line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Linux::NFS::BigDir; |
2
|
2
|
|
|
2
|
|
32632
|
use strict; |
|
2
|
|
|
|
|
5
|
|
|
2
|
|
|
|
|
48
|
|
3
|
2
|
|
|
2
|
|
9
|
use warnings; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
48
|
|
4
|
2
|
|
|
2
|
|
11
|
use Exporter 'import'; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
44
|
|
5
|
2
|
|
|
2
|
|
9
|
use Carp; |
|
2
|
|
|
|
|
3
|
|
|
2
|
|
|
|
|
87
|
|
6
|
2
|
|
|
2
|
|
10
|
use Fcntl; |
|
2
|
|
|
|
|
2
|
|
|
2
|
|
|
|
|
351
|
|
7
|
2
|
|
|
2
|
|
11
|
use Config; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
62
|
|
8
|
|
|
|
|
|
|
|
9
|
2
|
|
|
2
|
|
10
|
use constant BUF_SIZE => 4096; |
|
2
|
|
|
|
|
2
|
|
|
2
|
|
|
|
|
122
|
|
10
|
|
|
|
|
|
|
use constant SYS_getdents => do { |
11
|
2
|
|
|
2
|
|
948
|
use Inline 0.80 C => <<'...'; |
|
2
|
|
|
|
|
25886
|
|
|
2
|
|
|
|
|
14
|
|
12
|
|
|
|
|
|
|
#include |
13
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
int _get_syscall_num() { |
15
|
|
|
|
|
|
|
return SYS_getdents; |
16
|
|
|
|
|
|
|
} |
17
|
|
|
|
|
|
|
... |
18
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
_get_syscall_num(); |
20
|
|
|
|
|
|
|
}; |
21
|
|
|
|
|
|
|
|
22
|
|
|
|
|
|
|
our $VERSION = '0.002'; # VERSION |
23
|
|
|
|
|
|
|
|
24
|
|
|
|
|
|
|
=pod |
25
|
|
|
|
|
|
|
|
26
|
|
|
|
|
|
|
=head1 NAME |
27
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
Linux::NFS::BigDir - use Linux getdents syscall to read large directories over NFS |
29
|
|
|
|
|
|
|
|
30
|
|
|
|
|
|
|
=head1 SYNOPSIS |
31
|
|
|
|
|
|
|
|
32
|
|
|
|
|
|
|
use Linux::NFS::BigDir qw(getdents); |
33
|
|
|
|
|
|
|
# entries_ref is an array reference |
34
|
|
|
|
|
|
|
my $entries_ref = getdents($very_large_dir); |
35
|
|
|
|
|
|
|
|
36
|
|
|
|
|
|
|
=head1 DESCRIPTION |
37
|
|
|
|
|
|
|
|
38
|
|
|
|
|
|
|
This module was created to solve a very specific problem: you have a directory over NFS, mounted by |
39
|
|
|
|
|
|
|
a Linux OS, and that directory has a very large number of items (files, directories, etc). The number of entries |
40
|
|
|
|
|
|
|
is so large that you have trouble to list the contents with C or even C from the shell. In extreme |
41
|
|
|
|
|
|
|
cases, the operation just "hangs" and will provide a feedback hours later. |
42
|
|
|
|
|
|
|
|
43
|
|
|
|
|
|
|
I observed this behavior only with NFS version 3 (and wasn't able to simulate it with local EXT3/EXT4): you might find in different situations, |
44
|
|
|
|
|
|
|
but in that case it migh be a wrong configuration regarding the filesystem. Ask your administrator first. |
45
|
|
|
|
|
|
|
|
46
|
|
|
|
|
|
|
If you can't fix (or get fixed) the problem, then you might want to try to use this module. It will use the C |
47
|
|
|
|
|
|
|
syscall from Linux. You can check the documentation about this syscall with C in a shell. |
48
|
|
|
|
|
|
|
|
49
|
|
|
|
|
|
|
In short, this syscall will return a data structure, but you probably will want to use only the name of each entry in the directory. |
