zarr_safe_store.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections.abc import MutableMapping\n",
    "\n",
    "\n",
    "class SafeStoreWrapper(MutableMapping):\n",
    "    \n",
    "    def __init__(self, store):\n",
    "        self.store = store\n",
    "        \n",
    "    def __getitem__(self, key):\n",
    "        try:\n",
    "            return self.store[key]\n",
    "        except KeyError as e:\n",
    "            # always raise a runtime error to ensure zarr propagates the exception\n",
    "            raise RuntimeError(e)\n",
    "    \n",
    "    def __setitem__(self, key, value):\n",
    "        self.store[key] = value\n",
    "        \n",
    "    def __delitem__(self, key):\n",
    "        del self.store[key]\n",
    "        \n",
    "    def __contains__(self, key):\n",
    "        return key in self.store\n",
    "    \n",
    "    def __iter__(self):\n",
    "        return iter(self.store)\n",
    "    \n",
    "    def __len__(self):\n",
    "        return len(self.store)\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# SafeStoreWrapper can be used to wrap any kind of zarr store, e.g., a GCSMap ...\n",
    "\n",
    "import gcsfs\n",
    "gcs_bucket_fs = gcsfs.GCSFileSystem(project='malariagen-jupyterhub', token='anon', access='read_only')\n",
    "storage_path = 'ag1000g-release/phase1.AR3/variation/main/zarr/ag1000g.phase1.ar3'\n",
    "store = gcsfs.mapping.GCSMap(storage_path, gcs=gcs_bucket_fs, check=False, create=False)\n",
    "safe_store = SafeStoreWrapper(store)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'KeyError'>\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    # demonstrate what the original store does with a key that doesn't exist\n",
    "    store['foo']\n",
    "except Exception as e:\n",
    "    print(type(e))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'RuntimeError'>\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    # demonstrate what the wrapped store does with a key that doesn't exist\n",
    "    safe_store['foo']\n",
    "except Exception as e:\n",
    "    print(type(e))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([     103,      163,      192, ..., 49361791, 49361815, 49362329],\n",
       "      dtype=int32)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use like this, can guarantee that zarr will raise exception if any chunk is missing, \n",
    "# rather than fill in with fill value ...\n",
    "import zarr\n",
    "root = zarr.Group(store=safe_store)\n",
    "pos = root['2L/variants/POS'][:]\n",
    "pos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# etc. ..."
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}