Add extract_raw_text function

2 files changed, 25 insertions, 1 deletions

diff --git a/mammoth/__init__.py b/mammoth/__init__.py
index 59e349f..05e2167 100644
--- a/mammoth/__init__.py
+++ b/mammoth/__init__.py
@@ -1,5 +1,5 @@
 from .results import Result
-from . import docx, conversion, options
+from . import docx, conversion, options, documents
 
 
 def convert_to_html(fileobj, transform_document=None, **kwargs):
@@ -11,3 +11,18 @@ def convert_to_html(fileobj, transform_document=None, **kwargs):
             **options.read_options(kwargs)
         )
     )
+
+
+def extract_raw_text(fileobj):
+    return docx.read(fileobj).map(extract_raw_text_from_element)
+
+
+def extract_raw_text_from_element(element):
+    if isinstance(element, documents.Text):
+        return element.value
+    else:
+        text = "".join(map(extract_raw_text_from_element, element.children))
+        if isinstance(element, documents.Paragraph):
+            return text + "\n\n"
+        else:
+            return text
diff --git a/tests/mammoth_tests.py b/tests/mammoth_tests.py
index 0b3bf0f..9721d15 100644
--- a/tests/mammoth_tests.py
+++ b/tests/mammoth_tests.py
@@ -54,3 +54,12 @@ def transform_document_is_applied_to_document_before_conversion():
         result = mammoth.convert_to_html(fileobj=fileobj, transform_document=transform_document)
         assert_equal("<h1>Walking on imported air</h1>", result.value)
         assert_equal([], result.messages)
+
+
+@istest
+def can_extract_raw_text():
+    with open(test_path("simple-list.docx"), "rb") as fileobj:
+        result = mammoth.extract_raw_text(fileobj=fileobj)
+        assert_equal([], result.messages)
+        assert_equal("Apple\n\nBanana\n\n", result.value)
+