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 Data::Dumper; |
14 |
|
15 |
|
16 |
our $VERSION = '0.03'; |
17 |
|
18 |
=head1 NAME |
19 |
|
20 |
Fuse::DBI - mount your database as filesystem and use it |
21 |
|
22 |
=head1 SYNOPSIS |
23 |
|
24 |
use Fuse::DBI; |
25 |
Fuse::DBI->mount( ... ); |
26 |
|
27 |
See C<run> below for examples how to set parametars. |
28 |
|
29 |
=head1 DESCRIPTION |
30 |
|
31 |
This module will use C<Fuse> module, part of C<FUSE (Filesystem in USErspace)> |
32 |
available at L<http://sourceforge.net/projects/avf> to mount |
33 |
your database as file system. |
34 |
|
35 |
That will give you posibility to use normal file-system tools (cat, grep, vi) |
36 |
to manipulate data in database. |
37 |
|
38 |
It's actually opposite of Oracle's intention to put everything into database. |
39 |
|
40 |
|
41 |
=head1 METHODS |
42 |
|
43 |
=cut |
44 |
|
45 |
=head2 mount |
46 |
|
47 |
Mount your database as filesystem. |
48 |
|
49 |
my $mnt = Fuse::DBI->mount({ |
50 |
filenames => 'select name from files_table as filenames', |
51 |
read => 'sql read', |
52 |
update => 'sql update', |
53 |
dsn => 'DBI:Pg:dbname=webgui', |
54 |
user => 'database_user', |
55 |
password => 'database_password' |
56 |
}); |
57 |
|
58 |
=cut |
59 |
|
60 |
my $dbh; |
61 |
my $sth; |
62 |
my $ctime_start; |
63 |
|
64 |
sub read_filenames; |
65 |
sub fuse_module_loaded; |
66 |
|
67 |
# evil, evil way to solve this. It makes this module non-reentrant. But, since |
68 |
# fuse calls another copy of this script for each mount anyway, this shouldn't |
69 |
# be a problem. |
70 |
my $fuse_self; |
71 |
|
72 |
sub mount { |
73 |
my $class = shift; |
74 |
my $self = {}; |
75 |
bless($self, $class); |
76 |
|
77 |
my $arg = shift; |
78 |
|
79 |
print Dumper($arg); |
80 |
|
81 |
carp "mount needs 'dsn' to connect to (e.g. dsn => 'DBI:Pg:dbname=test')" unless ($arg->{'dsn'}); |
82 |
carp "mount needs 'mount' as mountpoint" unless ($arg->{'mount'}); |
83 |
|
84 |
# save (some) arguments in self |
85 |
foreach (qw(mount invalidate)) { |
86 |
$self->{$_} = $arg->{$_}; |
87 |
$fuse_self->{$_} = $arg->{$_}; |
88 |
} |
89 |
|
90 |
foreach (qw(filenames read update)) { |
91 |
carp "mount needs '$_' SQL" unless ($arg->{$_}); |
92 |
} |
93 |
|
94 |
$ctime_start = time(); |
95 |
|
96 |
my $pid; |
97 |
if ($arg->{'fork'}) { |
98 |
$pid = fork(); |
99 |
die "fork() failed: $!" unless defined $pid; |
100 |
# child will return to caller |
101 |
if ($pid) { |
102 |
return $self; |
103 |
} |
104 |
} |
105 |
|
106 |
$dbh = DBI->connect($arg->{'dsn'},$arg->{'user'},$arg->{'password'}, {AutoCommit => 0, RaiseError => 1}) || die $DBI::errstr; |
107 |
|
108 |
$sth->{filenames} = $dbh->prepare($arg->{'filenames'}) || die $dbh->errstr(); |
109 |
|
110 |
$sth->{'read'} = $dbh->prepare($arg->{'read'}) || die $dbh->errstr(); |
111 |
$sth->{'update'} = $dbh->prepare($arg->{'update'}) || die $dbh->errstr(); |
112 |
|
113 |
$self->read_filenames; |
114 |
|
115 |
Fuse::main( |
116 |
mountpoint=>$arg->{'mount'}, |
117 |
getattr=>\&e_getattr, |
118 |
getdir=>\&e_getdir, |
119 |
open=>\&e_open, |
120 |
statfs=>\&e_statfs, |
121 |
read=>\&e_read, |
122 |
write=>\&e_write, |
123 |
utime=>\&e_utime, |
124 |
