Read complex field checkboxes, always unchecked

5 files changed, 98 insertions, 16 deletions

diff --git a/mammoth/docx/body_xml.py b/mammoth/docx/body_xml.py
index e1e70a5..45dc59f 100644
--- a/mammoth/docx/body_xml.py
+++ b/mammoth/docx/body_xml.py
@@ -7,7 +7,7 @@ from .. import results
 from .. import lists
 from . import complex_fields
 from .dingbats import dingbats
-from .xmlparser import node_types, XmlElement
+from .xmlparser import node_types, XmlElement, null_xml_element
 from .styles_xml import Styles
 from .uris import replace_fragment, uri_to_zip_entry_name
 
@@ -192,29 +192,45 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
     def read_fld_char(element):
         fld_char_type = element.attributes.get("w:fldCharType")
         if fld_char_type == "begin":
-            complex_field_stack.append(complex_fields.unknown)
+            complex_field_stack.append(complex_fields.begin(fld_char=element))
             del current_instr_text[:]
+
         elif fld_char_type == "end":
-            complex_field_stack.pop()
+            complex_field = complex_field_stack.pop()
+            if isinstance(complex_field, complex_fields.Begin):
+                complex_field = parse_current_instr_text(complex_field)
+
+            if isinstance(complex_field, complex_fields.Checkbox):
+                return _success(documents.checkbox(checked=complex_field.checked))
+
         elif fld_char_type == "separate":
-            instr_text = "".join(current_instr_text)
-            hyperlink_kwargs = parse_hyperlink_field_code(instr_text)
-            if hyperlink_kwargs is None:
-                complex_field = complex_fields.unknown
-            else:
-                complex_field = complex_fields.hyperlink(hyperlink_kwargs)
-            complex_field_stack.pop()
+            complex_field_separate = complex_field_stack.pop()
+            complex_field = parse_current_instr_text(complex_field_separate)
             complex_field_stack.append(complex_field)
         return _empty_result
 
-    def parse_hyperlink_field_code(instr_text):
+    def parse_current_instr_text(complex_field):
+        instr_text = "".join(current_instr_text)
+
+        if isinstance(complex_field, complex_fields.Begin):
+            fld_char = complex_field.fld_char
+        else:
+            fld_char = null_xml_element
+
+        return parse_instr_text(instr_text, fld_char=fld_char)
+
+    def parse_instr_text(instr_text, *, fld_char):
         external_link_result = re.match(r'\s*HYPERLINK "(.*)"', instr_text)
         if external_link_result is not None:
-            return dict(href=external_link_result.group(1))
+            return complex_fields.hyperlink(dict(href=external_link_result.group(1)))
 
         internal_link_result = re.match(r'\s*HYPERLINK\s+\\l\s+"(.*)"', instr_text)
         if internal_link_result is not None:
-            return dict(anchor=internal_link_result.group(1))
+            return complex_fields.hyperlink(dict(anchor=internal_link_result.group(1)))
+
+        checkbox_result = re.match(r'\s*FORMCHECKBOX\s*', instr_text)
+        if checkbox_result is not None:
+            return complex_fields.checkbox(checked=False)
 
         return None
 
diff --git a/mammoth/docx/complex_fields.py b/mammoth/docx/complex_fields.py
index 75bc05a..4112cba 100644
--- a/mammoth/docx/complex_fields.py
+++ b/mammoth/docx/complex_fields.py
@@ -2,6 +2,15 @@ class unknown(object):
     pass
 
