1  #!/usr/bin/env python 
  2   
  3   
  4  import os 
  5  import re 
  6  import timeseries 
  7  import corestats 
  8   
  9   
 10   
 11 -class LogParser: 
 12           
 13 -    def __init__(self, input_type, query):    
 14          self.commandline_opts = '-i:' + input_type + ' -o:CSV ' + '"' + query + '"' 
 15          self.log_parser = 'LogParser.exe'  # make sure its on your Path or else put the full Path here 
 16          self.data_set = [] 
 17          self.headers = '' 
 18          self.footers = '' 
 19           
 20          # load the object with data from Log Parser query results 
 21          self.query() 
 22           
 23      
 24 -    def mslogparser_wrapper(self): 
 25          # wrapper for the Log Parser CLI 
 26          # execute the command as a subprocess and return file objects (child_stdin, child_stdout_and_stderr) 
 27          cmd = os.popen4(self.log_parser + ' ' + self.commandline_opts) 
 28          # read contents of the file object (child_stdout_and_stderr) until EOF 
 29          cmd_output = cmd[1].read() 
 30          # close the handles 
 31          for fd in cmd: fd.close() 
 32          return cmd_output 
 33       
 34       
 35 -    def query(self): 
 36          # we call the Log Parser wrapper to execute the query.   
 37          # rows returned from the query are split into csv's and stored in a list 
 38          data_set = self.mslogparser_wrapper().splitlines() 
 39          # store the output column headings 
 40          self.headers = data_set[:1] 
 41          # store the output footers (statistics) 
 42          self.footers = data_set[-6:-1][2:] 
 43          # trim the header and footers from the list so all we have left are rows of csv data 
 44          # store the list in the object's data_set instance var 
 45          self.data_set = data_set[2:-7] 
 46       
 47           
 48 -    def print_dataset(self): 
 49          # print the data_set (newline delimited) 
 50          for row in self.data_set: 
 51              print row 
 52           
 53       
 54 -    def dataset(self): 
 55          return data_set 
 56       
 57       
 58 -    def dataset_row(self, line_number): 
 59          # get a single line from our data_set (zero indexed) 
 60          return self.data_set[line_number] 
 61           
 62       
 63 -    def dataset_column(self, column_number): 
 64          # get a single column from our data_set (zero indexed) 
 65          data_column = [re.split(',', row)[column_number] for row in self.data_set] 
 66          return data_column 
 67   
 68       
 69 -    def sum(self, column_number): 
 70          col = self.dataset_column(column_number) 
 71          stats = corestats.Stats(col) 
 72          return stats.sum() 
 73        
 74   
 75 -    def count(self, column_number): 
 76          col = self.dataset_column(column_number) 
 77          stats = corestats.Stats(col) 
 78          return stats.count() 
 79   
 80   
 81 -    def min(self, column_number): 
 82          col = self.dataset_column(column_number) 
 83          stats = corestats.Stats(col) 
 84          return stats.min() 
 85           
 86       
 87 -    def max(self, column_number): 
 88          col = self.dataset_column(column_number) 
 89          stats = corestats.Stats(col) 
 90          return stats.max() 
 91           
 92           
 93 -    def avg(self, column_number): 
 94          col = self.dataset_column(column_number) 
 95          stats = corestats.Stats(col) 
 96          return stats.avg() 
 97           
 98           
 99 -    def median(self, column_number): 
100          col = self.dataset_column(column_number) 
101          stats = corestats.Stats(col) 
102          return stats.median() 
103   
104   
105 -    def stdev(self, column_number): 
106          col = self.dataset_column(column_number) 
107          stats = corestats.Stats(col) 
108          return stats.stdev() 
109           
110           
111 -    def percentile(self, percentile, column_number): 
112          col = self.dataset_column(column_number) 
113          stats = corestats.Stats(col) 
114          return stats.percentile(percentile) 
115       
116       
117 -    def timeseries(self, interval): 
118          return timeseries.TimeSeries(interval, self.data_set)