{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copyright 2024 Google LLC\n",
    "#\n",
    "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
    "# you may not use this file except in compliance with the License.\n",
    "# You may obtain a copy of the License at\n",
    "#\n",
    "#     https://www.apache.org/licenses/LICENSE-2.0\n",
    "#\n",
    "# Unless required by applicable law or agreed to in writing, software\n",
    "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
    "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
    "# See the License for the specific language governing permissions and\n",
    "# limitations under the License."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyzing package downloads from PyPI with BigQuery DataFrames\n",
    "\n",
    "In this notebook, you'll use the [PyPI public dataset](https://console.cloud.google.com/marketplace/product/gcp-public-data-pypi/pypi) and the [deps.dev public dataset](https://deps.dev/) to visualize Python package downloads for a package and its dependencies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Choose a package which you want to visualize.\n",
    "package_name = \"pandas\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import bigframes.pandas as bpd\n",
    "\n",
    "# Use `ordering_mode=\"partial\"` for more efficient query generation, but\n",
    "# some pandas-compatible methods may not be possible without a total ordering.\n",
    "bpd.options.bigquery.ordering_mode = \"partial\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Counting downloads and tracking dependencies\n",
    "\n",
    "The [PyPI `file_downloads`](https://console.cloud.google.com/bigquery?ws=!1m5!1m4!4m3!1sbigquery-public-data!2spypi!3sfile_downloads) table contains a row for each time there is a download request for a package. The [deps.dev Dependencies](https://console.cloud.google.com/bigquery?ws=!1m5!1m4!4m3!1sbigquery-public-data!2sdeps_dev_v1!3sDependencies) table contains a row for each dependency of each package.\n",
    "\n",
    "When `ordering_mode = \"partial\"`, `read_gbq_table` creates a DataFrame representing the table, but the DataFrame has no native ordering or index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import bigframes.enums\n",
    "\n",
    "# Without ordering_mode = \"partial\" it is recommended that you set\n",
    "# the \"filters\" parameter to limit the number of rows subsequent queries\n",
    "# have to read.\n",
    "pypi = bpd.read_gbq_table(\n",
    "    \"bigquery-public-data.pypi.file_downloads\",\n",
    "\n",
    "    # Using ordering_mode = \"partial\" changes the default index to a \"NULL\"\n",
    "    # index, meaning no index is available for implicit joins.\n",
    "    #\n",
    "    # Setting this explicitly avoids a DefaultIndexWarning.\n",
    "    index_col=bigframes.enums.DefaultIndexKind.NULL,\n",
    ")\n",
    "deps = bpd.read_gbq_table(\n",
    "    \"bigquery-public-data.deps_dev_v1.Dependencies\",\n",
    "    index_col=bigframes.enums.DefaultIndexKind.NULL,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Limit to the most recent 7 days of data\n",
    "\n",
    "The PyPI and deps.dev tables are partitioned by date. Query only the most recent 7 days of data to reduce the number of bytes scanned.\n",
    "\n",
    "Just as with the default ordering mode, filters can be describe in a pandas-compatible way by passing a Boolean Series to the DataFrame's `__getitem__` accessor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "\n",
    "now = datetime.datetime.now(datetime.timezone.utc)\n",
    "last_7_days = now - datetime.timedelta(days=7)\n",
    "last_30_days = now - datetime.timedelta(days=30)\n",
    "pypi = pypi[pypi[\"timestamp\"] > last_7_days]\n",
    "deps = deps[deps[\"SnapshotAt\"] > last_30_days]  # deps are refreshed less frequently\n",
    "deps = deps[deps[\"System\"] == \"PYPI\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "**⚠ Warning**\n",
    "\n",
    "Without `ordering_mode = \"partial\"`, these filters do not change the number of bytes scanned. Instead, add column and row filters at \"read\" time. For example,\n",
    "\n",
    "```\n",
    "import datetime\n",
    "\n",
    "last_7_days = datetime.datetime.now(datetime.timezone.utc) - datetime.timedelta(days=7)\n",
    "\n",
