
+    def test_can_read_pictures_with_hyperlink_specified_in_document_properties(self):
+        drawing_element = _create_inline_image(
+            blip=_embedded_blip(self.IMAGE_RELATIONSHIP_ID),
+            doc_pr_children=[
+                xml_element("a:hlinkClick", {"r:id": "rId42"})
+            ],
+        )
+
+        result = self._read_embedded_image(
+            drawing_element,
+            relationships=[_hyperlink_relationship("rId42", "http://example.com")],
+        )
+
+        assert_that(result, is_hyperlink(
+            href="http://example.com",
+            children=is_sequence(
+                is_image(
+                    content_type="image/png",
+                )
+            )
+        ))
+        image = result.children[0]
+        with image.open() as image_file:
+            assert_equal(self.IMAGE_BYTES, image_file.read())
+
 
 def test_footnote_reference_has_id_read():
     footnote_xml = xml_element("w:footnoteReference", {"w:id": "4"})
@@ -1710,9 +1736,18 @@ def _text_element(value):
     return xml_element("w:t", {}, [xml_text(value)])
 
 
-def _create_inline_image(blip, description=None, title=None):
+def _create_inline_image(blip, description=None, doc_pr_children=None, title=None):
     return xml_element("w:drawing", {}, [
-        xml_element("wp:inline", {}, _create_image_elements(blip, description=description, title=title))
+        xml_element(
+            "wp:inline",
+            {},
+            _create_image_elements(
+                blip,
+                description=description,
+                doc_pr_children=doc_pr_children,
+                title=title,
+            )
+        ),
     ])
 
 
@@ -1722,15 +1757,15 @@ def _create_anchored_image(description, blip):
     ])
 
 
-def _create_image_elements(blip, description=None, title=None):
+def _create_image_elements(blip, description=None, doc_pr_children=None, title=None):
     properties = {}
     if description is not None:
         properties["descr"] = description
     if title is not None:
         properties["title"] = title
 
     return [
-        xml_element("wp:docPr", properties),
+        xml_element("wp:docPr", properties, doc_pr_children or []),
         xml_element("a:graphic", {}, [
             xml_element("a:graphicData", {}, [
                 xml_element("pic:pic", {}, [

tests/docx/document_matchers.py

@@ -19,6 +19,7 @@ def matcher(**kwargs):
 is_checkbox = create_element_matcher(documents.Checkbox)
 is_table = create_element_matcher(documents.Table)
 is_row = create_element_matcher(documents.TableRow)
+is_image = create_element_matcher(documents.Image)
 
 
 is_empty_run = is_run(children=[])