truncate=>\&e_truncate, |
125 |
unlink=>\&e_unlink, |
126 |
debug=>0, |
127 |
); |
128 |
|
129 |
exit(0) if ($arg->{'fork'}); |
130 |
|
131 |
return 1; |
132 |
|
133 |
}; |
134 |
|
135 |
=head2 umount |
136 |
|
137 |
Unmount your database as filesystem. |
138 |
|
139 |
$mnt->umount; |
140 |
|
141 |
This will also kill background process which is translating |
142 |
database to filesystem. |
143 |
|
144 |
=cut |
145 |
|
146 |
sub umount { |
147 |
my $self = shift; |
148 |
|
149 |
system "fusermount -u ".$self->{'mount'} || croak "umount error: $!"; |
150 |
|
151 |
return 1; |
152 |
} |
153 |
|
154 |
#$SIG{'INT'} = sub { |
155 |
# print STDERR "umount called by SIG INT\n"; |
156 |
# umount; |
157 |
#}; |
158 |
|
159 |
sub DESTROY { |
160 |
my $self = shift; |
161 |
print STDERR "umount called by DESTROY\n"; |
162 |
$self->umount; |
163 |
} |
164 |
|
165 |
=head2 fuse_module_loaded |
166 |
|
167 |
Checks if C<fuse> module is loaded in kernel. |
168 |
|
169 |
die "no fuse module loaded in kernel" |
170 |
unless (Fuse::DBI::fuse_module_loaded); |
171 |
|
172 |
This function in called by L<mount>, but might be useful alone also. |
173 |
|
174 |
=cut |
175 |
|
176 |
sub fuse_module_loaded { |
177 |
my $lsmod = `lsmod`; |
178 |
die "can't start lsmod: $!" unless ($lsmod); |
179 |
if ($lsmod =~ m/fuse/s) { |
180 |
return 1; |
181 |
} else { |
182 |
return 0; |
183 |
} |
184 |
} |
185 |
|
186 |
my %files; |
187 |
my %dirs; |
188 |
|
189 |
sub read_filenames { |
190 |
my $self = shift; |
191 |
|
192 |
# create empty filesystem |
193 |
(%files) = ( |
194 |
'.' => { |
195 |
type => 0040, |
196 |
mode => 0755, |
197 |
}, |
198 |
# a => { |
199 |
# cont => "File 'a'.\n", |
200 |
# type => 0100, |
201 |
# ctime => time()-2000 |
202 |
# }, |
203 |
); |
204 |
|
205 |
# fetch new filename list from database |
206 |
$sth->{'filenames'}->execute() || die $sth->{'filenames'}->errstr(); |
207 |
|
208 |
# read them in with sesible defaults |
209 |
while (my $row = $sth->{'filenames'}->fetchrow_hashref() ) { |
210 |
$files{$row->{'filename'}} = { |
211 |
size => $row->{'size'}, |
212 |
mode => $row->{'writable'} ? 0644 : 0444, |
213 |
id => $row->{'id'} || 99, |
214 |
}; |
215 |
|
216 |
my $d; |
217 |
foreach (split(m!/!, $row->{'filename'})) { |
218 |
# first, entry is assumed to be file |
219 |
if ($d) { |
220 |
$files{$d} = { |
221 |
size => $dirs{$d}++, |
222 |
mode => 0755, |
223 |
type => 0040 |
224 |
}; |
225 |
$files{$d.'/.'} = { |
226 |
mode => 0755, |
227 |
type => 0040 |
228 |
}; |
229 |
$files{$d.'/..'} = { |
230 |
mode => 0755, |
231 |
type => 0040 |
232 |
}; |
233 |
} |
234 |
$d .= "/" if ($d); |
235 |
$d .= "$_"; |
236 |
} |
237 |
} |
238 |
|
239 |
print "found ",scalar(keys %files)-scalar(keys %dirs)," files, ",scalar(keys %dirs), " dirs\n"; |
240 |
} |
241 |
|
242 |
|
243 |
sub filename_fixup { |
244 |
my ($file) = shift; |
245 |
$file =~ s,^/,,; |
246 |
$file = '.' unless length($file); |
247 |
return $file; |
248 |
} |
249 |
|
250 |
sub e_getattr { |
251 |
my ($file) = filename_fixup(shift); |
252 |
$file =~ s,^/,,; |
253 |
$file = '.' unless length($file); |
254 |
return -ENOENT() unless exists($files{$file}); |
255 |
my ($size) = $files{$file}{size} || 1; |
256 |
my ($dev, $ino, $rdev, $blocks, $gid, $uid, $nlink, $blksize) = (0,0,0,1,0,0,1,1024); |
257 |
my ($atime, $ctime, $mtime); |
258 |
$atime = $ctime = $mtime = $files{$file}{ctime} || $ctime_start; |
259 |
|
260 |
my ($modes) = (($files{$file}{type} || 0100)<<9) + $files{$file}{mode}; |
261 |
|
262 |
# 2 possible types of return values: |
263 |
#return -ENOENT(); # or any other error you care to |
264 |
#print(join(",",($dev,$ino,$modes,$nlink,$uid,$gid,$rdev,$size,$atime,$mtime,$ctime,$blksize,$blocks)),"\n"); |
265 |
return ($dev,$ino,$modes,$nlink,$uid,$gid,$rdev,$size,$atime,$mtime,$ctime,$blksize,$blocks); |
266 |
} |
267 |
|
268 |
sub e_getdir { |
269 |
my ($dirname) = shift; |
270 |
$dirname =~ s!^/!!; |
271 |
# return as many text filenames as you like, followed by the retval. |
272 |
print((scalar keys %files)." files total\n"); |
273 |
my %out; |
274 |
foreach my $f (sort keys %files) { |
275 |
if ($dirname) { |
276 |
if ($f =~ s/^\E$dirname\Q\///) { |
277 |
$out{$f}++ if ($f =~ /^[^\/]+$/); |
278 |
} |
279 |
} else { |
280 |
$out{$f}++ if ($f =~ /^[^\/]+$/); |
281 |
} |
282 |
} |
283 |
if (! %out) { |
284 |
$out{'no files? bug?'}++; |
285 |
} |
286 |
print scalar keys %out," files in dir '$dirname'\n"; |
287 |
print "## ",join(" ",keys %out),"\n"; |
288 |
return (keys %out),0; |
289 |
} |
290 |
|
291 |
sub read_content { |
292 |
my ($file,$id) = @_; |
293 |
|
294 |
die "read_content needs file and id" unless ($file && $id); |
295 |
|
296 |
$sth->{'read'}->execute($id) || die $sth->{'read'}->errstr; |
297 |
$files{$file}{cont} = $sth->{'read'}->fetchrow_array; |
298 |
print "file '$file' content [",length($files{$file}{cont})," bytes] read in cache\n"; |
299 |
} |
300 |
|
301 |
|
302 |
sub e_open { |
303 |
# VFS sanity check; it keeps all the necessary state, not much to do here. |
304 |
my $file = filename_fixup(shift); |
305 |
my $flags = shift; |
306 |
|
307 |
return -ENOENT() unless exists($files{$file}); |
308 |
return -EISDIR() unless exists($files{$file}{id}); |
309 |
|
310 |
read_content($file,$files{$file}{id}) unless exists($files{$file}{cont}); |
311 |
|
312 |
print "open '$file' ",length($files{$file}{cont})," bytes\n"; |
313 |
return 0; |
314 |
} |
315 |
|
316 |
sub e_read { |
317 |
# return an error numeric, or binary/text string. |
318 |
# (note: 0 means EOF, "0" will give a byte (ascii "0") |
319 |
# to the reading program) |
320 |
my ($file) = filename_fixup(shift); |
321 |
my ($buf_len,$off) = @_; |
322 |
|
323 |
return -ENOENT() unless exists($files{$file}); |
324 |
|
325 |
my $len = length($files{$file}{cont}); |
326 |
|
327 |
print "read '$file' [$len bytes] offset $off length $buf_len\n"; |
328 |
|
329 |
return -EINVAL() if ($off > $len); |
330 |
return 0 if ($off == $len); |
331 |
|
332 |
$buf_len = $len-$off if ($len - $off < $buf_len); |
333 |
|
334 |
return substr($files{$file}{cont},$off,$buf_len); |
335 |
} |
336 |
|
337 |
sub clear_cont { |
338 |
print "transaction rollback\n"; |
339 |
$dbh->rollback || die $dbh->errstr; |
340 |
print "invalidate all cached content\n"; |
341 |
foreach my $f (keys %files) { |
342 |
delete $files{$f}{cont}; |
343 |
} |
344 |
print "begin new transaction\n"; |
345 |
