{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "initial_id",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:32:54.760793Z",
     "start_time": "2026-01-22T10:32:51.598659Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import tempfile\n",
    "import atexit\n",
    "from pathlib import Path\n",
    "\n",
    "import urgap\n",
    "\n",
    "\n",
    "tmp_dir_obj = tempfile.TemporaryDirectory()\n",
    "tmp_dir = Path(tmp_dir_obj.name)\n",
    "\n",
    "atexit.register(tmp_dir_obj.cleanup)\n",
    "\n",
    "print(f\"Temp directory: {tmp_dir}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4ef2bf6d34e62595",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:48:51.000110Z",
     "start_time": "2026-01-22T10:48:50.997793Z"
    }
   },
   "outputs": [],
   "source": [
    "notebook_dir = Path.cwd()\n",
    "test_data_dir = notebook_dir.parent.parent / \"tests\" / \"data\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c6c171ddcdc2aec7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:08.147245Z",
     "start_time": "2026-01-22T10:49:08.138685Z"
    }
   },
   "outputs": [],
   "source": [
    "ufiles = urgap.UFileList(\n",
    "    [\n",
    "        urgap.UFile(\n",
    "            uri=f\"file://{test_data_dir}?uftype={urgap.uftypes.any.CSV}\"\n",
    "            f\"#unified_csvs/BSA1_xtandem_alanine_unified.csv\",\n",
    "        ),\n",
    "    ],\n",
    ")\n",
    "print(ufiles)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d65f0fc0fac9c3ba",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:11.374309Z",
     "start_time": "2026-01-22T10:49:11.371454Z"
    }
   },
   "outputs": [],
   "source": [
    "urun_dict = urgap.URunDict(\n",
    "    {\n",
    "        \"parameters\": {\n",
    "            \"FilterTabularToCSV:1.0.0\": {\n",
    "                \"-q\": \"500 < `exp_mz` < 1000\",\n",
    "            },\n",
    "            \"CompressToTar:1.0.0\": {},\n",
    "        },\n",
    "        \"unode_parameters\": {\n",
    "            \"storage_base_uri\": f\"file://{tmp_dir}\",\n",
    "        },\n",
    "    },\n",
    ")\n",
    "print(urun_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4f4d369f8052e676",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:13.138704Z",
     "start_time": "2026-01-22T10:49:13.128446Z"
    }
   },
   "outputs": [],
   "source": [
    "filter_node = urgap.init_unode(\"FilterTabularToCSV:1.0.0\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e010a2826315407a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:15.533851Z",
     "start_time": "2026-01-22T10:49:14.580485Z"
    }
   },
   "outputs": [],
   "source": [
    "filter_results = filter_node.run(urun_dict=urun_dict, ufiles=ufiles)\n",
    "print(filter_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "307ee4ed67723d3d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:29.573440Z",
     "start_time": "2026-01-22T10:49:29.562141Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(filter_results[0].path)\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8f3f52de7d04bf3e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:30.839560Z",
     "start_time": "2026-01-22T10:49:30.833066Z"
    }
   },
   "outputs": [],
   "source": [
    "compress_node = urgap.init_unode(\"CompressToTar:1.0.0\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3a2d461fb5812a5b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2026-01-22T10:49:32.458858Z",
     "start_time": "2026-01-22T10:49:32.327011Z"
    }
   },
   "outputs": [],
   "source": [
    "compress_results = compress_node.run(urun_dict=urun_dict, ufiles=filter_results)\n",
    "print(compress_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8e3f385ee462aecc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.13.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}