source: tools/perf/process_stat_array.py @ 98233b3

#!/usr/bin/python3

import argparse, os, sys, re

def dir_path(string):
    if os.path.isdir(string):
        return string
    else:
        raise NotADirectoryError(string)

parser = argparse.ArgumentParser()
parser.add_argument('--path', type=dir_path, default=".cfadata", help= 'paste path to biog.txt file')

try :
        args = parser.parse_args()
except NotADirectoryError:
        print("Must use option --path to existing directory or have .cfadata in current directory", file=sys.stderr)
        sys.exit(1)

root, _, filenames = next(os.walk(args.path))

merged = []
counters = {}

#open the files
for filename in filenames:
        try:
                m = re.search('[A-z]+0x([0-9a-f]+)\.data', filename)
                me = m.group(1)
                counters[me] = 0
                with open(os.path.join(root, filename), 'r') as file:
                        for line in file:
                                # data = [int(x.strip()) for x in line.split(',')]
                                data = [int(line.strip())]
                                data = [me, *data]
                                merged.append(data)

        except:
                pass


# Sort by timestamp (the second element)
# take second element for sort
def takeSecond(elem):
    return elem[1]

merged.sort(key=takeSecond)

# for m in merged:
#       print(m)

single = []
curr = 0

# merge the data
# for (me, time, value) in merged:
for (me, value) in merged:
        # check now much this changes
        old = counters[me]
        change = value - old
        counters[me] = value

        # add change to the current
        curr = curr + change
        single.append( value )

        pass

print(single)

# single = sorted(single)[:len(single)-100]
# ms = max(single)
# single = [float(x) / 2500.0 for x in single]

#print
# for t, v in single:
#       print([t, v])
# print(len(single))
# print(max(single))
# print(min(single))

# bins = [0, 5.37751600e+04, 1.06903320e+05, 1.60031480e+05, 2.13159640e+05, 2.66287800e+05, 3.19415960e+05, 3.72544120e+05, 4.25672280e+05, 4.78800440e+05, 5.31928600e+05, 5.85056760e+05, 6.38184920e+05, 6.91313080e+05, 7.44441240e+05, 7.97569400e+05, 8.50697560e+05, 9.03825720e+05, 9.56953880e+05, 1.01008204e+06, 1.06321020e+06, 1.11633836e+06, 1.16946652e+06, 1.22259468e+06, 1.27572284e+06, 1.32885100e+06, 1.38197916e+06, 1.43510732e+06, 1.48823548e+06, 1.54136364e+06, 1.59449180e+06, 1.64761996e+06, 1.70074812e+06, 1.75387628e+06, 1.80700444e+06, 1.86013260e+06, 1.91326076e+06, 1.96638892e+06, 2.01951708e+06, 2.07264524e+06, 2.12577340e+06, 2.17890156e+06, 2.23202972e+06, 2.28515788e+06, 2.33828604e+06, 2.39141420e+06, 2.44454236e+06, 2.49767052e+06, 2.55079868e+06, 2.60392684e+06, 3.0e+06]
# # bins = [float(x) / 2500.0 for x in bins]
# # print([round(b, 2) for b in bins])

# import numpy
# # hist1, _ = numpy.histogram(single, density=True, bins=50)
# hist2, _ = numpy.histogram(single, density=True, bins=bins)
# # print(hist1)
# print([1000.0 * h for h in hist2])
# # for v in single:
# #     print([v])