#$dbh->begin_work || die $dbh->errstr; |
346 |
} |
347 |
|
348 |
|
349 |
sub update_db { |
350 |
my $file = shift || die; |
351 |
|
352 |
$files{$file}{ctime} = time(); |
353 |
|
354 |
my ($cont,$id) = ( |
355 |
$files{$file}{cont}, |
356 |
$files{$file}{id} |
357 |
); |
358 |
|
359 |
if (!$sth->{'update'}->execute($cont,$id)) { |
360 |
print "update problem: ",$sth->{'update'}->errstr; |
361 |
clear_cont; |
362 |
return 0; |
363 |
} else { |
364 |
if (! $dbh->commit) { |
365 |
print "ERROR: commit problem: ",$sth->{'update'}->errstr; |
366 |
clear_cont; |
367 |
return 0; |
368 |
} |
369 |
print "updated '$file' [",$files{$file}{id},"]\n"; |
370 |
|
371 |
$fuse_self->{'invalidate'}->() if (ref $fuse_self->{'invalidate'}); |
372 |
} |
373 |
return 1; |
374 |
} |
375 |
|
376 |
sub e_write { |
377 |
my $file = filename_fixup(shift); |
378 |
my ($buffer,$off) = @_; |
379 |
|
380 |
return -ENOENT() unless exists($files{$file}); |
381 |
|
382 |
my $cont = $files{$file}{cont}; |
383 |
my $len = length($cont); |
384 |
|
385 |
print "write '$file' [$len bytes] offset $off length ",length($buffer),"\n"; |
386 |
|
387 |
$files{$file}{cont} = ""; |
388 |
|
389 |
$files{$file}{cont} .= substr($cont,0,$off) if ($off > 0); |
390 |
$files{$file}{cont} .= $buffer; |
391 |
$files{$file}{cont} .= substr($cont,$off+length($buffer),$len-$off-length($buffer)) if ($off+length($buffer) < $len); |
392 |
|
393 |
$files{$file}{size} = length($files{$file}{cont}); |
394 |
|
395 |
if (! update_db($file)) { |
396 |
return -ENOSYS(); |
397 |
} else { |
398 |
return length($buffer); |
399 |
} |
400 |
} |
401 |
|
402 |
sub e_truncate { |
403 |
my $file = filename_fixup(shift); |
404 |
my $size = shift; |
405 |
|
406 |
print "truncate to $size\n"; |
407 |
|
408 |
$files{$file}{cont} = substr($files{$file}{cont},0,$size); |
409 |
$files{$file}{size} = $size; |
410 |
return 0 |
411 |
}; |
412 |
|
413 |
|
414 |
sub e_utime { |
415 |
my ($atime,$mtime,$file) = @_; |
416 |
$file = filename_fixup($file); |
417 |
|
418 |
return -ENOENT() unless exists($files{$file}); |
419 |
|
420 |
print "utime '$file' $atime $mtime\n"; |
421 |
|
422 |
$files{$file}{time} = $mtime; |
423 |
return 0; |
424 |
} |
425 |
|
426 |
sub e_statfs { return 255, 1, 1, 1, 1, 2 } |
427 |
|
428 |
sub e_unlink { |
429 |
my $file = filename_fixup(shift); |
430 |
|
431 |
return -ENOENT() unless exists($files{$file}); |
432 |
|
433 |
print "unlink '$file' will invalidate cache\n"; |
434 |
|
435 |
read_content($file,$files{$file}{id}); |
436 |
|
437 |
return 0; |
438 |
} |
439 |
1; |
440 |
__END__ |
441 |
|
442 |
=head1 EXPORT |
443 |
|
444 |
Nothing. |
445 |
|
446 |
=head1 SEE ALSO |
447 |
|
448 |
C<FUSE (Filesystem in USErspace)> website |
449 |
L<http://sourceforge.net/projects/avf> |
450 |
|
451 |
=head1 AUTHOR |
452 |
|
453 |
Dobrica Pavlinusic, E<lt>dpavlin@rot13.orgE<gt> |
454 |
|
455 |
=head1 COPYRIGHT AND LICENSE |
456 |
|
457 |
Copyright (C) 2004 by Dobrica Pavlinusic |
458 |
|
459 |
This library is free software; you can redistribute it and/or modify |
460 |
it under the same terms as Perl itself, either Perl version 5.8.4 or, |
461 |
at your option, any later version of Perl 5 you may have available. |
462 |
|
463 |
|
464 |
=cut |
465 |
|