Merge branch 'main' into patch-1

.devcontainer/devcontainer.json

@@ -0,0 +1,29 @@
+// For format details, see https://aka.ms/devcontainer.json. For config options, see the
+// README at: https://github.com/devcontainers/templates/tree/main/src/docker-existing-dockerfile
+{
+	"name": "Existing Dockerfile",
+	"build": {
+		// Sets the run context to one level up instead of the .devcontainer folder.
+		"context": "..",
+		// Update the 'dockerFile' property if you aren't using the standard 'Dockerfile' filename.
+		"dockerfile": "../Dockerfile"
+	},
+
+	// Features to add to the dev container. More info: https://containers.dev/features.
+	// "features": {},
+	"features": {
+		"ghcr.io/devcontainers-extra/features/hatch:2": {}
+	},
+
+	// Use 'forwardPorts' to make a list of ports inside the container available locally.
+	// "forwardPorts": [],
+
+	// Uncomment the next line to run commands after the container is created.
+	// "postCreateCommand": "cat /etc/os-release",
+
+	// Configure tool-specific properties.
+	// "customizations": {},
+
+	// Uncomment to connect as an existing user other than the container default. More info: https://aka.ms/dev-containers-non-root.
+	"remoteUser": "root"
+}

Dockerfile

@@ -1,9 +1,11 @@
-FROM python:3.13-alpine
+FROM python:3.13-slim-bullseye
 
 USER root
 
 # Runtime dependency
-RUN apk add --no-cache ffmpeg
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    ffmpeg \
+ && rm -rf /var/lib/apt/lists/*
 
 RUN pip install markitdown
 

README.md

@@ -2,65 +2,47 @@
 
 [![PyPI](https://img.shields.io/pypi/v/markitdown.svg)](https://pypi.org/project/markitdown/)
 
-The MarkItDown library is a utility tool for converting various files to Markdown (e.g., for indexing, text analysis, etc.)
+MarkItDown is a utility for converting various files to Markdown (e.g., for indexing, text analysis, etc).
+It supports:
+- PDF
+- PowerPoint
+- Word
+- Excel
+- Images (EXIF metadata and OCR)
+- Audio (EXIF metadata and speech transcription)
+- HTML
+- Text-based formats (CSV, JSON, XML)
+- ZIP files (iterates over contents)
 
-It presently supports:
+To install MarkItDown, use pip: `pip install markitdown`. Alternatively, you can install it from the source: `pip install -e .`
 
-- PDF (.pdf)
-- PowerPoint (.pptx)
-- Word (.docx)
-- Excel (.xlsx)
-- Images (EXIF metadata, and OCR)
-- Audio (EXIF metadata, and speech transcription)
-- HTML (special handling of Wikipedia, etc.)
-- Various other text-based formats (csv, json, xml, etc.)
-- ZIP (Iterates over contents and converts each file)
+## Usage
 
-# Installation
-
-You can install `markitdown` using pip:
-
-```python
-pip install markitdown
-```
-
-or from the source
-
-```sh
-pip install -e .
-```
-
-# Usage
-The API is simple:
-
-```python
-from markitdown import MarkItDown
-
-markitdown = MarkItDown()
-result = markitdown.convert("test.xlsx")
-print(result.text_content)
-```
-
-To use this as a command-line utility, install it and then run it like this:
-
-```bash
-markitdown path-to-file.pdf
-```
-
-This will output Markdown to standard output. You can save it like this:
+### Command-Line
 
 ```bash
 markitdown path-to-file.pdf > document.md
 ```
 
-You can pipe content to standard input by omitting the argument:
+You can also pipe content:
 
 ```bash
 cat path-to-file.pdf | markitdown
 ```
 
-You can also configure markitdown to use Large Language Models to describe images. To do so you must provide `llm_client` and `llm_model` parameters to MarkItDown object, according to your specific client.
+### Python API
 
+Basic usage in Python:
+
+```python
+from markitdown import MarkItDown
+
+md = MarkItDown()
+result = md.convert("test.xlsx")
+print(result.text_content)
+```
+
+To use Large Language Models for image descriptions, provide `llm_client` and `llm_model`:
 
 ```python
 from markitdown import MarkItDown
@@ -72,7 +54,7 @@ result = md.convert("example.jpg")
 print(result.text_content)
 ```
 
