{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b2c9066b-db07-4ebc-aebe-41691abad1c0",
   "metadata": {},
   "source": [
    "## Writing to a file target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c32a1f28-fd92-42df-ba34-5e5d410fa4ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "test.grib:   0%|          | 0.00/1.03k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get input GRIB data\n",
    "import os\n",
    "import earthkit.data as ekd\n",
    "ds = ekd.from_source(\"sample\", \"test.grib\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd7bfa87-808c-4fa4-b4c4-1a60fbdeb467",
   "metadata": {},
   "source": [
    "#### Using a file path"
   ]
  },
  {
   "cell_type": "raw",
   "id": "ad3c3648-f7a3-4bb8-bb50-47be0f1dbf6a",
   "metadata": {
    "editable": true,
    "raw_mimetype": "text/restructuredtext",
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "source": [
    "We use :func:`to_target` to write the GRIB fieldlist/field into a file. The :ref:`encoder <encoders>` is automatically guessed from the input data. Alternatively, we can create a target object with :func:`create_target` and directly write to it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "216afc9e-c6b3-48c2-a62a-9aca346447df",
   "metadata": {},
   "outputs": [],
   "source": [
    "# calling to_target\n",
    "ds.to_target(\"file\", \"_res_t_file_handler.grib\")\n",
    "\n",
    "# using write on the target object\n",
    "with ekd.create_target(\"file\", \"_res_t_file_handler.grib\") as t:\n",
    "    t.write(ds)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "253b32b5-d6cd-41f7-96d4-5933931aec3a",
   "metadata": {},
   "source": [
    "#### Using a file-like object"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41473048-e8a1-42b2-a39a-9cc7cf46617e",
   "metadata": {},
   "source": [
    "A file-like object passed to the target is not closed, even when the target is closed or created with a context manager."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "90375bdd-2f72-40ad-a94c-4ab635acb54d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False\n"
     ]
    }
   ],
   "source": [
    "fp = open(\"_res_t_file_handler.grib\", \"wb\")\n",
    "ds.to_target(\"file\", fp)\n",
    "\n",
    "# the file object is still open\n",
    "print(fp.closed)\n",
    "\n",
    "# we need to close it manually\n",
    "fp.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "348834e3-0197-4c79-ba27-738855c4d2eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False\n"
     ]
    }
   ],
   "source": [
    "fp = open(\"_res_t_file_handler.grib\", \"wb\")\n",
    "\n",
    "# the context manager will call close() on the target\n",
    "with ekd.create_target(\"file\", fp) as t:\n",
    "    t.write(ds)\n",
    "\n",
    "# the file object is still open\n",
    "print(fp.closed)\n",
    "\n",
    "# we need to close it manually\n",
    "fp.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "278183b2-f502-4b24-9975-f4e704fc14a9",
   "metadata": {},
   "source": [
    "The simplest solution to this problem is to use a context manager for the file-like object.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "63a34551-35d8-4262-8759-3e69fca05bc7",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"_res_t_file_handler.grib\", \"wb\") as fp:\n",
    "    ds.to_target(\"file\", fp)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "233e46d1-c052-405c-bbf8-9bad9f512497",
   "metadata": {},
   "source": [
    "#### Appending to a file"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92fad9cb-f9f4-4e79-9357-b3d888504890",
   "metadata": {},
   "source": [
    "When using a file path we can use the ``append=True`` option to append to the output."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "afe1de08-866a-4943-9918-4755f1e5ce68",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calling to_target\n",
    "out_file = \"_res_t1_file_handler.grib\"\n",
    "if os.path.isfile(out_file):\n",
    "    os.remove(out_file)\n",
    "    \n",
    "ds[0].to_target(\"file\", out_file, append=True)\n",
    "len(ekd.from_source(\"file\", out_file))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "96583f35-48bd-4bc9-a4e6-811758d5ea06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds[1].to_target(\"file\", out_file, append=True)\n",
    "len(ekd.from_source(\"file\", out_file))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3709c465-551b-4369-9b03-61ea5f50e56f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dev_ecc",
   "language": "python",
   "name": "dev_ecc"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}