ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/BDB/BDB.pm
(Generate patch)

Comparing BDB/BDB.pm (file contents):
Revision 1.7 by root, Mon Mar 5 19:47:01 2007 UTC vs.
Revision 1.57 by root, Tue Jul 14 19:29:23 2009 UTC

4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use BDB; 7 use BDB;
8 8
9 my $env = db_env_create;
10
11 mkdir "bdtest", 0700;
12 db_env_open
13 $env,
14 "bdtest",
15 BDB::INIT_LOCK | BDB::INIT_LOG | BDB::INIT_MPOOL
16 | BDB::INIT_TXN | BDB::RECOVER | BDB::USE_ENVIRON | BDB::CREATE,
17 0600;
18
19 $env->set_flags (BDB::AUTO_COMMIT | BDB::TXN_NOSYNC, 1);
20
21 my $db = db_create $env;
22 db_open $db, undef, "table", undef, BDB::BTREE, BDB::AUTO_COMMIT | BDB::CREATE
23 | BDB::READ_UNCOMMITTED, 0600;
24 db_put $db, undef, "key", "data", 0, sub {
25 db_del $db, undef, "key";
26 };
27 db_sync $db;
28
29 # when you also use Coro, management is easy:
30 use Coro::BDB;
31
32 # automatic event loop intergration with AnyEvent:
33 use AnyEvent::BDB;
34
35 # automatic result processing with EV:
36 my $WATCHER = EV::io BDB::poll_fileno, EV::READ, \&BDB::poll_cb;
37
38 # with Glib:
39 add_watch Glib::IO BDB::poll_fileno,
40 in => sub { BDB::poll_cb; 1 };
41
42 # or simply flush manually
43 BDB::flush;
44
45
9=head1 DESCRIPTION 46=head1 DESCRIPTION
10 47
11See the eg/ directory in the distribution and the berkeleydb C 48See the BerkeleyDB documentation (L<http://www.oracle.com/technology/documentation/berkeley-db/db/index.html>).
12documentation. This is inadequate, but the only sources of documentation 49The BDB API is very similar to the C API (the translation has been very faithful).
13known for this module so far.
14 50
15=head2 EXAMPLE 51See also the example sections in the document below and possibly the eg/
52subdirectory of the BDB distribution. Last not least see the IO::AIO
53documentation, as that module uses almost the same asynchronous request
54model as this module.
55
56I know this is woefully inadequate documentation. Send a patch!
57
16 58
17=head1 REQUEST ANATOMY AND LIFETIME 59=head1 REQUEST ANATOMY AND LIFETIME
18 60
19Every request method creates a request. which is a C data structure not 61Every request method creates a request. which is a C data structure not
20directly visible to Perl. 62directly visible to Perl.
61 103
62=cut 104=cut
63 105
64package BDB; 106package BDB;
65 107
66no warnings; 108use common::sense;
67use strict 'vars';
68 109
69use base 'Exporter'; 110use base 'Exporter';
70 111
112our $VERSION;
113
71BEGIN { 114BEGIN {
72 our $VERSION = '0.1'; 115 $VERSION = '1.84';
73 116
74 our @BDB_REQ = qw( 117 our @BDB_REQ = qw(
75 db_env_open db_env_close db_env_txn_checkpoint db_env_lock_detect 118 db_env_open db_env_close db_env_txn_checkpoint db_env_lock_detect
76 db_env_memp_sync db_env_memp_trickle 119 db_env_memp_sync db_env_memp_trickle db_env_dbrename db_env_dbremove
77 db_open db_close db_compact db_sync db_put db_get db_pget db_del db_key_range 120 db_env_log_archive
121 db_open db_close db_compact db_sync db_verify db_upgrade
122 db_put db_exists db_get db_pget db_del db_key_range
78 db_txn_commit db_txn_abort 123 db_txn_commit db_txn_abort db_txn_finish
79 db_c_close db_c_count db_c_put db_c_get db_c_pget db_c_del 124 db_c_close db_c_count db_c_put db_c_get db_c_pget db_c_del
80 db_sequence_open db_sequence_close 125 db_sequence_open db_sequence_close
81 db_sequence_get db_sequence_remove 126 db_sequence_get db_sequence_remove
82 ); 127 );
83 our @EXPORT = (@BDB_REQ, qw(dbreq_pri dbreq_nice db_env_create db_create)); 128 our @EXPORT = (@BDB_REQ, qw(dbreq_pri dbreq_nice db_env_create db_create));
90 135
91 require XSLoader; 136 require XSLoader;
92 XSLoader::load ("BDB", $VERSION); 137 XSLoader::load ("BDB", $VERSION);
93} 138}
94 139
140=head1 BERKELEYDB FUNCTIONS
141
142All of these are functions. The create functions simply return a new
143object and never block. All the remaining functions take an optional
144callback as last argument. If it is missing, then the function will be
145executed synchronously. In both cases, C<$!> will reflect the return value
146of the function.