 
+class Begin:
+    def __init__(self, *, fld_char):
+        self.fld_char = fld_char
+
+
+def begin(*, fld_char):
+    return Begin(fld_char=fld_char)
+
+
 class Hyperlink(object):
     def __init__(self, kwargs):
         self.kwargs = kwargs
@@ -9,3 +18,12 @@ class Hyperlink(object):
 
 def hyperlink(kwargs):
     return Hyperlink(kwargs=kwargs)
+
+
+class Checkbox:
+    def __init__(self, *, checked):
+        self.checked = checked
+
+
+def checkbox(*, checked):
+    return Checkbox(checked=checked)
diff --git a/mammoth/docx/xmlparser.py b/mammoth/docx/xmlparser.py
index bb84afd..2e22cec 100644
--- a/mammoth/docx/xmlparser.py
+++ b/mammoth/docx/xmlparser.py
@@ -10,7 +10,7 @@ class XmlElement(object):
     children = cobble.field()
 
     def find_child_or_null(self, name):
-        return self.find_child(name) or _null_xml_element
+        return self.find_child(name) or null_xml_element
 
     def find_child(self, name):
         for child in self.children:
@@ -50,7 +50,7 @@ class NullXmlElement(object):
         return None
 
 
-_null_xml_element = NullXmlElement()
+null_xml_element = NullXmlElement()
 
 
 @cobble.data
diff --git a/tests/docx/body_xml_tests.py b/tests/docx/body_xml_tests.py
index 117d966..cb5a9b5 100644
--- a/tests/docx/body_xml_tests.py
+++ b/tests/docx/body_xml_tests.py
@@ -18,6 +18,7 @@ from .document_matchers import (
     is_empty_run,
     is_run,
     is_hyperlink,
+    is_checkbox,
     is_text,
     is_table,
     is_row,
@@ -630,6 +631,52 @@ class ComplexFieldTests(object):
         )))
 
 
+class CheckboxTests:
+    def test_complex_field_checkbox_without_separate_is_read(self):
+        element = xml_element("w:p", {}, [
+            xml_element("w:r", {}, [
+                xml_element("w:fldChar", {"w:fldCharType": "begin"})
+            ]),
+            xml_element("w:instrText", {}, [
+                xml_text(' FORMCHECKBOX ')
+            ]),
+            xml_element("w:r", {}, [
+                xml_element("w:fldChar", {"w:fldCharType": "end"})
+            ])
+        ])
+
+        paragraph = _read_and_get_document_xml_element(element);
+
+        assert_that(paragraph, is_paragraph(children=is_sequence(
+            is_empty_run,
+            is_run(children=is_sequence(is_checkbox())),
+        )))
+
+    def test_complex_field_checkbox_with_separate_is_read(self):
+        element = xml_element("w:p", {}, [
+            xml_element("w:r", {}, [
+                xml_element("w:fldChar", {"w:fldCharType": "begin"})
+            ]),
+            xml_element("w:instrText", {}, [
+                xml_text(' FORMCHECKBOX ')
+            ]),
+            xml_element("w:r", {}, [
+                xml_element("w:fldChar", {"w:fldCharType": "separate"})
+            ]),
+            xml_element("w:r", {}, [
+                xml_element("w:fldChar", {"w:fldCharType": "end"})
+            ])
+        ])
+
+        paragraph = _read_and_get_document_xml_element(element);
+
+        assert_that(paragraph, is_paragraph(children=is_sequence(
+            is_empty_run,
+            is_empty_run,
+            is_run(children=is_sequence(is_checkbox())),
+        )))
+
+
 def test_can_read_tab_element():
     element = xml_element("w:tab")
     tab = _read_and_get_document_xml_element(element)
diff --git a/tests/docx/document_matchers.py b/tests/docx/document_matchers.py
index d2e5d5c..6b02cf9 100644
--- a/tests/docx/document_matchers.py
+++ b/tests/docx/document_matchers.py
@@ -9,13 +9,14 @@ def create_element_matcher(element_type):
             instance_of(element_type),
             has_attrs(**kwargs),
         )
-    
+
     return matcher
 
 
 is_paragraph = create_element_matcher(documents.Paragraph)
 is_run = create_element_matcher(documents.Run)
 is_hyperlink = create_element_matcher(documents.Hyperlink)
+is_checkbox = create_element_matcher(documents.Checkbox)
 is_table = create_element_matcher(documents.Table)
 is_row = create_element_matcher(documents.TableRow)