    "# Without ordering_mode = \"partial\", one must limit the data at \"read\" time to reduce bytes scanned.\n",
    "pypi = bpd.read_gbq_table(\n",
    "    \"bigquery-public-data.pypi.file_downloads\",\n",
    "    columns=[\"timestamp\", \"project\"],\n",
    "    filters=[(\"timestamp\", \">\", last_7_days)],\n",
    ")\n",
    "```\n",
    "\n",
    "`head()` is not available when no ordering has been established. It fails with `OrderRequiredError`. Use `peek()` instead to download a sample of the data. This will be much more efficient, as the query doesn't need to order all rows to determine which are first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Query job 423d9d93-1495-4c76-b8c2-e830a6e19ff4 is DONE. 110.3 MB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:423d9d93-1495-4c76-b8c2-e830a6e19ff4&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>country_code</th>\n",
       "      <th>url</th>\n",
       "      <th>project</th>\n",
       "      <th>file</th>\n",
       "      <th>details</th>\n",
       "      <th>tls_protocol</th>\n",
       "      <th>tls_cipher</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-09-18 18:15:04+00:00</td>\n",
       "      <td>US</td>\n",
       "      <td>/packages/ff/c8/4cd4b2834012ffc71ae3fd69187f08...</td>\n",
       "      <td>aiobreaker</td>\n",
       "      <td>{'filename': 'aiobreaker-1.2.0-py3-none-any.wh...</td>\n",
       "      <td>{'installer': {'name': 'pip', 'version': '21.1...</td>\n",
       "      <td>TLSv1.3</td>\n",
       "      <td>TLS_AES_128_GCM_SHA256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-09-18 18:29:50+00:00</td>\n",
       "      <td>US</td>\n",
       "      <td>/packages/21/8e/4562029e179226051cd4aa3135444d...</td>\n",
       "      <td>aiobotocore</td>\n",
       "      <td>{'filename': 'aiobotocore-1.3.0.tar.gz', 'proj...</td>\n",
       "      <td>{'installer': {'name': 'pip', 'version': '24.1...</td>\n",
       "      <td>TLSv1.2</td>\n",
       "      <td>ECDHE-RSA-AES128-GCM-SHA256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-09-18 18:22:14+00:00</td>\n",
       "      <td>US</td>\n",
       "      <td>/packages/11/16/4226e59bb72e096d9809ccedf349a1...</td>\n",
       "      <td>aiobotocore</td>\n",
       "      <td>{'filename': 'aiobotocore-2.0.1.tar.gz', 'proj...</td>\n",
       "      <td>{'installer': {'name': 'pip', 'version': '24.2...</td>\n",
       "      <td>TLSv1.2</td>\n",
       "      <td>ECDHE-RSA-AES128-GCM-SHA256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-09-18 18:22:08+00:00</td>\n",
       "      <td>US</td>\n",
       "      <td>/packages/11/16/4226e59bb72e096d9809ccedf349a1...</td>\n",
       "      <td>aiobotocore</td>\n",
       "      <td>{'filename': 'aiobotocore-2.0.1.tar.gz', 'proj...</td>\n",
       "      <td>{'installer': {'name': 'pip', 'version': '24.2...</td>\n",
       "      <td>TLSv1.2</td>\n",
       "      <td>ECDHE-RSA-AES128-GCM-SHA256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-09-18 18:29:22+00:00</td>\n",
       "      <td>US</td>\n",
       "      <td>/packages/54/b7/453119271cc4c36b07fdeab9b0ff25...</td>\n",
       "      <td>aiobotocore</td>\n",
       "      <td>{'filename': 'aiobotocore-2.3.3.tar.gz', 'proj...</td>\n",
       "      <td>{'installer': {'name': 'pip', 'version': '24.1...</td>\n",
       "      <td>TLSv1.2</td>\n",
       "      <td>ECDHE-RSA-AES128-GCM-SHA256</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   timestamp country_code  \\\n",
       "0  2024-09-18 18:15:04+00:00           US   \n",
       "1  2024-09-18 18:29:50+00:00           US   \n",
       "2  2024-09-18 18:22:14+00:00           US   \n",
       "3  2024-09-18 18:22:08+00:00           US   \n",
       "4  2024-09-18 18:29:22+00:00           US   \n",
       "\n",
       "                                                 url      project  \\\n",
       "0  /packages/ff/c8/4cd4b2834012ffc71ae3fd69187f08...   aiobreaker   \n",
       "1  /packages/21/8e/4562029e179226051cd4aa3135444d...  aiobotocore   \n",
       "2  /packages/11/16/4226e59bb72e096d9809ccedf349a1...  aiobotocore   \n",
       "3  /packages/11/16/4226e59bb72e096d9809ccedf349a1...  aiobotocore   \n",
       "4  /packages/54/b7/453119271cc4c36b07fdeab9b0ff25...  aiobotocore   \n",
       "\n",
       "                                                file  \\\n",