-You can also use the project as Docker Image:
+### Docker
 
 ```sh
 docker build -t markitdown:latest .
@@ -141,28 +123,27 @@ This project has adopted the [Microsoft Open Source Code of Conduct](https://ope
 For more information see the [Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/) or
 contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or comments.
 
-### Running Tests
+### Running Tests and Checks
 
-To run tests, install `hatch` using `pip` or other methods as described [here](https://hatch.pypa.io/dev/install).
+- Install `hatch` in your environment and run tests:
+    ```sh
+    pip install hatch  # Other ways of installing hatch: https://hatch.pypa.io/dev/install/
+    hatch shell
+    hatch test
+    ```
 
-```sh
-pip install hatch
-hatch shell
-hatch test
-```
+  (Alternative) Use the Devcontainer which has all the dependencies installed:
+    ```sh
+    # Reopen the project in Devcontainer and run:
+    hatch test
+    ```
 
-### Running Pre-commit Checks
-
-Please run the pre-commit checks before submitting a PR.
-
-```sh
-pre-commit run --all-files
-```
+- Run pre-commit checks before submitting a PR: `pre-commit run --all-files`
 
 ## Trademarks
 
-This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft 
-trademarks or logos is subject to and must follow 
+This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft
+trademarks or logos is subject to and must follow
 [Microsoft's Trademark & Brand Guidelines](https://www.microsoft.com/en-us/legal/intellectualproperty/trademarks/usage/general).
 Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship.
 Any use of third-party trademarks or logos are subject to those third-party's policies.

pyproject.toml

@@ -39,6 +39,7 @@ dependencies = [
   "SpeechRecognition",
   "pathvalidate",
   "charset-normalizer",
+  "openai",
 ]
 
 [project.urls]
@@ -77,3 +78,6 @@ exclude_lines = [
   "if __name__ == .__main__.:",
   "if TYPE_CHECKING:",
 ]
+
+[tool.hatch.build.targets.sdist]
+only-include = ["src/markitdown"]

src/markitdown/__about__.py

@@ -1,4 +1,4 @@
 # SPDX-FileCopyrightText: 2024-present Adam Fourney <adamfo@microsoft.com>
 #
 # SPDX-License-Identifier: MIT
-__version__ = "0.0.1a1"
+__version__ = "0.0.1a3"

src/markitdown/_markitdown.py

@@ -13,9 +13,10 @@ import sys
 import tempfile
 import traceback
 import zipfile
+from xml.dom import minidom
 from typing import Any, Dict, List, Optional, Union
 from urllib.parse import parse_qs, quote, unquote, urlparse, urlunparse
-from warnings import catch_warnings
+from warnings import warn, resetwarnings, catch_warnings
 
 import mammoth
 import markdownify
@@ -44,6 +45,8 @@ try:
     IS_AUDIO_TRANSCRIPTION_CAPABLE = True
 except ModuleNotFoundError:
     pass
+finally:
+    resetwarnings()
 
 # Optional YouTube transcription support
 try:
@@ -220,6 +223,143 @@ class HtmlConverter(DocumentConverter):
         )
 
 
+class RSSConverter(DocumentConverter):
+    """Convert RSS / Atom type to markdown"""
+
+    def convert(
+        self, local_path: str, **kwargs
+    ) -> Union[None, DocumentConverterResult]:
+        # Bail if not RSS type
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() not in [".xml", ".rss", ".atom"]:
+            return None
+        try:
+            doc = minidom.parse(local_path)
+        except BaseException as _:
+            return None
+        result = None
+        if doc.getElementsByTagName("rss"):
+            # A RSS feed must have a root element of <rss>
+            result = self._parse_rss_type(doc)
+        elif doc.getElementsByTagName("feed"):
+            root = doc.getElementsByTagName("feed")[0]
+            if root.getElementsByTagName("entry"):
+                # An Atom feed must have a root element of <feed> and at least one <entry>
+                result = self._parse_atom_type(doc)
+            else:
+                return None
+        else:
+            # not rss or atom
+            return None
+
+        return result
+
+    def _parse_atom_type(
+        self, doc: minidom.Document
+    ) -> Union[None, DocumentConverterResult]:
+        """Parse the type of an Atom feed.
+
+        Returns None if the feed type is not recognized or something goes wrong.
+        """
+        try:
+            root = doc.getElementsByTagName("feed")[0]
+            title = self._get_data_by_tag_name(root, "title")
+            subtitle = self._get_data_by_tag_name(root, "subtitle")
+            entries = root.getElementsByTagName("entry")
+            md_text = f"# {title}\n"
+            if subtitle:
+                md_text += f"{subtitle}\n"
+            for entry in entries:
+                entry_title = self._get_data_by_tag_name(entry, "title")
+                entry_summary = self._get_data_by_tag_name(entry, "summary")
+                entry_updated = self._get_data_by_tag_name(entry, "updated")
+                entry_content = self._get_data_by_tag_name(entry, "content")
+
+                if entry_title:
+                    md_text += f"\n## {entry_title}\n"
+                if entry_updated:
+                    md_text += f"Updated on: {entry_updated}\n"
+                if entry_summary:
+                    md_text += self._parse_content(entry_summary)
+                if entry_content:
+                    md_text += self._parse_content(entry_content)
+
+            return DocumentConverterResult(
+                title=title,
+                text_content=md_text,
+            )
+        except BaseException as _:
+            return None
+
+    def _parse_rss_type(
+        self, doc: minidom.Document
+    ) -> Union[None, DocumentConverterResult]:
+        """Parse the type of an RSS feed.
+
+        Returns None if the feed type is not recognized or something goes wrong.
+        """
+        try:
+            root = doc.getElementsByTagName("rss")[0]
+            channel = root.getElementsByTagName("channel")
+            if not channel:
+                return None
+            channel = channel[0]
+            channel_title = self._get_data_by_tag_name(channel, "title")
+            channel_description = self._get_data_by_tag_name(channel, "description")
+            items = channel.getElementsByTagName("item")
+            if channel_title:
+                md_text = f"# {channel_title}\n"
+            if channel_description:
+                md_text += f"{channel_description}\n"
+            if not items:
+                items = []
+            for item in items:
+                title = self._get_data_by_tag_name(item, "title")
+                description = self._get_data_by_tag_name(item, "description")
+                pubDate = self._get_data_by_tag_name(item, "pubDate")
+                content = self._get_data_by_tag_name(item, "content:encoded")
+
+                if title:
+                    md_text += f"\n## {title}\n"
+                if pubDate:
+                    md_text += f"Published on: {pubDate}\n"
+                if description:
+                    md_text += self._parse_content(description)
+                if content:
+                    md_text += self._parse_content(content)
+
+            return DocumentConverterResult(
+                title=channel_title,
+                text_content=md_text,
+            )
+        except BaseException as _:
+            print(traceback.format_exc())
+            return None
+
+    def _parse_content(self, content: str) -> str:
+        """Parse the content of an RSS feed item"""
+        try:
+            # using bs4 because many RSS feeds have HTML-styled content
+            soup = BeautifulSoup(content, "html.parser")
+            return _CustomMarkdownify().convert_soup(soup)
+        except BaseException as _:
+            return content
+
+    def _get_data_by_tag_name(
+        self, element: minidom.Element, tag_name: str
+    ) -> Union[str, None]:
+        """Get data from first child element with the given tag name.
+        Returns None when no such element is found.
+        """
+        nodes = element.getElementsByTagName(tag_name)
+        if not nodes:
+            return None
+        fc = nodes[0].firstChild
+        if fc:
+            return fc.data
+        return None
+
+
 class WikipediaConverter(DocumentConverter):
     """Handle Wikipedia pages separately, focusing only on the main document content."""
 
@@ -401,6 +541,67 @@ class YouTubeConverter(DocumentConverter):
         return None
 
 
+class IpynbConverter(DocumentConverter):
+    """Converts Jupyter Notebook (.ipynb) files to Markdown."""
+
+    def convert(
+        self, local_path: str, **kwargs: Any
+    ) -> Union[None, DocumentConverterResult]:
+        # Bail if not ipynb
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".ipynb":
+            return None
+
+        # Parse and convert the notebook
+        result = None
+        with open(local_path, "rt", encoding="utf-8") as fh:
+            notebook_content = json.load(fh)
+            result = self._convert(notebook_content)
+
+        return result
+
+    def _convert(self, notebook_content: dict) -> Union[None, DocumentConverterResult]:
+        """Helper function that converts notebook JSON content to Markdown."""
+        try:
+            md_output = []
+            title = None
+
+            for cell in notebook_content.get("cells", []):
+                cell_type = cell.get("cell_type", "")
+                source_lines = cell.get("source", [])
+
+                if cell_type == "markdown":
+                    md_output.append("".join(source_lines))
+
+                    # Extract the first # heading as title if not already found
+                    if title is None:
+                        for line in source_lines:
+                            if line.startswith("# "):
+                                title = line.lstrip("# ").strip()
+                                break
+
+                elif cell_type == "code":
+                    # Code cells are wrapped in Markdown code blocks
+                    md_output.append(f"```python\n{''.join(source_lines)}\n```")
+                elif cell_type == "raw":