50
|
|
|
|
|
|
|
|
51
|
|
|
|
|
|
|
How can this be useful? Here are some directions: |
52
|
|
|
|
|
|
|
|
53
|
|
|
|
|
|
|
=over |
54
|
|
|
|
|
|
|
|
55
|
|
|
|
|
|
|
=item 1. |
56
|
|
|
|
|
|
|
|
57
|
|
|
|
|
|
|
You want to remove all directory content. |
58
|
|
|
|
|
|
|
|
59
|
|
|
|
|
|
|
=item 2. |
60
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
You want to remove files from the directory with a pattern in their filename (using regular expressions, for example). |
62
|
|
|
|
|
|
|
|
63
|
|
|
|
|
|
|
=item 3. |
64
|
|
|
|
|
|
|
|
65
|
|
|
|
|
|
|
You want to select specific files by their filenames and then test something else (like atime). |
66
|
|
|
|
|
|
|
|
67
|
|
|
|
|
|
|
=back |
68
|
|
|
|
|
|
|
|
69
|
|
|
|
|
|
|
These are examples, but it should cover the vast majority of what you want to do. C syscall will be more effective because |
70
|
|
|
|
|
|
|
it will not call C of each of those files before returning the information to you. That means, you will have the opportunity to filter |
71
|
|
|
|
|
|
|
whatever you need and then call C if you really need. |
72
|
|
|
|
|
|
|
|
73
|
|
|
|
|
|
|
I came up at C after researching about "how to remove million of files". After a while I reached an C program example that uses C |
74
|
|
|
|
|
|
|
to print the filenames under the directory. By using it, I was able to cleanup directories with thousands (or even millions) of files in a couple of minutes, |
75
|
|
|
|
|
|
|
instead of many hours. |
76
|
|
|
|
|
|
|
|
77
|
|
|
|
|
|
|
This module is a Perl implementation of that. |
78
|
|
|
|
|
|
|
|
79
|
|
|
|
|
|
|
=head1 FUNCTIONS |
80
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
The sub C and C are exported on demand. |
82
|
|
|
|
|
|
|
|
83
|
|
|
|
|
|
|
=cut |
84
|
|
|
|
|
|
|
|
85
|
|
|
|
|
|
|
our @EXPORT_OK = qw(getdents getdents_safe); |
86
|
|
|
|
|
|
|
|
87
|
|
|
|
|
|
|
=head2 getdents |
88
|
|
|
|
|
|
|
|
89
|
|
|
|
|
|
|
Expects the complete path to the directory as a parameter. |
90
|
|
|
|
|
|
|
|
91
|
|
|
|
|
|
|
Returns an array reference with all files inside that directory but the 'dot' files. |
92
|
|
|
|
|
|
|
|
93
|
|
|
|
|
|
|
Meanwhile simple (and probably faster), you should be careful regarding memory restrictions when using this functions. |
94
|
|
|
|
|
|
|
|
95
|
|
|
|
|
|
|
If you have too many files, you program may try to allocate too much memory, with all the undesired effects. See C. |
96
|
|
|
|
|
|
|
|
97
|
|
|
|
|
|
|
=cut |
98
|
|
|
|
|
|
|
|
99
|
|
|
|
|
|
|
sub getdents { |
100
|
|
|
|
|
|
|
my $dir = shift; |
101
|
|
|
|
|
|
|
confess "directory $dir is not available" unless ( -d $dir ); |
102
|
|
|
|
|
|
|
sysopen( my $fd, $dir, O_RDONLY | O_DIRECTORY ); |
103
|
|
|
|
|
|
|
my @items; |
104
|
|
|
|
|
|
|
|
105
|
|
|
|
|
|
|
while (1) { |
106
|
|
|
|
|
|
|
my $buf = "\0" x BUF_SIZE; |
107
|
|
|
|
|
|
|
my $read = syscall( SYS_getdents, fileno($fd), $buf, BUF_SIZE ); |
108
|
|
|
|
|
|
|
|
109
|
|
|
|
|
|
|
if ( ( $read == -1 ) and ( $! != 0 ) ) { |