       "0  {'filename': 'aiobreaker-1.2.0-py3-none-any.wh...   \n",
       "1  {'filename': 'aiobotocore-1.3.0.tar.gz', 'proj...   \n",
       "2  {'filename': 'aiobotocore-2.0.1.tar.gz', 'proj...   \n",
       "3  {'filename': 'aiobotocore-2.0.1.tar.gz', 'proj...   \n",
       "4  {'filename': 'aiobotocore-2.3.3.tar.gz', 'proj...   \n",
       "\n",
       "                                             details tls_protocol  \\\n",
       "0  {'installer': {'name': 'pip', 'version': '21.1...      TLSv1.3   \n",
       "1  {'installer': {'name': 'pip', 'version': '24.1...      TLSv1.2   \n",
       "2  {'installer': {'name': 'pip', 'version': '24.2...      TLSv1.2   \n",
       "3  {'installer': {'name': 'pip', 'version': '24.2...      TLSv1.2   \n",
       "4  {'installer': {'name': 'pip', 'version': '24.1...      TLSv1.2   \n",
       "\n",
       "                    tls_cipher  \n",
       "0       TLS_AES_128_GCM_SHA256  \n",
       "1  ECDHE-RSA-AES128-GCM-SHA256  \n",
       "2  ECDHE-RSA-AES128-GCM-SHA256  \n",
       "3  ECDHE-RSA-AES128-GCM-SHA256  \n",
       "4  ECDHE-RSA-AES128-GCM-SHA256  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Warning: Ensure bpd.options.bigquery.ordering_mode = \"partial\" or else\n",
    "# this query() will cause a full table scan because of the sequential index.\n",
    "assert bpd.options.bigquery.ordering_mode == \"partial\"\n",
    "pypi.peek()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Query job 3a421217-59e2-4722-8382-0930f0a3b9ee is DONE. 1.5 MB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:3a421217-59e2-4722-8382-0930f0a3b9ee&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SnapshotAt</th>\n",
       "      <th>System</th>\n",
       "      <th>Name</th>\n",
       "      <th>Version</th>\n",
       "      <th>Dependency</th>\n",
       "      <th>MinimumDepth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-08-29 04:39:16.121656+00:00</td>\n",
       "      <td>PYPI</td>\n",
       "      <td>zxkane-cdk-construct-simple-nat</td>\n",
       "      <td>0.2.89</td>\n",
       "      <td>{'System': 'PYPI', 'Name': 'attrs', 'Version':...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-08-29 04:39:16.121656+00:00</td>\n",
       "      <td>PYPI</td>\n",
       "      <td>zxkane-cdk-construct-simple-nat</td>\n",
       "      <td>0.2.82</td>\n",
       "      <td>{'System': 'PYPI', 'Name': 'attrs', 'Version':...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-08-29 04:39:16.121656+00:00</td>\n",
       "      <td>PYPI</td>\n",
       "      <td>zxkane-cdk-construct-simple-nat</td>\n",
       "      <td>0.2.88</td>\n",
       "      <td>{'System': 'PYPI', 'Name': 'attrs', 'Version':...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-08-29 04:39:16.121656+00:00</td>\n",
       "      <td>PYPI</td>\n",
       "      <td>zxkane-cdk-construct-simple-nat</td>\n",
       "      <td>0.2.91</td>\n",
       "      <td>{'System': 'PYPI', 'Name': 'attrs', 'Version':...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-08-29 04:39:16.121656+00:00</td>\n",
       "      <td>PYPI</td>\n",
       "      <td>zxkane-cdk-construct-simple-nat</td>\n",
       "      <td>0.2.77</td>\n",
       "      <td>{'System': 'PYPI', 'Name': 'attrs', 'Version':...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         SnapshotAt System                             Name  \\\n",
       "0  2024-08-29 04:39:16.121656+00:00   PYPI  zxkane-cdk-construct-simple-nat   \n",
       "1  2024-08-29 04:39:16.121656+00:00   PYPI  zxkane-cdk-construct-simple-nat   \n",
       "2  2024-08-29 04:39:16.121656+00:00   PYPI  zxkane-cdk-construct-simple-nat   \n",
       "3  2024-08-29 04:39:16.121656+00:00   PYPI  zxkane-cdk-construct-simple-nat   \n",
       "4  2024-08-29 04:39:16.121656+00:00   PYPI  zxkane-cdk-construct-simple-nat   \n",
       "\n",
       "  Version                                         Dependency  MinimumDepth  \n",
       "0  0.2.89  {'System': 'PYPI', 'Name': 'attrs', 'Version':...             2  \n",
       "1  0.2.82  {'System': 'PYPI', 'Name': 'attrs', 'Version':...             2  \n",
       "2  0.2.88  {'System': 'PYPI', 'Name': 'attrs', 'Version':...             2  \n",
       "3  0.2.91  {'System': 'PYPI', 'Name': 'attrs', 'Version':...             2  \n",