+                    md_output.append(f"```\n{''.join(source_lines)}\n```")
+
+            md_text = "\n\n".join(md_output)
+
+            # Check for title in notebook metadata
+            title = notebook_content.get("metadata", {}).get("title", title)
+
+            return DocumentConverterResult(
+                title=title,
+                text_content=md_text,
+            )
+
+        except Exception as e:
+            raise FileConversionException(
+                f"Error converting .ipynb file: {str(e)}"
+            ) from e
+
+
 class BingSerpConverter(DocumentConverter):
     """
     Handle Bing results pages (only the organic search results).
@@ -681,7 +882,7 @@ class WavConverter(MediaConverter):
     """
 
     def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
-        # Bail if not a XLSX
+        # Bail if not a WAV
         extension = kwargs.get("file_extension", "")
         if extension.lower() != ".wav":
             return None
@@ -797,7 +998,7 @@ class ImageConverter(MediaConverter):
     """
 
     def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
-        # Bail if not a XLSX
+        # Bail if not an image
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".jpg", ".jpeg", ".png"]:
             return None
@@ -847,8 +1048,6 @@ class ImageConverter(MediaConverter):
         if prompt is None or prompt.strip() == "":
             prompt = "Write a detailed caption for this image."
 
-        sys.stderr.write(f"llm Prompt:\n{prompt}\n")
-
         data_uri = ""
         with open(local_path, "rb") as image_file:
             content_type, encoding = mimetypes.guess_type("_dummy" + extension)
@@ -1010,14 +1209,46 @@ class MarkItDown:
         self,
         requests_session: Optional[requests.Session] = None,
         llm_client: Optional[Any] = None,
-        llm_model: Optional[Any] = None,
+        llm_model: Optional[str] = None,
         style_map: Optional[str] = None,
+        # Deprecated
+        mlm_client: Optional[Any] = None,
+        mlm_model: Optional[str] = None,
     ):
         if requests_session is None:
             self._requests_session = requests.Session()
         else:
             self._requests_session = requests_session
 
+        # Handle deprecation notices
+        #############################
+        if mlm_client is not None:
+            if llm_client is None:
+                warn(
+                    "'mlm_client' is deprecated, and was renamed 'llm_client'.",
+                    DeprecationWarning,
+                )
+                llm_client = mlm_client
+                mlm_client = None
+            else:
+                raise ValueError(
+                    "'mlm_client' is deprecated, and was renamed 'llm_client'. Do not use both at the same time. Just use 'llm_client' instead."
+                )
+
+        if mlm_model is not None:
+            if llm_model is None:
+                warn(
+                    "'mlm_model' is deprecated, and was renamed 'llm_model'.",
+                    DeprecationWarning,
+                )
+                llm_model = mlm_model
+                mlm_model = None
+            else:
+                raise ValueError(
+                    "'mlm_model' is deprecated, and was renamed 'llm_model'. Do not use both at the same time. Just use 'llm_model' instead."
+                )
+        #############################
+
         self._llm_client = llm_client
         self._llm_model = llm_model
         self._style_map = style_map
@@ -1029,6 +1260,7 @@ class MarkItDown:
         # To this end, the most specific converters should appear below the most generic converters
         self.register_page_converter(PlainTextConverter())
         self.register_page_converter(HtmlConverter())
+        self.register_page_converter(RSSConverter())
         self.register_page_converter(WikipediaConverter())
         self.register_page_converter(YouTubeConverter())
         self.register_page_converter(BingSerpConverter())
@@ -1038,6 +1270,7 @@ class MarkItDown:
         self.register_page_converter(WavConverter())
         self.register_page_converter(Mp3Converter())
         self.register_page_converter(ImageConverter())
+        self.register_page_converter(IpynbConverter())
         self.register_page_converter(PdfConverter())
         self.register_page_converter(ZipConverter())
 

tests/test_files/test_notebook.ipynb

@@ -0,0 +1,89 @@
+{
+    "cells": [
+        {
+            "cell_type": "markdown",
+            "id": "0f61db80",
+            "metadata": {},
+            "source": [
+                "# Test Notebook"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 11,
+            "id": "3f2a5bbd",
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "markitdown\n"
+                    ]
+                }
+            ],
+            "source": [
+                "print('markitdown')"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "id": "9b9c0468",
+            "metadata": {},
+            "source": [
