| 1 | #! /usr/bin/python2.2 |
| 2 | |
| 3 | # Copyright (C) 2002 by Martin Pool <mbp@samba.org> |
| 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 version |
| 7 | # 2 as published by the Free Software Foundation. |
| 8 | # |
| 9 | # This program is distributed in the hope that it will be useful, but |
| 10 | # WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 12 | # Lesser General Public License for more details. |
| 13 | # |
| 14 | # You should have received a copy of the GNU Lesser General Public |
| 15 | # License along with this program; if not, write to the Free Software |
| 16 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
| 17 | |
| 18 | # Populate a tree with pseudo-randomly distributed files to test |
| 19 | # rsync. |
| 20 | |
| 21 | from __future__ import generators |
| 22 | import random, string, os, os.path |
| 23 | |
| 24 | nfiles = 10000 |
| 25 | depth = 5 |
| 26 | n_children = 20 |
| 27 | n_files = 20 |
| 28 | n_symlinks = 10 |
| 29 | |
| 30 | name_chars = string.digits + string.letters |
| 31 | |
| 32 | abuffer = 'a' * 1024 |
| 33 | |
| 34 | def random_name_chars(): |
| 35 | a = "" |
| 36 | for i in range(10): |
| 37 | a = a + random.choice(name_chars) |
| 38 | return a |
| 39 | |
| 40 | |
| 41 | def generate_names(): |
| 42 | n = 0 |
| 43 | while 1: |
| 44 | yield "%05d_%s" % (n, random_name_chars()) |
| 45 | n += 1 |
| 46 | |
| 47 | |
| 48 | class TreeBuilder: |
| 49 | def __init__(self): |
| 50 | self.n_children = 20 |
| 51 | self.n_files = 100 |
| 52 | self.total_entries = 100000 # long(1e8) |
| 53 | self.actual_size = 0 |
| 54 | self.name_gen = generate_names() |
| 55 | self.all_files = [] |
| 56 | self.all_dirs = [] |
| 57 | self.all_symlinks = [] |
| 58 | |
| 59 | |
| 60 | def random_size(self): |
| 61 | return random.lognormvariate(4, 4) |
| 62 | |
| 63 | |
| 64 | def random_symlink_target(self): |
| 65 | what = random.choice(['directory', 'file', 'symlink', 'none']) |
| 66 | try: |
| 67 | if what == 'directory': |
| 68 | return random.choice(self.all_dirs) |
| 69 | elif what == 'file': |
| 70 | return random.choice(self.all_files) |
| 71 | elif what == 'symlink': |
| 72 | return random.choice(self.all_symlinks) |
| 73 | elif what == 'none': |
| 74 | return self.name_gen.next() |
| 75 | except IndexError: |
| 76 | return self.name_gen.next() |
| 77 | |
| 78 | |
| 79 | def can_continue(self): |
| 80 | self.total_entries -= 1 |
| 81 | return self.total_entries > 0 |
| 82 | |
| 83 | |
| 84 | def build_tree(self, prefix, depth): |
| 85 | """Generate a breadth-first tree""" |
| 86 | for count, function in [[n_files, self.make_file], |
| 87 | [n_children, self.make_child_recurse], |
| 88 | [n_symlinks, self.make_symlink]]: |
| 89 | for i in range(count): |
| 90 | if not self.can_continue(): |
| 91 | return |
| 92 | name = os.path.join(prefix, self.name_gen.next()) |
| 93 | function(name, depth) |
| 94 | |
| 95 | |
| 96 | def print_summary(self): |
| 97 | print "total bytes: %d" % self.actual_size |
| 98 | |
| 99 | |
| 100 | def make_child_recurse(self, dname, depth): |
| 101 | if depth > 1: |
| 102 | self.make_dir(dname) |
| 103 | self.build_tree(dname, depth-1) |
| 104 | |
| 105 | |
| 106 | def make_dir(self, dname, depth='ignore'): |
| 107 | print "%s/" % (dname) |
| 108 | os.mkdir(dname) |
| 109 | self.all_dirs.append(dname) |
| 110 | |
| 111 | |
| 112 | def make_symlink(self, lname, depth='ignore'): |
| 113 | print "%s -> %s" % (lname, self.random_symlink_target()) |
| 114 | |
| 115 | |
| 116 | def make_file(self, fname, depth='ignore'): |
| 117 | size = long(self.random_size()) |
| 118 | print "%-70s %d" % (fname, size) |
| 119 | f = open(fname, 'w') |
| 120 | f.truncate(size) |
| 121 | self.fill_file(f, size) |
| 122 | self.all_files.append(fname) |
| 123 | self.actual_size += size |
| 124 | |
| 125 | def fill_file(self, f, size): |
| 126 | while size > 0: |
| 127 | f.write(abuffer[:size]) |
| 128 | size -= len(abuffer) |
| 129 | |
| 130 | |
| 131 | tb = TreeBuilder() |
| 132 | tb.build_tree('/tmp/foo', 3) |
| 133 | tb.print_summary() |