feat: add DOC converter to MarkItDown

src/markitdown/_markitdown.py

@@ -17,6 +17,7 @@
 from typing import Any, Dict, List, Optional, Union
 from urllib.parse import parse_qs, quote, unquote, urlparse, urlunparse
 from warnings import warn, resetwarnings, catch_warnings
+from pathlib import Path
 
 import mammoth
 import markdownify
@@ -712,6 +713,52 @@ def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
         return result
 
 
+class DocConverter(HtmlConverter):
+    """
+    Converts DOC files to Markdown.
+    """
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not a DOC
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".doc":
+            return None
+
+        if not (soffice := shutil.which("soffice")):
+            return DocumentConverterResult(
+                title=None,
+                text_content="[ERROR] LibreOffice not found. Please install LibreOffice and try again.",
+            )
+
+        local_path = Path(local_path)
+        outdir = local_path.parent
+        result = None
+
+        libreoffice_command = [
+            soffice,
+            "--headless",
+            "--convert-to",
+            "txt:Text",
+            "--outdir",
+            outdir,
+            local_path,
+        ]
+        try:
+            result = subprocess.run(libreoffice_command, capture_output=True, text=True)
+            output_file = outdir / (local_path.stem + ".txt")
+            with open(output_file, "r", encoding="utf-8") as f:
+                output_content = f.read()
+            output_file.unlink(missing_ok=True)
+            result = self._convert(output_content)
+            return result
+
+        except Exception as e:
+            return DocumentConverterResult(
+                title=None,
+                text_content=f"[ERROR] Failed to process DOC file {local_path}: {str(e)}",
+            )
+
+
 class XlsxConverter(HtmlConverter):
     """
     Converts XLSX files to Markdown, with each sheet presented as a separate Markdown table.
@@ -1276,6 +1323,7 @@ def __init__(
         self.register_page_converter(YouTubeConverter())
         self.register_page_converter(BingSerpConverter())
         self.register_page_converter(DocxConverter())
+        self.register_page_converter(DocConverter())
         self.register_page_converter(XlsxConverter())
         self.register_page_converter(PptxConverter())
         self.register_page_converter(WavConverter())

tests/test_markitdown.py

@@ -25,6 +25,9 @@
 # Skip exiftool tests if not installed
 skip_exiftool = shutil.which("exiftool") is None
 
+# Skip soffice tests if not installed
+skip_soffice = shutil.which("soffice") is None
+
 TEST_FILES_DIR = os.path.join(os.path.dirname(__file__), "test_files")
 
 JPG_TEST_EXIFTOOL = {
@@ -74,6 +77,15 @@
     "Yet another comment in the doc. 55yiyi-asd09",
 ]
 
+DOC_TEST_STRINGS = [
+    "314b0a30-5b04-470b-b9f7-eed2c2bec74a",
+    "49e168b7-d2ae-407f-a055-2167576f39a1",
+    "d666f1f7-46cb-42bd-9a39-9a39cf2a509f",
+    "Abstract",
+    "Introduction",
+    "AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation",
+]
+
 PPTX_TEST_STRINGS = [
     "2cdda5c8-e50e-4db4-b5f0-9722a649f455",
     "04191ea8-5c73-4215-a1d3-1cfb43aaaf12",
@@ -257,6 +269,28 @@ def test_markitdown_exiftool() -> None:
         target = f"{key}: {JPG_TEST_EXIFTOOL[key]}"
         assert target in result.text_content
 
+@pytest.mark.skipif(
+    skip_soffice,
+    reason="do not run if soffice is not installed",
+)
+def test_markitdown_soffice() -> None:
+    markitdown = MarkItDown()
+
+    # Test DOC processing
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.doc"))
+    for test_string in DOC_TEST_STRINGS:
+        text_content = result.text_content.replace("\\", "")
+        assert test_string in text_content
+
+def test_markitdown_soffice_mocked(mocker) -> None:
+    mocker.patch("shutil.which", return_value=None)
+    markitdown = MarkItDown()
+
+    # Test DOC processing
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.doc"))
+    text_content = result.text_content.replace("\\", "")
+    assert text_content == "[ERROR] LibreOffice not found. Please install LibreOffice and try again."
+
 
 def test_markitdown_deprecation() -> None:
     try: