1 |
#!/usr/bin/perl |
2 |
|
3 |
package Fuse::DBI; |
4 |
|
5 |
use 5.008; |
6 |
use strict; |
7 |
use warnings; |
8 |
|
9 |
use POSIX qw(ENOENT EISDIR EINVAL ENOSYS O_RDWR); |
10 |
use Fuse; |
11 |
use DBI; |
12 |
use Carp; |
13 |
use Proc::Simple; |
14 |
use Data::Dumper; |
15 |
|
16 |
|
17 |
our $VERSION = '0.01'; |
18 |
|
19 |
=head1 NAME |
20 |
|
21 |
Fuse::DBI - mount your database as filesystem and use it |
22 |
|
23 |
=head1 SYNOPSIS |
24 |
|
25 |
use Fuse::DBI; |
26 |
Fuse::DBI->mount( ... ); |
27 |
|
28 |
See L<run> below for examples how to set parametars. |
29 |
|
30 |
=head1 DESCRIPTION |
31 |
|
32 |
This module will use L<Fuse> module, part of C<FUSE (Filesystem in USErspace)> |
33 |
available at L<http://sourceforge.net/projects/avf> to mount |
34 |
your database as file system. |
35 |
|
36 |
That will give you posibility to use normal file-system tools (cat, grep, vi) |
37 |
to manipulate data in database. |
38 |
|
39 |
It's actually opposite of Oracle's intention to put everything into database. |
40 |
|
41 |
|
42 |
=head1 METHODS |
43 |
|
44 |
=cut |
45 |
|
46 |
=head2 mount |
47 |
|
48 |
Mount your database as filesystem. |
49 |
|
50 |
my $mnt = Fuse::DBI->mount({ |
51 |
filenames => 'select name from filenamefilenames, |
52 |
read => 'sql read', |
53 |
update => 'sql update', |
54 |
dsn => 'DBI:Pg:dbname=webgui', |
55 |
user => 'database_user', |
56 |
password => 'database_password' |
57 |
}); |
58 |
|
59 |
=cut |
60 |
|
61 |
my $dbh; |
62 |
my $sth; |
63 |
my $ctime_start; |
64 |
|
65 |
sub read_filenames; |
66 |
|
67 |
sub mount { |
68 |
my $class = shift; |
69 |
my $self = {}; |
70 |
bless($self, $class); |
71 |
|
72 |
my $arg = shift; |
73 |
|
74 |
print Dumper($arg); |
75 |
|
76 |
carp "mount needs 'dsn' to connect to (e.g. dsn => 'DBI:Pg:dbname=test')" unless ($arg->{'dsn'}); |
77 |
carp "mount needs 'mount' as mountpoint" unless ($arg->{'mount'}); |
78 |
|
79 |
foreach (qw(filenames read update)) { |
80 |
carp "mount needs '$_' SQL" unless ($arg->{$_}); |
81 |
} |
82 |
|
83 |
$dbh = DBI->connect($arg->{'dsn'},$arg->{'user'},$arg->{'password'}, { AutoCommit => 0 }) || die $DBI::errstr; |
84 |
|
85 |
print "start transaction\n"; |
86 |
#$dbh->begin_work || die $dbh->errstr; |
87 |
|
88 |
$sth->{filenames} = $dbh->prepare($arg->{'filenames'}) || die $dbh->errstr(); |
89 |
|
90 |
$sth->{'read'} = $dbh->prepare($arg->{'read'}) || die $dbh->errstr(); |
91 |
$sth->{'update'} = $dbh->prepare($arg->{'update'}) || die $dbh->errstr(); |
92 |
|
93 |
$ctime_start = time(); |
94 |
|
95 |
read_filenames; |
96 |
|
97 |
$self->{'proc'} = Proc::Simple->new(); |
98 |
$self->{'proc'}->kill_on_destroy(1); |
99 |
|
100 |
$self->{'proc'}->start( sub { |
101 |
Fuse::main( |
102 |
mountpoint=>$arg->{'mount'}, |
103 |
getattr=>\&e_getattr, |
104 |
getdir=>\&e_getdir, |
105 |
open=>\&e_open, |
106 |
statfs=>\&e_statfs, |
107 |
read=>\&e_read, |
108 |
write=>\&e_write, |
109 |
utime=>\&e_utime, |
110 |
truncate=>\&e_truncate, |
111 |
debug=>0, |
112 |
); |
113 |
} ); |
114 |
|
115 |
$self ? return $self : return undef; |
116 |
}; |
117 |
|
118 |
=head2 umount |
119 |
|
120 |
Unmount your database as filesystem. |
121 |
|
122 |
$mnt->umount; |
123 |
|
124 |
This will also kill background process which is translating |
125 |
database to filesystem. |
126 |
|
127 |
=cut |
128 |
|
129 |
sub umount { |
130 |
my $self = shift; |
131 |
|
132 |
confess "no process running?" unless ($self->{'proc'}); |
133 |
$self->{'proc'}->kill; |
134 |
} |
135 |
|
136 |
|
137 |
my %files; |
138 |
my %dirs; |
139 |
|
140 |
sub read_filenames { |
141 |
my $self = shift; |
142 |
|
143 |
# create empty filesystem |
144 |
(%files) = ( |
145 |
'.' => { |
146 |
type => 0040, |
147 |
mode => 0755, |
148 |
}, |
149 |
# a => { |
150 |
# cont => "File 'a'.\n", |
151 |
# type => 0100, |
152 |
# ctime => time()-2000 |
153 |
# }, |
154 |
); |
155 |
|
156 |
# fetch new filename list from database |
157 |
$sth->{'filenames'}->execute() || die $sth->{'filenames'}->errstr(); |
158 |
|
159 |
# read them in with sesible defaults |
160 |
while (my $row = $sth->{'filenames'}->fetchrow_hashref() ) { |
161 |
$files{$row->{'filename'}} = { |
162 |
size => $row->{'size'}, |
163 |
mode => $row->{'writable'} ? 0644 : 0444, |
164 |
id => $row->{'id'} || 99, |
165 |
}; |
166 |
|
167 |
my $d; |
168 |
foreach (split(m!/!, $row->{'filename'})) { |
169 |
# first, entry is assumed to be file |
170 |
if ($d) { |
171 |
$files{$d} = { |
172 |
size => $dirs{$d}++, |
173 |
mode => 0755, |
174 |
type => 0040 |
175 |
}; |
176 |
$files{$d.'/.'} = { |
177 |
mode => 0755, |
178 |
type => 0040 |
179 |
}; |
180 |
$files{$d.'/..'} = { |
181 |
mode => 0755, |
182 |
type => 0040 |
183 |
}; |
184 |
} |
185 |
$d .= "/" if ($d); |
186 |
$d .= "$_"; |
187 |
} |
188 |
} |
189 |
|
190 |
print "found ",scalar(keys %files)-scalar(keys %dirs)," files, ",scalar(keys %dirs), " dirs\n"; |
191 |
} |
192 |
|
193 |
|
194 |
sub filename_fixup { |
195 |
my ($file) = shift; |
196 |
$file =~ s,^/,,; |
197 |
$file = '.' unless length($file); |
198 |
return $file; |
199 |
} |
200 |
|
201 |
sub e_getattr { |
202 |
my ($file) = filename_fixup(shift); |
203 |
$file =~ s,^/,,; |
204 |
$file = '.' unless length($file); |
205 |
return -ENOENT() unless exists($files{$file}); |
206 |
my ($size) = $files{$file}{size} || 1; |
207 |
my ($dev, $ino, $rdev, $blocks, $gid, $uid, $nlink, $blksize) = (0,0,0,1,0,0,1,1024); |
208 |
my ($atime, $ctime, $mtime); |
209 |
$atime = $ctime = $mtime = $files{$file}{ctime} || $ctime_start; |
210 |
|
211 |
my ($modes) = (($files{$file}{type} || 0100)<<9) + $files{$file}{mode}; |
212 |
|
213 |
# 2 possible types of return values: |
214 |
#return -ENOENT(); # or any other error you care to |
215 |
#print(join(",",($dev,$ino,$modes,$nlink,$uid,$gid,$rdev,$size,$atime,$mtime,$ctime,$blksize,$blocks)),"\n"); |
216 |
return ($dev,$ino,$modes,$nlink,$uid,$gid,$rdev,$size,$atime,$mtime,$ctime,$blksize,$blocks); |
217 |
} |
218 |
|
219 |
sub e_getdir { |
220 |
my ($dirname) = shift; |
221 |
$dirname =~ s!^/!!; |
222 |
# return as many text filenames as you like, followed by the retval. |
223 |
print((scalar keys %files)." files total\n"); |
224 |
my %out; |
225 |
foreach (keys %files) { |
226 |
my $f = $_; |
227 |
$f =~ s/^\E$dirname\Q//; |
228 |
$f =~ s/^\///; |
229 |
if ($dirname) { |
230 |
$out{$f}++ if (/^\E$dirname\Q/ && $f =~ /^[^\/]+$/); |
231 |
} else { |
232 |
$out{$f}++ if ($f =~ /^[^\/]+$/); |
233 |
} |
234 |
} |
235 |
if (! %out) { |
236 |
$out{'no files? bug?'}++; |
237 |
} |
238 |
print scalar keys %out," files in dir '$dirname'\n"; |
239 |
return (keys %out),0; |
240 |
} |
241 |
|
242 |
sub e_open { |
243 |
# VFS sanity check; it keeps all the necessary state, not much to do here. |
244 |
my $file = filename_fixup(shift); |
245 |
my $flags = shift; |
246 |
|
247 |
return -ENOENT() unless exists($files{$file}); |
248 |
return -EISDIR() unless exists($files{$file}{id}); |
249 |
|
250 |
if (!exists($files{$file}{cont})) { |
251 |
$sth->{'read'}->execute($files{$file}{id}) || die $sth->{'read'}->errstr; |
252 |
$files{$file}{cont} = $sth->{'read'}->fetchrow_array; |
253 |
print "file '$file' content read in cache\n"; |
254 |
} |
255 |
print "open '$file' ",length($files{$file}{cont})," bytes\n"; |
256 |
return 0; |
257 |
} |
258 |
|
259 |
sub e_read { |
260 |
# return an error numeric, or binary/text string. |
261 |
# (note: 0 means EOF, "0" will give a byte (ascii "0") |
262 |
# to the reading program) |
263 |
my ($file) = filename_fixup(shift); |
264 |
my ($buf_len,$off) = @_; |
265 |
|
266 |
return -ENOENT() unless exists($files{$file}); |
267 |
|
268 |
my $len = length($files{$file}{cont}); |
269 |
|
270 |
print "read '$file' [$len bytes] offset $off length $buf_len\n"; |
271 |
|
272 |
return -EINVAL() if ($off > $len); |
273 |
return 0 if ($off == $len); |
274 |
|
275 |
$buf_len = $buf_len-$off if ($off+$buf_len > $len); |
276 |
|
277 |
return substr($files{$file}{cont},$off,$buf_len); |
278 |
} |
279 |
|
280 |
sub clear_cont { |
281 |
print "transaction rollback\n"; |
282 |
$dbh->rollback || die $dbh->errstr; |
283 |
print "invalidate all cached content\n"; |
284 |
foreach my $f (keys %files) { |
285 |
delete $files{$f}{cont}; |
286 |
} |
287 |
print "begin new transaction\n"; |
288 |
$dbh->begin_work || die $dbh->errstr; |
289 |
} |
290 |
|
291 |
|
292 |
sub update_db { |
293 |
my $file = shift || die; |
294 |
|
295 |
$files{$file}{ctime} = time(); |
296 |
|
297 |
if (!$sth->{'update'}->execute($files{$file}{cont},$files{$file}{id})) { |
298 |
print "update problem: ",$sth->{'update'}->errstr; |
299 |
clear_cont; |
300 |
return 0; |
301 |
} else { |
302 |
if (! $dbh->commit) { |
303 |
print "ERROR: commit problem: ",$sth->{'update'}->errstr; |
304 |
clear_cont; |
305 |
return 0; |
306 |
} |
307 |
print "updated '$file' [",$files{$file}{id},"]\n"; |
308 |
} |
309 |
return 1; |
310 |
} |
311 |
|
312 |
sub e_write { |
313 |
my $file = filename_fixup(shift); |
314 |
my ($buf_len,$off) = @_; |
315 |
|
316 |
return -ENOENT() unless exists($files{$file}); |
317 |
|
318 |
my $len = length($files{$file}{cont}); |
319 |
|
320 |
print "write '$file' [$len bytes] offset $off length\n"; |
321 |
|
322 |
$files{$file}{cont} = |
323 |
substr($files{$file}{cont},0,$off) . |
324 |
$buf_len . |
325 |
substr($files{$file}{cont},$off+length($buf_len)); |
326 |
|
327 |
if (! update_db($file)) { |
328 |
return -ENOSYS(); |
329 |
} else { |
330 |
return length($buf_len); |
331 |
} |
332 |
} |
333 |
|
334 |
sub e_truncate { |
335 |
my $file = filename_fixup(shift); |
336 |
my $size = shift; |
337 |
|
338 |
$files{$file}{cont} = substr($files{$file}{cont},0,$size); |
339 |
return 0 |
340 |
}; |
341 |
|
342 |
|
343 |
sub e_utime { |
344 |
my ($atime,$mtime,$file) = @_; |
345 |
$file = filename_fixup($file); |
346 |
|
347 |
return -ENOENT() unless exists($files{$file}); |
348 |
|
349 |
print "utime '$file' $atime $mtime\n"; |
350 |
|
351 |
$files{$file}{time} = $mtime; |
352 |
return 0; |
353 |
} |
354 |
|
355 |
sub e_statfs { return 255, 1, 1, 1, 1, 2 } |
356 |
|
357 |
1; |
358 |
__END__ |
359 |
|
360 |
=head1 EXPORT |
361 |
|
362 |
Nothing. |
363 |
|
364 |
=head1 SEE ALSO |
365 |
|
366 |
C<FUSE (Filesystem in USErspace)> website |
367 |
L<http://sourceforge.net/projects/avf> |
368 |
|
369 |
=head1 AUTHOR |
370 |
|
371 |
Dobrica Pavlinusic, E<lt>dpavlin@rot13.orgE<gt> |
372 |
|
373 |
=head1 COPYRIGHT AND LICENSE |
374 |
|
375 |
Copyright (C) 2004 by Dobrica Pavlinusic |
376 |
|
377 |
This library is free software; you can redistribute it and/or modify |
378 |
it under the same terms as Perl itself, either Perl version 5.8.4 or, |
379 |
at your option, any later version of Perl 5 you may have available. |
380 |
|
381 |
|
382 |
=cut |
383 |
|