147
148BDB functions that cannot block (mostly functions that manipulate
149settings) are method calls on the relevant objects, so the rule of thumb
150is: if it's a method, it's not blocking, if it's a function, it takes a
151callback as last argument.
152
153In the following, C<$int> signifies an integer return value,
154C<bdb_filename> is a "filename" (octets on unix, madness on windows),
155C<U32> is an unsigned 32 bit integer, C<int> is some integer, C<NV> is a
156floating point value.
157
158Most C<SV *> types are generic perl scalars (for input and output of data
159values).
160
161The various C<DB_ENV> etc. arguments are handles return by
162C<db_env_create>, C<db_create>, C<txn_begin> and so on. If they have an
163appended C<_ornull> this means they are optional and you can pass C<undef>
164for them, resulting a NULL pointer on the C level.
165
166The C<SV *callback> is the optional callback function to call when the
167request is completed. This last callback argument is special: the callback
168is simply the last argument passed. If there are "optional" arguments
169before the callback they can be left out. The callback itself can be left
170out or specified as C<undef>, in which case the function will be executed
171synchronously.
172
173For example, C<db_env_txn_checkpoint> usually is called with all integer
174arguments zero. These can be left out, so all of these specify a call
175to C<< DB_ENV->txn_checkpoint >>, to be executed asynchronously with a
176callback to be called:
177
178 db_env_txn_checkpoint $db_env, 0, 0, 0, sub { };
179 db_env_txn_checkpoint $db_env, 0, 0, sub { };
180 db_env_txn_checkpoint $db_env, sub { };
181
182While these all specify a call to C<< DB_ENV->txn_checkpoint >> to be
183executed synchronously:
184
185 db_env_txn_checkpoint $db_env, 0, 0, 0, undef;
186 db_env_txn_checkpoint $db_env, 0, 0, 0;
187 db_env_txn_checkpoint $db_env, 0;
188
189=head2 BDB functions
190
191Functions in the BDB namespace, exported by default:
192
193 $env = db_env_create (U32 env_flags = 0)
194 flags: RPCCLIENT
195
196 db_env_open (DB_ENV *env, bdb_filename db_home, U32 open_flags, int mode, SV *callback = 0)
197 open_flags: INIT_CDB INIT_LOCK INIT_LOG INIT_MPOOL INIT_REP INIT_TXN RECOVER RECOVER_FATAL USE_ENVIRON USE_ENVIRON_ROOT CREATE LOCKDOWN PRIVATE REGISTER SYSTEM_MEM
198 db_env_close (DB_ENV *env, U32 flags = 0, SV *callback = 0)
199 db_env_txn_checkpoint (DB_ENV *env, U32 kbyte = 0, U32 min = 0, U32 flags = 0, SV *callback = 0)
200 flags: FORCE
201 db_env_lock_detect (DB_ENV *env, U32 flags = 0, U32 atype = DB_LOCK_DEFAULT, SV *dummy = 0, SV *callback = 0)
202 atype: LOCK_DEFAULT LOCK_EXPIRE LOCK_MAXLOCKS LOCK_MAXWRITE LOCK_MINLOCKS LOCK_MINWRITE LOCK_OLDEST LOCK_RANDOM LOCK_YOUNGEST
