From b10a644021dafc56f259b850590980acbb637d0d Mon Sep 17 00:00:00 2001
From: Pol Dellaiera <pol.dellaiera@protonmail.com>
Date: Mon, 16 Dec 2024 21:08:30 +0100
Subject: [PATCH] python312Packages.markitdown: 0-unstable-2024-12-15

---
 .../python-modules/markitdown/default.nix     | 74 +++++++++++++++++++
 pkgs/top-level/python-packages.nix            |  2 +
 2 files changed, 76 insertions(+)
 create mode 100644 pkgs/development/python-modules/markitdown/default.nix

diff --git a/pkgs/development/python-modules/markitdown/default.nix b/pkgs/development/python-modules/markitdown/default.nix
new file mode 100644
index 0000000000000..b27e23b846778
--- /dev/null
+++ b/pkgs/development/python-modules/markitdown/default.nix
@@ -0,0 +1,74 @@
+{
+  lib,
+  buildPythonPackage,
+  fetchFromGitHub,
+  hatchling,
+  beautifulsoup4,
+  ffmpeg-headless,
+  mammoth,
+  markdownify,
+  numpy,
+  openpyxl,
+  pandas,
+  pathvalidate,
+  pdfminer-six,
+  puremagic,
+  pydub,
+  python-pptx,
+  requests,
+  speechrecognition,
+  youtube-transcript-api,
+  pytestCheckHook,
+  gitUpdater,
+}:
+
+buildPythonPackage {
+  pname = "markitdown";
+  version = "unstable-2024-12-15";
+  pyproject = true;
+
+  src = fetchFromGitHub {
+    owner = "microsoft";
+    repo = "markitdown";
+    rev = "6dad1cca967f0d8eb15233104ab5b30fc37e6649";
+    hash = "sha256-oview6cA4bFPq5fp6UegW34uKHF1PybAh+FUcfgPa+Q=";
+  };
+
+  build-system = [ hatchling ];
+
+  dependencies = [
+    beautifulsoup4
+    ffmpeg-headless
+    mammoth
+    markdownify
+    numpy
+    openpyxl
+    pandas
+    pathvalidate
+    pdfminer-six
+    puremagic
+    pydub
+    python-pptx
+    requests
+    speechrecognition
+    youtube-transcript-api
+  ];
+
+  pythonImportsCheck = [ "markitdown" ];
+
+  nativeCheckInputs = [ pytestCheckHook ];
+
+  disabledTests = [
+    # Require network access
+    "test_markitdown_remote"
+  ];
+
+  passthru.updateScripts = gitUpdater { };
+
+  meta = {
+    description = "Python tool for converting files and office documents to Markdown";
+    homepage = "https://github.com/microsoft/markitdown";
+    license = lib.licenses.mit;
+    maintainers = with lib.maintainers; [ drupol ];
+  };
+}
diff --git a/pkgs/top-level/python-packages.nix b/pkgs/top-level/python-packages.nix
index 4a807b9781e5c..fb8e06097ad75 100644
--- a/pkgs/top-level/python-packages.nix
+++ b/pkgs/top-level/python-packages.nix
@@ -7869,6 +7869,8 @@ self: super: with self; {
 
   markdownify  = callPackage ../development/python-modules/markdownify { };
 
+  markitdown  = callPackage ../development/python-modules/markitdown { };
+
   marko = callPackage ../development/python-modules/marko { };
 
   markupsafe = callPackage ../development/python-modules/markupsafe { };