From 22776c4d0fa64148172eebea04b7045474304cbe Mon Sep 17 00:00:00 2001
From: John-Paul Robinson <jpr@uab.edu>
Date: Fri, 26 Jul 2024 11:48:39 -0500
Subject: [PATCH] Notebook to convert policy run output to parquet data sets
This is intended to be run on URL encoded output lines from a
gpfs list policy run. It creates panda structures that are
then saved as parquet format for ease of downstream processing.
Can be run in parallel across many inputs by wrapping with papermill
and have upstream split the input file.
---
parquet-list-policy-data.ipynb | 246 +++++++++++++++++++++++++++++++++
1 file changed, 246 insertions(+)
create mode 100644 parquet-list-policy-data.ipynb
diff --git a/parquet-list-policy-data.ipynb b/parquet-list-policy-data.ipynb
new file mode 100644
index 0000000..72e5963
--- /dev/null
+++ b/parquet-list-policy-data.ipynb
@@ -0,0 +1,246 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "073ef418",
+ "metadata": {},
+ "source": [
+ "# Convert raw policy lists into parquet\n",
+ "\n",
+ "Having the raw list-policy output data converted to parquet binary reduces storage space, centralized and speeds later parallel processing and reporting via dask.\n",
+ "\n",
+ "The script reads files that match the `glob_pattern` from the provided `dirname` and writes identical file names in pickled format to the `parquet`, optionally filtering lines by the `line_regex_filter`. If the default parameters aren't changed no files are read or written.\n",
+ "\n",
+ "Some parsing progress is available via the `verbose` flag.\n",
+ "\n",
+ "This converter assumes a policy show format defined in the [list-paths-external policy](https://code.rc.uab.edu/rc/gpfs-policy/-/blob/main/policy/list-path-external):\n",
+ "```\n",
+ " SHOW ('|size=' || varchar(FILE_SIZE) ||\n",
+ " '|kballoc='|| varchar(KB_ALLOCATED) ||\n",
+ " '|access=' || varchar(ACCESS_TIME) ||\n",
+ " '|create=' || varchar(CREATION_TIME) ||\n",
+ " '|modify=' || varchar(MODIFICATION_TIME) ||\n",
+ " '|uid=' || varchar(USER_ID) ||\n",
+ " '|gid=' || varchar(GROUP_ID) ||\n",
+ " '|heat=' || varchar(FILE_HEAT) ||\n",
+ " '|pool=' || varchar(POOL_NAME) ||\n",
+ " '|mode=' || varchar(MODE) ||\n",
+ " '|misc=' || varchar(MISC_ATTRIBUTES) ||\n",
+ " '|'\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "af015950",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import datetime\n",
+ "import pandas as pd\n",
+ "import matplotlib.pyplot as plt\n",
+ "from urllib.parse import unquote\n",
+ "import sys\n",
+ "import os\n",
+ "import pathlib\n",
+ "import re"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3781a0d6",
+ "metadata": {},
+ "source": [
+ "## input vars"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "932707e6",
+ "metadata": {
+ "tags": [
+ "parameters"
+ ]
+ },
+ "outputs": [],
+ "source": [
+ "dirname=\"data/list-20191520.list.gather-info.d\" # directory to fine files to pickle\n",
+ "glob_pattern = \"*.gz\" # file name glob pattern to match, can be file name for individual file\n",
+ "line_regex_filter = \".*\" # regex to match lines of interest in file\n",
+ "\n",
+ "verbose = True"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "833be559",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pickledir=f\"{dirname}/parquet\""
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "47ea1d93",
+ "metadata": {},
+ "source": [
+ "dirname=\"data/list-17404604.list.gather-info.d/\" # directory to fine files to pickle\n",
+ "glob_pattern = \"list-*.gz\" # file name glob pattern to match, can be file name for individual file\n",
+ "line_regex_filter = \".*\" # regex to match lines of interest in file\n",
+ "pickledir=f\"{dirname}/pickles\"\n",
+ "\n",
+ "verbose = True"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "07ef745a",
+ "metadata": {},
+ "source": [
