Add anchor on hyperlinks as fragment if present

4 files changed, 29 insertions, 2 deletions

diff --git a/NEWS b/NEWS
index da03d08..b5dd58d 100644
--- a/NEWS
+++ b/NEWS
@@ -2,6 +2,8 @@
 
 * Extract the default data URI image converter to the images module.
 
+* Add anchor on hyperlinks as fragment if present.
+
 # 1.3.5
 
 * Handle complex fields that do not have a "separate" fldChar.
diff --git a/mammoth/compat.py b/mammoth/compat.py
new file mode 100644
index 0000000..5fb9acf
--- /dev/null
+++ b/mammoth/compat.py
@@ -0,0 +1,9 @@
+import sys
+
+if sys.version_info < (3, ):
+    from urlparse import urldefrag
+else:
+    from urllib.parse import urldefrag
+    
+    
+__all__ = ["urldefrag"]
diff --git a/mammoth/docx/body_xml.py b/mammoth/docx/body_xml.py
index b8d94cf..0724ac8 100644
--- a/mammoth/docx/body_xml.py
+++ b/mammoth/docx/body_xml.py
@@ -4,6 +4,7 @@ import re
 from .. import documents
 from .. import results
 from .. import lists
+from ..compat import urldefrag
 from . import complex_fields
 from .xmlparser import node_types, XmlElement
 from .styles_xml import Styles
@@ -294,6 +295,9 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
         children_result = _read_xml_elements(element.children)
         if relationship_id is not None:
             href = relationships[relationship_id].target
+            if anchor is not None:
+                href = urldefrag(href)[0] + "#" + anchor
+            
             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))
diff --git a/tests/docx/body_xml_tests.py b/tests/docx/body_xml_tests.py
index d112349..1277238 100644
--- a/tests/docx/body_xml_tests.py
+++ b/tests/docx/body_xml_tests.py
@@ -600,7 +600,7 @@ def _assert_children_are_converted_normally(tag_name):
 @istest
 class HyperlinkTests(object):
     @istest
-    def hyperlink_is_read_if_it_has_a_relationship_id(self):
+    def hyperlink_is_read_as_external_hyperlink_if_it_has_a_relationship_id(self):
         relationships = Relationships({
             "r42": Relationship(target="http://example.com")
         })
@@ -612,7 +612,19 @@ class HyperlinkTests(object):
         )
         
     @istest
-    def hyperlink_is_read_if_it_has_an_anchor_attribute(self):
+    def hyperlink_is_read_as_external_hyperlink_if_it_has_a_relationship_id_and_an_anchor(self):
+        relationships = Relationships({
+            "r42": Relationship(target="http://example.com/")
+        })
+        run_element = xml_element("w:r")
+        element = xml_element("w:hyperlink", {"r:id": "r42", "w:anchor": "fragment"}, [run_element])
+        assert_equal(
+            documents.hyperlink(href="http://example.com/#fragment", children=[documents.run([])]),
+            _read_and_get_document_xml_element(element, relationships=relationships)
+        )
+        
+    @istest
+    def hyperlink_is_read_as_internal_hyperlink_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(