       "4  0.2.77  {'System': 'PYPI', 'Name': 'attrs', 'Version':...             2  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "deps.peek()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Find dependencies for pandas\n",
    "\n",
    "Use assign to add columns to the DataFrame after a scalar operations, such as extracting a sub-field from a `STRUCT` column.\n",
    "\n",
    "Because the DataFrame has no index, this does not work if the new column belongs to a different table expression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "deps = deps.assign(DependencyName=deps[\"Dependency\"].struct.field(\"Name\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use an aggregation to identify the unique `DependencyName`s for the `pandas` package. Note: `drop_duplicates()` is not supported, as the order-based behavior such as `keep=\"first\"` is not applicable when using `ordering_mode = \"partial\"`.\n",
    "\n",
    "A DataFrame with no index still supports aggregation operations. Set `as_index=False` to keep the GROUP BY keys as regular columns, instead of turning them into an index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Query job 5b69917f-9ed7-483a-9241-0083acea9990 is DONE. 1.1 GB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:5b69917f-9ed7-483a-9241-0083acea9990&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Query job ac94c55d-ce8e-4694-ad97-55c933cf3053 is DONE. 123 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:ac94c55d-ce8e-4694-ad97-55c933cf3053&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>DependencyName</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>pandas</td>\n",
       "      <td>pytz</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>pandas</td>\n",
       "      <td>numpy</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>pandas</td>\n",
       "      <td>python-dateutil</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pandas</td>\n",
       "      <td>six</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>pandas</td>\n",
       "      <td>tzdata</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Name   DependencyName  size\n",
       "0  pandas             pytz   168\n",
       "1  pandas            numpy   168\n",
       "2  pandas  python-dateutil   168\n",
       "3  pandas              six   168\n",
       "4  pandas           tzdata    56"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "package_deps = deps[deps[\"Name\"] == package_name].groupby([\"Name\", \"DependencyName\"], as_index=False).size()\n",
    "package_deps.peek()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Count downloads for pandas and its dependencies\n",
    "\n",
    "The previous step created `pandas_deps` with all the dependencies of `pandas` but not pandas itself.\n",
    "\n",
    "Combine two DataFrames with the same column names with the `bigframes.pandas.concat` function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "pandas_and_deps = bpd.concat(\n",
    "    [\n",
    "        package_deps.drop(columns=[\"Name\", \"size\"]).rename(columns={\"DependencyName\": \"Name\"}),\n",
    "        bpd.DataFrame({\"Name\": [package_name]}),\n",
    "    ],\n",
    "\n",
    "    # To join DataFrames that have a NULL index, set ignore_index = True.\n",
    "    ignore_index=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since there is no index to implicitly join on, use the `merge` method to join two DataFrames by column name."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "pandas_pypi = pandas_and_deps.merge(pypi, how=\"inner\", left_on=\"Name\", right_on=\"project\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a time series to visualize by grouping by the date, extracted from the `timestamp` column.\n",
    "\n",
    "**Note:** If you don't `peek()` at your data and only do grouped aggregations, BigQuery DataFrames can eliminate unnecessary ordering from the compilation even without `ordering_mode = \"partial\"`.\n",
    "\n",
    "When BigQuery DataFrames aggregates over columns, those columns provide a\n",
    "unique key post-aggregation that is used for ordering. Any ordering applied before is overridden. By aggregating over\n",
