check_cached.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ec620c8b-9a8f-4ee2-8886-62c811954a74",
   "metadata": {},
   "source": [
    "# Check if all files for a particular iid are cached\n",
    "\n",
    "This is related to Rich Signell's high res request, which is currently incredibly slowly advancing...(1 worker with 4% CPU and 25% RAM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "aa10929a-7f0a-4a26-bada-b0c9fc3ec784",
   "metadata": {},
   "outputs": [],
   "source": [
    "import gcsfs\n",
    "import cftime\n",
    "import numpy as np\n",
    "import datetime\n",
    "fs = gcsfs.GCSFileSystem()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e9c2c2ac-277c-4817-9e38-b0e8da84d13e",
   "metadata": {},
   "outputs": [],
   "source": [
    "iids = ['CMIP6.HighResMIP.MOHC.HadGEM3-GC31-HH.highres-future.r1i1p1f1.Omon.so.gn.v20200514',\n",
    " 'CMIP6.HighResMIP.NERC.HadGEM3-GC31-HH.hist-1950.r1i1p1f1.Omon.so.gn.v20200514',\n",
    " 'CMIP6.HighResMIP.NERC.HadGEM3-GC31-HH.hist-1950.r1i1p1f1.Omon.thetao.gn.v20200514',\n",
    " 'CMIP6.HighResMIP.MOHC.HadGEM3-GC31-HH.highres-future.r1i1p1f1.Omon.thetao.gn.v20200514']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "fc192eeb-78af-4a52-b18d-1d7e1c10f3d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "iid = iids[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "9760b70b-f262-4e12-b70f-d744c473c3fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_cached = fs.ls('gs://leap-scratch/data-library/cmip6-pgf-ingestion/cache')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "96e5a791-c4a9-4a3b-8455-f36a7bc02d72",
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_all_cached(iid:str, all_cached:list[str]) -> bool:\n",
    "    iid_modified = iid.lower().replace('.', '_')\n",
    "    cached_iid = [c for c in all_cached if iid_modified in c]\n",
    "    start_date_str_iid = [c.split('_')[-1].replace('.nc', '').split('-')[0] for c in cached_iid]\n",
    "    start_dates = [cftime.Datetime360Day(int(s[0:4]), int(s[4:6]), 1) for s in sorted(start_date_str_iid)]\n",
    "    start_dates_diff = np.diff(start_dates)\n",
    "    assert np.all(start_dates_diff==datetime.timedelta(days=30))\n",
    "    pri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "b7b2e87c-ffd3-45fc-bc08-7127b6360b60",
   "metadata": {},
   "outputs": [],
   "source": [
    "for iid in iids:\n",
    "    check_all_cached(iid, all_cached)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4ab4c5b5-9ee5-495c-b16a-ae6402331a4a",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}