+ "dirname=\"data/list-16144464.list.gather-info.d/\" # directory to fine files to pickle\n",
+ "glob_pattern = \"list-*\" # file name glob pattern to match, can be file name for individual file\n",
+ "line_regex_filter = \".*\" # regex to match lines of interest in file\n",
+ "pickledir=f\"{dirname}/pickles\"\n",
+ "\n",
+ "verbose = True"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "5599e260",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# parse files with read_csv optionally filtering specific lines via regex\n",
+ "\n",
+ "def parse_file(filename, pattern=\".*\"):\n",
+ " \n",
+ " gen = pd.read_csv(filename, sep='\\n', header=None, iterator=True)\n",
+ " df = pd.concat((x[x[0].str.contains(pattern, regex=True)] for x in gen), ignore_index=True)\n",
+ "\n",
+ " return df"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "6542cb23",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# parse rows according to the list-policy-external format\n",
+ "\n",
+ "def parse_rows(df):\n",
+ " # split content on white space\n",
+ " df=df.rename(columns={0:\"details\"})\n",
+ " new=df[\"details\"].str.split(expand=True)\n",
+ " \n",
+ " # create a new dataframe and populate with parsed data\n",
+ " df = pd.DataFrame()\n",
+ "\n",
+ " # split attribuignoring filename= prefix\n",
+ " df[\"showattr\"] = new[3].map(lambda x: re.sub(\"\\w+=\", \"\", unquote(x)))\n",
+ " df[[\"ignore1\", \"size\", \"kballoc\", \"access\", \"create\", \"modify\", \n",
+ " \"uid\", \"gid\", \"heat\", \"pool\", \"mode\", \"misc\", \"ignore2\"]] = df[\"showattr\"].str.split(\"|\", expand=True)\n",
+ " df[\"path\"] = new[5].map(lambda x: unquote(x))\n",
+ "\n",
+ " # drop temp columns\n",
+ " df = df.drop([\"showattr\", \"ignore1\", \"ignore2\"], axis=1)\n",
+ "\n",
+ " df.reset_index(drop=True, inplace=True)\n",
+ "\n",
+ " df = set_types(df)\n",
+ "\n",
+ " return df"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "9730f207",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# convert data to native pandas types\n",
+ "def set_types(df):\n",
+ " df[\"size\"] = df[\"size\"].astype('int64')\n",
+ " df[\"kballoc\"] = df[\"kballoc\"].astype('int64')\n",
+ " df[\"uid\"] = df[\"uid\"].astype('int64')\n",
+ " df[\"gid\"] = df[\"gid\"].astype('int64')\n",
+ " df[\"access\"] = df[\"access\"].astype('datetime64')\n",
+ " df[\"create\"] = df[\"create\"].astype('datetime64')\n",
+ " df[\"modify\"] = df[\"modify\"].astype('datetime64')\n",
+ " \n",
+ " return df"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "2ed6bdc8",
+ "metadata": {},
+ "source": [
+ "## Gather the files according to glob_pattern"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "7297f0d2",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "dirpath = pathlib.Path(dirname)\n",
+ "\n",
+ "files = list()\n",
+ "for file in list(dirpath.glob(glob_pattern)):\n",
+ " files.append(str(file))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e4929a0f",
+ "metadata": {},
+ "source": [
+ "## Read, parse and pickle files"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "2ab7f7f5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "for file in files:\n",
+ " if (verbose): print(f\"parse: {file}\")\n",
+ " filename=os.path.basename(file)\n",
+ " df = parse_rows(parse_file(file))\n",
+ " \n",
+ " # rename for parquet (drop the read glob_pattern extension)\n",
+ " filename, _ = filename.split(\".\", 1)\n",
+ "\n",
+ " ## Write the pickled data\n",
+ "\n",
+ " # only create dir if there is data to pickle\n",
+ " if (not os.path.isdir(pickledir)):\n",
+ " os.mkdir(pickledir)\n",
+ "\n",
+ " if (verbose): print(f\"pickling: {filename}\")\n",
+ " df.to_parquet(f\"{pickledir}/{filename}.parquet\", engine=\"pyarrow\")"
+ ]
+ }
+ ],
+ "metadata": {
+ "language_info": {
+ "name": "python",
+ "pygments_lexer": "ipython3"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
--
GitLab