    "a time series, the line plots will render in the expected order."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Query job 57037a4c-5b8b-4f30-a5c6-bfeb9731a38f is DONE. 270.4 GB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:57037a4c-5b8b-4f30-a5c6-bfeb9731a38f&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Query job a8ea0b8e-2260-4175-b80d-668a2411c6ad is DONE. 2.0 kB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:a8ea0b8e-2260-4175-b80d-668a2411c6ad&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "date        project        \n",
       "2024-09-17  numpy               2572982\n",
       "            pandas              2195692\n",
       "            python-dateutil     3608119\n",
       "            pytz                1863133\n",
       "            six                 3064640\n",
       "            tzdata               991989\n",
       "2024-09-18  numpy              13282573\n",
       "            pandas             10856758\n",
       "            python-dateutil    17877058\n",
       "            pytz                9450103\n",
       "            six                15225000\n",
       "            tzdata              5230039\n",
       "2024-09-19  numpy              13637868\n",
       "            pandas             11077817\n",
       "            python-dateutil    18449777\n",
       "            pytz                9690329\n",
       "            six                15706263\n",
       "            tzdata              5473910\n",
       "2024-09-20  numpy              12609524\n",
       "            pandas             10758593\n",
       "            python-dateutil    17257536\n",
       "            pytz                9082050\n",
       "            six                14489456\n",
       "            tzdata              5206738\n",
       "2024-09-21  numpy               8316481\n",
       "            pandas              7483241\n",
       "            python-dateutil    11604691\n",
       "            pytz                5494178\n",
       "            six                 8814983\n",
       "            tzdata              3141578\n",
       "2024-09-22  numpy               7768078\n",
       "            pandas              6566272\n",
       "            python-dateutil    10835755\n",
       "            pytz                5130018\n",
       "            six                 8297507\n",
       "            tzdata              2811247\n",
       "2024-09-23  numpy              12389164\n",
       "            pandas             10758931\n",
       "            python-dateutil    17153013\n",
       "            pytz                9045824\n",
       "            six                14512209\n",
       "            tzdata              5214048\n",
       "2024-09-24  numpy              10385658\n",
       "            pandas              8830996\n",
       "            python-dateutil    14066307\n",
       "            pytz                7425446\n",
       "            six                11917222\n",
       "            tzdata              4550626\n",
       "dtype: Int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pandas_pypi = pandas_pypi.assign(date=pandas_pypi[\"timestamp\"].dt.date)\n",
    "downloads_per_day = pandas_pypi.groupby([\"date\", \"project\"]).size()\n",
    "\n",
    "# Cache after the aggregation so that the aggregation only runs once.\n",
    "downloads_per_day.cache()\n",
    "downloads_per_day.to_pandas()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "BigQuery DataFrames has several built-in visualization methods. Alternatively, download the time series with the `to_pandas()` method for further analysis and visualization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Query job 319558aa-e092-4fd0-a8aa-447fca216a57 is DONE. 1.6 kB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:319558aa-e092-4fd0-a8aa-447fca216a57&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Query job e64eb861-bd12-4748-9c29-d97990aa1241 is DONE. 1.2 kB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=swast-scratch&j=bq:US:e64eb861-bd12-4748-9c29-d97990aa1241&page=queryresults\">Open Job</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='date', ylabel='daily downloads'>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "downloads_per_day.unstack().plot.line(rot=45, ylabel=\"daily downloads\", ylim=(0, 2e7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.12.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}