+                "## Code Cell Below"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 10,
+            "id": "37d8088a",
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "42\n"
+                    ]
+                }
+            ],
+            "source": [
+                "# comment in code\n",
+                "print(42)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "id": "2e3177bd",
+            "metadata": {},
+            "source": [
+                "End\n",
+                "\n",
+                "---"
+            ]
+        }
+    ],
+    "metadata": {
+        "kernelspec": {
+            "display_name": "Python 3",
+            "language": "python",
+            "name": "python3"
+        },
+        "language_info": {
+            "codemirror_mode": {
+                "name": "ipython",
+                "version": 3
+            },
+            "file_extension": ".py",
+            "mimetype": "text/x-python",
+            "name": "python",
+            "nbconvert_exporter": "python",
+            "pygments_lexer": "ipython3",
+            "version": "3.12.8"
+        },
+        "title": "Test Notebook Title"
+    },
+    "nbformat": 4,
+    "nbformat_minor": 5
+}

tests/test_markitdown.py

@@ -6,11 +6,23 @@ import shutil
 import pytest
 import requests
 
+from warnings import catch_warnings, resetwarnings
+
 from markitdown import MarkItDown
 
 skip_remote = (
     True if os.environ.get("GITHUB_ACTIONS") else False
 )  # Don't run these tests in CI
+
+
+# Don't run the llm tests without a key and the client library
+skip_llm = False if os.environ.get("OPENAI_API_KEY") else True
+try:
+    import openai
+except ModuleNotFoundError:
+    skip_llm = True
+
+# Skip exiftool tests if not installed
 skip_exiftool = shutil.which("exiftool") is None
 
 TEST_FILES_DIR = os.path.join(os.path.dirname(__file__), "test_files")
@@ -78,6 +90,13 @@ BLOG_TEST_STRINGS = [
     "an example where high cost can easily prevent a generic complex",
 ]
 
+
+RSS_TEST_STRINGS = [
+    "The Official Microsoft Blog",
+    "In the case of AI, it is absolutely true that the industry is moving incredibly fast",
+]
+
+
 WIKIPEDIA_TEST_URL = "https://en.wikipedia.org/wiki/Microsoft"
 WIKIPEDIA_TEST_STRINGS = [
     "Microsoft entered the operating system (OS) business in 1980 with its own version of [Unix]",
@@ -107,6 +126,10 @@ CSV_CP932_TEST_STRINGS = [
     "髙橋淳,35,名古屋",
 ]
 
+LLM_TEST_STRINGS = [
+    "5bda1dd6",
+]
+
 
 @pytest.mark.skipif(
     skip_remote,
@@ -208,6 +231,12 @@ def test_markitdown_local() -> None:
     for test_string in SERP_TEST_STRINGS:
         assert test_string in text_content
 
+    # Test RSS processing
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_rss.xml"))
+    text_content = result.text_content.replace("\\", "")
+    for test_string in RSS_TEST_STRINGS:
+        assert test_string in text_content
+
     ## Test non-UTF-8 encoding
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_mskanji.csv"))
     text_content = result.text_content.replace("\\", "")
@@ -229,8 +258,63 @@ def test_markitdown_exiftool() -> None:
         assert target in result.text_content
 
 
+def test_markitdown_deprecation() -> None:
+    try:
+        with catch_warnings(record=True) as w:
+            test_client = object()
+            markitdown = MarkItDown(mlm_client=test_client)
+            assert len(w) == 1
+            assert w[0].category is DeprecationWarning
+            assert markitdown._llm_client == test_client
+    finally:
+        resetwarnings()
+
+    try:
+        with catch_warnings(record=True) as w:
+            markitdown = MarkItDown(mlm_model="gpt-4o")
+            assert len(w) == 1
+            assert w[0].category is DeprecationWarning
+            assert markitdown._llm_model == "gpt-4o"
+    finally:
+        resetwarnings()
+
+    try:
+        test_client = object()
+        markitdown = MarkItDown(mlm_client=test_client, llm_client=test_client)
+        assert False
+    except ValueError:
+        pass
+
+    try:
+        markitdown = MarkItDown(mlm_model="gpt-4o", llm_model="gpt-4o")
+        assert False
+    except ValueError:
+        pass
+
+
+@pytest.mark.skipif(
+    skip_llm,
+    reason="do not run llm tests without a key",
+)
+def test_markitdown_llm() -> None:
+    client = openai.OpenAI()
+    markitdown = MarkItDown(llm_client=client, llm_model="gpt-4o")
+
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_llm.jpg"))
+
+    for test_string in LLM_TEST_STRINGS:
+        assert test_string in result.text_content
+
+    # This is not super precise. It would also accept "red square", "blue circle",
+    # "the square is not blue", etc. But it's sufficient for this test.
+    for test_string in ["red", "circle", "blue", "square"]:
+        assert test_string in result.text_content.lower()
+
+
 if __name__ == "__main__":
     """Runs this file's tests from the command line."""
     test_markitdown_remote()
     test_markitdown_local()
     test_markitdown_exiftool()
+    test_markitdown_deprecation()
+    test_markitdown_llm()