support/scripts/size-stats

   1 #!/usr/bin/env python
   2
   3 # Copyright (C) 2014 by Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
   4
   5 # This program is free software; you can redistribute it and/or modify
   6 # it under the terms of the GNU General Public License as published by
   7 # the Free Software Foundation; either version 2 of the License, or
   8 # (at your option) any later version.
   9 #
  10 # This program is distributed in the hope that it will be useful,
  11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  13 # General Public License for more details.
  14 #
  15 # You should have received a copy of the GNU General Public License
  16 # along with this program; if not, write to the Free Software
  17 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
  18
  19 import sys
  20 import os
  21 import os.path
  22 import argparse
  23 import csv
  24 import collections
  25
  26 try:
  27     import matplotlib
  28     matplotlib.use('Agg')
  29     import matplotlib.font_manager as fm
  30     import matplotlib.pyplot as plt
  31 except ImportError:
  32     sys.stderr.write("You need python-matplotlib to generate the size graph\n")
  33     exit(1)
  34
  35 colors = ['#e60004', '#009836', '#2e1d86', '#ffed00',
  36           '#0068b5', '#f28e00', '#940084', '#97c000']
  37
  38 #
  39 # This function adds a new file to 'filesdict', after checking its
  40 # size. The 'filesdict' contain the relative path of the file as the
  41 # key, and as the value a tuple containing the name of the package to
  42 # which the file belongs and the size of the file.
  43 #
  44 # filesdict: the dict to which  the file is added
  45 # relpath: relative path of the file
  46 # fullpath: absolute path to the file
  47 # pkg: package to which the file belongs
  48 #
  49 def add_file(filesdict, relpath, abspath, pkg):
  50     if not os.path.exists(abspath):
  51         return
  52     if os.path.islink(abspath):
  53         return
  54     sz = os.stat(abspath).st_size
  55     filesdict[relpath] = (pkg, sz)
  56
  57 #
  58 # This function returns a dict where each key is the path of a file in
  59 # the root filesystem, and the value is a tuple containing two
  60 # elements: the name of the package to which this file belongs and the
  61 # size of the file.
  62 #
  63 # builddir: path to the Buildroot output directory
  64 #
  65 def build_package_dict(builddir):
  66     filesdict = {}
  67     with open(os.path.join(builddir, "build", "packages-file-list.txt")) as filelistf:
  68         for l in filelistf.readlines():
  69             pkg, fpath = l.split(",", 1)
  70             # remove the initial './' in each file path
  71             fpath = fpath.strip()[2:]
  72             fullpath = os.path.join(builddir, "target", fpath)
  73             add_file(filesdict, fpath, fullpath, pkg)
  74     return filesdict
  75
  76 #
  77 # This function builds a dictionary that contains the name of a
  78 # package as key, and the size of the files installed by this package
  79 # as the value.
  80 #
  81 # filesdict: dictionary with the name of the files as key, and as
  82 # value a tuple containing the name of the package to which the files
  83 # belongs, and the size of the file. As returned by
  84 # build_package_dict.
  85 #
  86 # builddir: path to the Buildroot output directory
  87 #
  88 def build_package_size(filesdict, builddir):
  89     pkgsize = collections.defaultdict(int)
  90
  91     for root, _, files in os.walk(os.path.join(builddir, "target")):
  92         for f in files:
  93             fpath = os.path.join(root, f)
  94             if os.path.islink(fpath):
  95                 continue
  96             frelpath = os.path.relpath(fpath, os.path.join(builddir, "target"))
  97             if not frelpath in filesdict:
  98                 print("WARNING: %s is not part of any package" % frelpath)
  99                 pkg = "unknown"
 100             else:
 101                 pkg = filesdict[frelpath][0]
 102
 103             pkgsize[pkg] += os.path.getsize(fpath)
 104
 105     return pkgsize
 106
 107 #
 108 # Given a dict returned by build_package_size(), this function
 109 # generates a pie chart of the size installed by each package.
 110 #
 111 # pkgsize: dictionary with the name of the package as a key, and the
 112 # size as the value, as returned by build_package_size.
 113 #
 114 # outputf: output file for the graph
 115 #
 116 def draw_graph(pkgsize, outputf):
 117     total = sum(pkgsize.values())
 118     labels = []
 119     values = []
 120     other_value = 0
 121     for (p, sz) in pkgsize.items():
 122         if sz < (total * 0.01):
 123             other_value += sz
 124         else:
 125             labels.append("%s (%d kB)" % (p, sz / 1000.))
