Read image size in wp:extent tag

Author: aarbouin

mammoth/docx/body_xml.py

@@ -390,26 +390,32 @@ def inline(element):
             alt_text = properties.get("descr")
         else:
             alt_text = properties.get("title")
+        dimensions = element.find_child_or_null("wp:extent").attributes
+        size = documents.Size(
+            width=str(_emu_to_pixel(dimensions.get("cx"))),
+            height=str(_emu_to_pixel(dimensions.get("cy")))
+        )
+
         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, size)
 
     def _emu_to_pixel(emu):
         return round(int(emu) * EMU_TO_PIXEL)
 
-    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, size):
+        return _ReadResult.concat(lists.map(lambda blip: _read_blip(blip, alt_text, size), blips))
 
-    def _read_blip(element, alt_text):
-        return _read_image(lambda: _find_blip_image(element), alt_text)
+    def _read_blip(element, alt_text, size):
+        return _read_image(lambda: _find_blip_image(element), alt_text, size)
 
-    def _read_image(find_image, alt_text):
+    def _read_image(find_image, alt_text, size=None):
         image_path, open_image = find_image()
         content_type = content_types.find_content_type(image_path)
-        image = documents.image(alt_text=alt_text, content_type=content_type, open=open_image)
+        image = documents.image(alt_text=alt_text, content_type=content_type, size=size, open=open_image)
 
         if content_type in ["image/png", "image/gif", "image/jpeg", "image/svg+xml", "image/tiff"]:
             messages = []

tests/docx/body_xml_tests.py

@@ -933,13 +933,15 @@ def can_read_inline_pictures(self):
         drawing_element = _create_inline_image(
             blip=_embedded_blip(self.IMAGE_RELATIONSHIP_ID),
             description="It's a hat",
+            extent=(9525, 19000)
         )
 
         image = self._read_embedded_image(drawing_element)
 
         assert_equal(documents.Image, type(image))
         assert_equal("It's a hat", image.alt_text)
         assert_equal("image/png", image.content_type)
+        assert_equal(documents.Size(width="1", height="2"), image.size)
         with image.open() as image_file:
             assert_equal(self.IMAGE_BYTES, image_file.read())
 
@@ -1229,9 +1231,9 @@ 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, title=None, extent=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, title=title, extent=extent))
     ])
 
 
@@ -1241,15 +1243,19 @@ def _create_anchored_image(description, blip):
     ])
 
 
-def _create_image_elements(blip, description=None, title=None):
+def _create_image_elements(blip, description=None, title=None, extent=None):
     properties = {}
     if description is not None:
         properties["descr"] = description
     if title is not None:
         properties["title"] = title
-
+    extent = {
+        "cx": extent[0] if extent else "0",
+        "cy": extent[1] if extent else "0"
+    }
     return [
         xml_element("wp:docPr", properties),
+        xml_element("wp:extent", extent),
         xml_element("a:graphic", {}, [
             xml_element("a:graphicData", {}, [
                 xml_element("pic:pic", {}, [