top_2_csv.py

# -*- coding:utf-8 -*- 
#!/usr/bin/python

"""
top_2_csv.py
linux top コマンドのアウトプットをgraph出力用途にcsvに変換する
"""

__author__ = "ken sakurai"
__status__ = "development"
__version__ = "0.0.1"
__date__    = "17 November 2016"

import sys
import re

def create_cpu_summary_csv(filename, lines): 
    
    f = open(filename + '_cpu_summary.csv', 'w')
    header_columns = []
    header_columns.append("時刻")
    header_columns.append("ユーザが利用しているCPU使用率")
    header_columns.append("システムが利用しているCPU使用率")
    header_columns.append("nice値を変更しているプロセスのCPU使用率")
    header_columns.append("待機中のCPU使用率")
    header_columns.append("IO待ちしているCPU使用率")
    header_columns.append("ハードウェア割り込みによるCPU使用率")
    header_columns.append("ソフトウェア割り込みによるCPU使用率")
    header_columns.append("ゲストOSが割り当て要求をしたが、割り当ててもらえなかったCPU使用率")
    f.write(",".join(header_columns) + "\n")
 
    for line in lines:
        
        if line.find("top") == 0:
            columns = []
            words = re.split(" +", line)
            time = words[2]
            columns.append(time)
        elif line.find("Cpu(s):") == 0:
            line = line.replace(",",", ")
            words = re.split(" +", line)
            for word in words[1:]:
                word = word.replace("%us,","").replace("%sy,","").replace("%ni,","")\
                           .replace("%id,","").replace("%wa,","").replace("%hi,","").replace("%si,","").replace("%st","")
                columns.append(word)
            f.write(",".join(columns) + "\n")

def create_memory_summary_csv(filename, lines):
 
     f = open(filename + '_memory_summary.csv', 'w')
     header_columns = []
     header_columns.append("時刻")
     header_columns.append("合計メモリ量")
     header_columns.append("利用メモリ量")
     header_columns.append("空きメモリ量")
     header_columns.append("利用メモリ量のうち、バッファ利用しているメモリ量")
     header_columns.append("合計スワップ量")
     header_columns.append("利用スワップ量")
     header_columns.append("空きスワップ量")
     header_columns.append("キャッシュしているスワップ量")
     f.write(",".join(header_columns) + "\n")
 
     for line in lines:
         if line.find("top") == 0:
             words = re.split(" +", line)
             time = words[2]
             columns = []
             columns.append(time)
         elif line.find("Mem:") == 0:
             words = re.split(" +", line)
             for i, word in enumerate(words):
                 if i % 2 == 1:
                     word = word.replace("k","")
                     columns.append(word)
         elif line.find("Swap:") == 0:
             words = re.split(" +", line)
             for i, word in enumerate(words):
                 if i % 2 == 1:
                     word = word.replace("k","")
                     columns.append(word)
             f.write(",".join(columns) + "\n")

def create_load_average_csv(filename, lines):

     f = open(filename + '_load_average.csv', 'w')
     header_columns = []
     header_columns.append("時刻")
     header_columns.append("ログインユーザ数")
     header_columns.append("ロードアベレージ過去1分")
     header_columns.append("ロードアベレージ過去5分")
     header_columns.append("ロードアベレージ過去15分")
     f.write(",".join(header_columns) + "\n")

     for line in lines:
         if line.find("top") == 0:
             words = re.split(" +", line)
             columns = []
             for i, word in enumerate(words): 
                 if i == 2 or i == 7 or i == 11 or i == 12 or i == 13:
                     columns.append(word.replace(",",""))
             f.write(",".join(columns) + "\n")

def create_task_summary_csv(filename, lines):
     
    f = open(filename + '_task_summary.csv', 'w')
    header_columns = []
    header_columns.append("時刻")
    header_columns.append("タスクの合計数")
    header_columns.append("稼働中のタスク数")
    header_columns.append("待機中のタスク数")
    header_columns.append("停止タスク数")
    header_columns.append("ゾンビタスク数")
    f.write(",".join(header_columns) + "\n")
    
    for line in lines:
        if line.find("top") == 0:
             words = re.split(" +", line)
             time = words[2]
             columns = []
             columns.append(time) 
        elif line.find("Tasks:") == 0:
             words = re.split(" +", line)
             for i, word in enumerate(words):
                 if i % 2 == 1:
                     columns.append(word)
             f.write(",".join(columns) + "\n")

def create_procees_id_csv(filename, lines, pid):
    
    f = open(filename + '_' + str(pid) + '.csv', 'w')
    header_columns = []
    header_columns.append("時刻")
    header_columns.append("PID")
    header_columns.append("USER")
    header_columns.append("優先度")
    header_columns.append("相対優先度")
    header_columns.append("仮想メモリ")
    header_columns.append("物理メモリ")
    header_columns.append("共有メモリ")
    header_columns.append("状態")
    header_columns.append("CPU使用率")
    header_columns.append("メモリ使用率")
    header_columns.append("実行時間")
    header_columns.append("コマンド")
    f.write(",".join(header_columns) + "\n")
    
    for line in lines:
        if line.find("top") == 0:
            columns = []
            words = re.split(" +", line)
            time = words[2]
            columns.append(time) 
        else:
            replaced_line = re.sub("^ +","",line)
            if replaced_line.find(pid) == 0:
                words = re.split(" +", replaced_line)   
                columns.extend(words)
                f.write(",".join(columns) + "\n") 

def print_usage():
    print "-----------------------------------------------"
    print "Usage: python top_2csv.py {1} {2}"
    print "arguments:"
    print "{1}: top result file name"
    print "{2}: mode 'SUMMARY' or 'PID (digit like 1234)'"
    print "-----------------------------------------------"

if __name__ == '__main__':
    filename = sys.argv[1]
    try:
       mode = sys.argv[2]
    except IndexError:
       print_usage()
       raise 
    
    lines = [line.rstrip('\n') for line in open(filename)]

    if mode == "SUMMARY":
        create_cpu_summary_csv(filename, lines)
        create_memory_summary_csv(filename, lines)
        create_load_average_csv(filename, lines)
        create_task_summary_csv(filename, lines)       

    elif mode.isdigit():
        create_procees_id_csv(filename, lines, mode)
    
    else:
        print_usage()
        raise