From dd977ca1d89d933682d6048d716e939ace7e459f Mon Sep 17 00:00:00 2001
From: Hemanth HM <hemanth.hm@gmail.com>
Date: Mon, 30 Dec 2024 21:30:22 -0800
Subject: [PATCH] Add serve command to start CORS enabled Flask server

Add a `serve` command to start a CORS-enabled Flask server for file conversion.

* **New Flask Server**: Add `src/markitdown/server.py` to define a Flask server with CORS enabled and a route to convert files to markdown.

* **Dependencies**: Update `pyproject.toml` to include `flask` and `flask-cors` as dependencies and add the `serve` command to the `[project.scripts]` section.
* **Documentation**: Update `README.md` with instructions on how to use the `serve` command.

* **Tests**: Add tests in `tests/test_markitdown.py` to verify the functionality of the `serve` command, including handling both URL and file POST data.

In case we decide not to have `serve` command we have [markdown-converter](https://pypi.org/project/markdown-converter/).
---
 README.md                | 17 +++++++++++++++++
 pyproject.toml           |  3 +++
 src/markitdown/server.py | 24 ++++++++++++++++++++++++
 tests/test_markitdown.py | 32 ++++++++++++++++++++++++++++++++
 4 files changed, 76 insertions(+)
 create mode 100644 src/markitdown/server.py

diff --git a/README.md b/README.md
index d2314c3..fbc9d12 100644
--- a/README.md
+++ b/README.md
@@ -72,6 +72,23 @@ print(result.text_content)
 docker build -t markitdown:latest .
 docker run --rm -i markitdown:latest < ~/your-file.pdf > output.md
 ```
+
+### Serve Command
+
+You can start a CORS-enabled Flask server to convert files to markdown using the `serve` command:
+
+```sh
+markitdown serve
+```
+
+The server will be available at `http://localhost:5000`. You can send a POST request to the `/convert` endpoint with a file to convert it to markdown.
+
+Example using `curl`:
+
+```sh
+curl -X POST -F 'file=@path-to-file.pdf' http://localhost:5000/convert
+```
+
 <details>
     
 <summary>Batch Processing Multiple Files</summary>
diff --git a/pyproject.toml b/pyproject.toml
index 3e14cec..8e156f0 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,6 +40,8 @@ dependencies = [
   "pathvalidate",
   "charset-normalizer",
   "openai",
+  "flask",
+  "flask-cors",
 ]
 
 [project.urls]
@@ -52,6 +54,7 @@ path = "src/markitdown/__about__.py"
 
 [project.scripts]
 markitdown = "markitdown.__main__:main"
+serve = "markitdown.server:app.run"
 
 [tool.hatch.envs.types]
 extra-dependencies = [
diff --git a/src/markitdown/server.py b/src/markitdown/server.py
new file mode 100644
index 0000000..55ede9f
--- /dev/null
+++ b/src/markitdown/server.py
@@ -0,0 +1,24 @@
+from flask import Flask, request, jsonify
+from flask_cors import CORS
+from markitdown import MarkItDown
+
+app = Flask(__name__)
+CORS(app)
+
+markitdown = MarkItDown()
+
+@app.route('/convert', methods=['POST'])
+def convert():
+    if 'file' in request.files:
+        file = request.files['file']
+        result = markitdown.convert(file.stream, file_extension=file.filename.split('.')[-1])
+        return jsonify({'content': result.text_content})
+    elif 'url' in request.form:
+        url = request.form['url']
+        result = markitdown.convert(url)
+        return jsonify({'content': result.text_content})
+    else:
+        return jsonify({'error': 'No file or URL provided'}), 400
+
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=5000)
diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index 4a981bd..b494dd8 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -130,6 +130,11 @@
     "5bda1dd6",
 ]
 
+# New test strings for the serve command
+SERVE_TEST_STRINGS = [
+    "While there is contemporaneous exploration of multi-agent approaches"
+]
+
 
 # --- Helper Functions ---
 def validate_strings(result, expected_strings, exclude_strings=None):
@@ -300,6 +305,32 @@ def test_markitdown_llm() -> None:
         assert test_string in result.text_content.lower()
 
 
+# New test for the serve command
+def test_markitdown_serve() -> None:
+    from src.markitdown.server import app
+
+    client = app.test_client()
+
+    # Test with file
+    response = client.post(
+        "/convert",
+        data={"file": (io.BytesIO(b"test content"), "test.pdf")},
+        content_type="multipart/form-data",
+    )
+    assert response.status_code == 200
+    for test_string in SERVE_TEST_STRINGS:
+        assert test_string in response.json["content"]
+
+    # Test with URL
+    response = client.post(
+        "/convert",
+        data={"url": PDF_TEST_URL},
+    )
+    assert response.status_code == 200
+    for test_string in SERVE_TEST_STRINGS:
+        assert test_string in response.json["content"]
+
+
 if __name__ == "__main__":
     """Runs this file's tests from the command line."""
     test_markitdown_remote()
@@ -307,3 +338,4 @@ def test_markitdown_llm() -> None:
     test_markitdown_exiftool()
     test_markitdown_deprecation()
     test_markitdown_llm()
+    test_markitdown_serve()