110
|
|
|
|
|
|
|
confess "failed to syscall getdents: $!"; |
111
|
|
|
|
|
|
|
} |
112
|
|
|
|
|
|
|
|
113
|
|
|
|
|
|
|
last if ( $read == 0 ); |
114
|
|
|
|
|
|
|
|
115
|
|
|
|
|
|
|
while ( $read != 0 ) { |
116
|
|
|
|
|
|
|
my ( $ino, $off, $len, $name ) = unpack( "L!L!SZ*", $buf ); |
117
|
|
|
|
|
|
|
push( @items, $name ); |
118
|
|
|
|
|
|
|
substr( $buf, 0, $len ) = ''; |
119
|
|
|
|
|
|
|
$read -= $len; |
120
|
|
|
|
|
|
|
} |
121
|
|
|
|
|
|
|
|
122
|
|
|
|
|
|
|
} |
123
|
|
|
|
|
|
|
|
124
|
|
|
|
|
|
|
# removing '.' and '..' |
125
|
|
|
|
|
|
|
shift(@items); |
126
|
|
|
|
|
|
|
shift(@items); |
127
|
|
|
|
|
|
|
return \@items; |
128
|
|
|
|
|
|
|
} |
129
|
|
|
|
|
|
|
|
130
|
|
|
|
|
|
|
=head2 getdents_safe |
131
|
|
|
|
|
|
|
|
132
|
|
|
|
|
|
|
"Safe" version of C because it will write each entry read to a text file instead of storing |
133
|
|
|
|
|
|
|
all the entries on memory. |
134
|
|
|
|
|
|
|
|
135
|
|
|
|
|
|
|
Expects as parameters: |
136
|
|
|
|
|
|
|
|
137
|
|
|
|
|
|
|
=over |
138
|
|
|
|
|
|
|
|
139
|
|
|
|
|
|
|
=item * |
140
|
|
|
|
|
|
|
|
141
|
|
|
|
|
|
|
The complete path to the directory to be read. |
142
|
|
|
|
|
|
|
|
143
|
|
|
|
|
|
|
=item * |
144
|
|
|
|
|
|
|
|
145
|
|
|
|
|
|
|
The complete path to the file that will be used to print each entry, one per line. As convenience, all filenames will be |
146
|
|
|
|
|
|
|
prepended with the complete path to the directory given as parameter. |
147
|
|
|
|
|
|
|
|
148
|
|
|
|
|
|
|
=back |
149
|
|
|
|
|
|
|
|
150
|
|
|
|
|
|
|
The filename given will be created. If it already exists, this function will C. |
151
|
|
|
|
|
|
|
|
152
|
|
|
|
|
|
|
This function returns the number of files read from the given directory. |
153
|
|
|
|
|
|
|
|
154
|
|
|
|
|
|
|
=cut |
155
|
|
|
|
|
|
|
|
156
|
|
|
|
|
|
|
sub getdents_safe { |
157
|
|
|
|
|
|
|
my ( $dir, $output ) = @_; |
158
|
|
|
|
|
|
|
confess "directory $dir is not available" unless ( -d $dir ); |
159
|
|
|
|
|
|
|
sysopen( my $fd, $dir, O_RDONLY | O_DIRECTORY ); |
160
|
|
|
|
|
|
|
sysopen( my $out, $output, O_CREAT | O_RDWR | O_EXCL ) |
161
|
|
|
|
|
|
|
or die "Cannot create $output: $!"; |
162
|
|
|
|
|
|
|
my $dots = 0; |
163
|
|
|
|
|
|
|
my $counter = 0; |
164
|
|
|
|
|
|
|
|
165
|
|
|
|
|
|
|
while (1) { |
166
|
|
|
|
|
|
|
my $buf = "\0" x BUF_SIZE; |
167
|
|
|
|
|
|
|
my $read = syscall( SYS_getdents, fileno($fd), $buf, BUF_SIZE ); |
168
|
|
|
|
|
|
|
|
169
|
|
|
|
|
|
|
if ( ( $read == -1 ) and ( $! != 0 ) ) { |
170
|
|
|
|
|
|
|
confess "failed to syscall getdents: $!"; |
171
|
|
|
|
|
|
|
} |
172
|
|
|
|
|
|
|
|
173
|
|
|
|
|
|
|
last if ( $read == 0 ); |
174
|
|
|
|
|
|
|
|
175
|
|
|
|
|
|
|
if ( $dots == 2 ) { |
176
|
|
|
|
|
|
|
|
177
|
|
|
|
|
|
|
while ( $read != 0 ) { |
178
|
|
|
|
|
|
|
my ( $ino, $off, $len, $name ) = unpack( "L!L!SZ*", $buf ); |
179
|
|
|
|
|
|
|
print $out $dir, '/', $name, "\n"; |
180
|
|
|
|
|
|
|
$counter++; |
181
|
|
|
|
|
|
|
substr( $buf, 0, $len ) = ''; |