 126             values.append(sz)
 127     labels.append("Other (%d kB)" % (other_value / 1000.))
 128     values.append(other_value)
 129
 130     plt.figure()
 131     patches, texts, autotexts = plt.pie(values, labels=labels,
 132                                         autopct='%1.1f%%', shadow=True,
 133                                         colors=colors)
 134     # Reduce text size
 135     proptease = fm.FontProperties()
 136     proptease.set_size('xx-small')
 137     plt.setp(autotexts, fontproperties=proptease)
 138     plt.setp(texts, fontproperties=proptease)
 139
 140     plt.suptitle("Filesystem size per package", fontsize=18, y=.97)
 141     plt.title("Total filesystem size: %d kB" % (total / 1000.), fontsize=10, y=.96)
 142     plt.savefig(outputf)
 143
 144 #
 145 # Generate a CSV file with statistics about the size of each file, its
 146 # size contribution to the package and to the overall system.
 147 #
 148 # filesdict: dictionary with the name of the files as key, and as
 149 # value a tuple containing the name of the package to which the files
 150 # belongs, and the size of the file. As returned by
 151 # build_package_dict.
 152 #
 153 # pkgsize: dictionary with the name of the package as a key, and the
 154 # size as the value, as returned by build_package_size.
 155 #
 156 # outputf: output CSV file
 157 #
 158 def gen_files_csv(filesdict, pkgsizes, outputf):
 159     total = 0
 160     for (p, sz) in pkgsizes.items():
 161         total += sz
 162     with open(outputf, 'w') as csvfile:
 163         wr = csv.writer(csvfile, delimiter=',', quoting=csv.QUOTE_MINIMAL)
 164         wr.writerow(["File name",
 165                      "Package name",
 166                      "File size",
 167                      "Package size",
 168                      "File size in package (%)",
 169                      "File size in system (%)"])
 170         for f, (pkgname, filesize) in filesdict.items():
 171             pkgsize = pkgsizes[pkgname]
 172             wr.writerow([f, pkgname, filesize, pkgsize,
 173                          "%.1f" % (float(filesize) / pkgsize * 100),
 174                          "%.1f" % (float(filesize) / total * 100)])
 175
 176
 177 #
 178 # Generate a CSV file with statistics about the size of each package,
 179 # and their size contribution to the overall system.
 180 #
 181 # pkgsize: dictionary with the name of the package as a key, and the
 182 # size as the value, as returned by build_package_size.
 183 #
 184 # outputf: output CSV file
 185 #
 186 def gen_packages_csv(pkgsizes, outputf):
 187     total = sum(pkgsizes.values())
 188     with open(outputf, 'w') as csvfile:
 189         wr = csv.writer(csvfile, delimiter=',', quoting=csv.QUOTE_MINIMAL)
 190         wr.writerow(["Package name", "Package size", "Package size in system (%)"])
 191         for (pkg, size) in pkgsizes.items():
 192             wr.writerow([pkg, size, "%.1f" % (float(size) / total * 100)])
 193
 194 parser = argparse.ArgumentParser(description='Draw size statistics graphs')
 195
 196 parser.add_argument("--builddir", '-i', metavar="BUILDDIR", required=True,
 197                     help="Buildroot output directory")
 198 parser.add_argument("--graph", '-g', metavar="GRAPH",
 199                     help="Graph output file (.pdf or .png extension)")
 200 parser.add_argument("--file-size-csv", '-f', metavar="FILE_SIZE_CSV",
 201                     help="CSV output file with file size statistics")
 202 parser.add_argument("--package-size-csv", '-p', metavar="PKG_SIZE_CSV",
 203                     help="CSV output file with package size statistics")
 204 args = parser.parse_args()
 205
 206 # Find out which package installed what files
 207 pkgdict = build_package_dict(args.builddir)
 208
 209 # Collect the size installed by each package
 210 pkgsize = build_package_size(pkgdict, args.builddir)
 211
 212 if args.graph:
 213     draw_graph(pkgsize, args.graph)
 214 if args.file_size_csv:
 215     gen_files_csv(pkgdict, pkgsize, args.file_size_csv)
 216 if args.package_size_csv:
 217     gen_packages_csv(pkgsize, args.package_size_csv)