182
|
|
|
|
|
|
|
$read -= $len; |
183
|
|
|
|
|
|
|
} |
184
|
|
|
|
|
|
|
|
185
|
|
|
|
|
|
|
} |
186
|
|
|
|
|
|
|
else { |
187
|
|
|
|
|
|
|
|
188
|
|
|
|
|
|
|
while ( $read != 0 ) { |
189
|
|
|
|
|
|
|
my ( $ino, $off, $len, $name ) = unpack( "L!L!SZ*", $buf ); |
190
|
|
|
|
|
|
|
|
191
|
|
|
|
|
|
|
unless ( ( $name eq '.' ) or ( $name eq '..' ) ) { |
192
|
|
|
|
|
|
|
print $out $dir, '/', $name, "\n"; |
193
|
|
|
|
|
|
|
$counter++; |
194
|
|
|
|
|
|
|
} |
195
|
|
|
|
|
|
|
else { |
196
|
|
|
|
|
|
|
$dots++; |
197
|
|
|
|
|
|
|
} |
198
|
|
|
|
|
|
|
|
199
|
|
|
|
|
|
|
substr( $buf, 0, $len ) = ''; |
200
|
|
|
|
|
|
|
$read -= $len; |
201
|
|
|
|
|
|
|
} |
202
|
|
|
|
|
|
|
|
203
|
|
|
|
|
|
|
} |
204
|
|
|
|
|
|
|
|
205
|
|
|
|
|
|
|
} |
206
|
|
|
|
|
|
|
|
207
|
|
|
|
|
|
|
close($out); |
208
|
|
|
|
|
|
|
return $counter; |
209
|
|
|
|
|
|
|
} |
210
|
|
|
|
|
|
|
|
211
|
|
|
|
|
|
|
|
212
|
|
|
|
|
|
|
=head1 TO DO |
213
|
|
|
|
|
|
|
|
214
|
|
|
|
|
|
|
Create C versions of C and C with L to see if they get close to C |
215
|
|
|
|
|
|
|
speed when running over a B file system (currently they are slower). |
216
|
|
|
|
|
|
|
|
217
|
|
|
|
|
|
|
=head1 SEE ALSO |
218
|
|
|
|
|
|
|
|
219
|
|
|
|
|
|
|
=over |
220
|
|
|
|
|
|
|
|
221
|
|
|
|
|
|
|
=item * |
222
|
|
|
|
|
|
|
|
223
|
|
|
|
|
|
|
L |
224
|
|
|
|
|
|
|
|
225
|
|
|
|
|
|
|
=item * |
226
|
|
|
|
|
|
|
|
227
|
|
|
|
|
|
|
L |
228
|
|
|
|
|
|
|
|
229
|
|
|
|
|
|
|
=item * |
230
|
|
|
|
|
|
|
|
231
|
|
|
|
|
|
|
The manual page of C. |
232
|
|
|
|
|
|
|
|
233
|
|
|
|
|
|
|
=item * |
234
|
|
|
|
|
|
|
|
235
|
|
|
|
|
|
|
L. |
236
|
|
|
|
|
|
|
|
237
|
|
|
|
|
|
|
=back |
238
|
|
|
|
|
|
|
|
239
|
|
|
|
|
|
|
=head1 AUTHOR |
240
|
|
|
|
|
|
|
|
241
|
|
|
|
|
|
|
Alceu Rodrigues de Freitas Junior, Earfreitas@cpan.orgE |
242
|
|
|
|
|
|
|
|
243
|
|
|
|
|
|
|
=head1 COPYRIGHT AND LICENSE |
244
|
|
|
|
|
|
|
|
245
|
|
|
|
|
|
|
This software is copyright (c) 2016 of Alceu Rodrigues de Freitas Junior, Earfreitas@cpan.orgE |
246
|
|
|
|
|
|
|
|
247
|
|
|
|
|
|
|
This file is part of Linux-NFS-BigDir distribution. |
248
|
|
|
|
|
|
|
|
249
|
|
|
|
|
|
|
Linux-NFS-BigDir is free software: you can redistribute it and/or modify |
250
|
|
|
|
|
|
|
it under the terms of the GNU General Public License as published by |
251
|
|
|
|
|
|
|
the Free Software Foundation, either version 3 of the License, or |
252
|
|
|
|
|
|
|
(at your option) any later version. |
253
|
|
|
|
|
|
|
|
254
|
|
|
|
|
|
|
Linux-NFS-BigDir is distributed in the hope that it will be useful, |
255
|
|
|
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of |
256
|
|
|
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
257
|
|
|
|
|
|
|
GNU General Public License for more details. |
258
|
|
|
|
|
|
|
|
259
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License |
260
|
|
|
|
|
|
|
along with Linux-NFS-BigDir. If not, see . |
261
|
|
|
|
|
|
|
|
262
|
|
|
|
|
|
|
=cut |
263
|
|
|
|
|
|
|
|
264
|
|
|
|
|
|
|
1; |