203 db_env_memp_sync (DB_ENV *env, SV *dummy = 0, SV *callback = 0)
204 db_env_memp_trickle (DB_ENV *env, int percent, SV *dummy = 0, SV *callback = 0)
205 db_env_dbremove (DB_ENV *env, DB_TXN_ornull *txnid, bdb_filename file, bdb_filename database, U32 flags = 0, SV *callback = 0)
206 db_env_dbrename (DB_ENV *env, DB_TXN_ornull *txnid, bdb_filename file, bdb_filename database, bdb_filename newname, U32 flags = 0, SV *callback = 0)
207 db_env_log_archive (DB_ENV *env, SV *listp, U32 flags = 0, SV *callback = 0)
208
209 $db = db_create (DB_ENV *env = 0, U32 flags = 0)
210 flags: XA_CREATE
211
212 db_open (DB *db, DB_TXN_ornull *txnid, bdb_filename file, bdb_filename database, int type, U32 flags, int mode, SV *callback = 0)
213 flags: AUTO_COMMIT CREATE EXCL MULTIVERSION NOMMAP RDONLY READ_UNCOMMITTED THREAD TRUNCATE
214 db_close (DB *db, U32 flags = 0, SV *callback = 0)
215 flags: DB_NOSYNC
216 db_verify (DB *db, bdb_filename file, bdb_filename database = 0, SV *dummy = 0, U32 flags = 0, SV *callback = 0)
217 db_upgrade (DB *db, bdb_filename file, U32 flags = 0, SV *callback = 0)
218 db_compact (DB *db, DB_TXN_ornull *txn = 0, SV *start = 0, SV *stop = 0, SV *unused1 = 0, U32 flags = DB_FREE_SPACE, SV *unused2 = 0, SV *callback = 0)
219 flags: FREELIST_ONLY FREE_SPACE
220 db_sync (DB *db, U32 flags = 0, SV *callback = 0)
221 db_key_range (DB *db, DB_TXN_ornull *txn, SV *key, SV *key_range, U32 flags = 0, SV *callback = 0)
222 db_put (DB *db, DB_TXN_ornull *txn, SV *key, SV *data, U32 flags = 0, SV *callback = 0)
223 flags: APPEND NODUPDATA NOOVERWRITE
224 db_exists (DB *db, DB_TXN_ornull *txn, SV *key, U32 flags = 0, SV *callback = 0) (v4.6)
225 db_get (DB *db, DB_TXN_ornull *txn, SV *key, SV *data, U32 flags = 0, SV *callback = 0)
226 flags: CONSUME CONSUME_WAIT GET_BOTH SET_RECNO MULTIPLE READ_COMMITTED READ_UNCOMMITTED RMW
227 db_pget (DB *db, DB_TXN_ornull *txn, SV *key, SV *pkey, SV *data, U32 flags = 0, SV *callback = 0)
228 flags: CONSUME CONSUME_WAIT GET_BOTH SET_RECNO MULTIPLE READ_COMMITTED READ_UNCOMMITTED RMW
229 db_del (DB *db, DB_TXN_ornull *txn, SV *key, U32 flags = 0, SV *callback = 0)
230 db_txn_commit (DB_TXN *txn, U32 flags = 0, SV *callback = 0)
231 flags: TXN_NOSYNC TXN_SYNC
232 db_txn_abort (DB_TXN *txn, SV *callback = 0)
233
234 db_c_close (DBC *dbc, SV *callback = 0)
235 db_c_count (DBC *dbc, SV *count, U32 flags = 0, SV *callback = 0)
236 db_c_put (DBC *dbc, SV *key, SV *data, U32 flags = 0, SV *callback = 0)
237 flags: AFTER BEFORE CURRENT KEYFIRST KEYLAST NODUPDATA
238 db_c_get (DBC *dbc, SV *key, SV *data, U32 flags = 0, SV *callback = 0)
239 flags: CURRENT FIRST GET_BOTH GET_BOTH_RANGE GET_RECNO JOIN_ITEM LAST NEXT NEXT_DUP NEXT_NODUP PREV PREV_DUP PREV_NODUP SET SET_RANGE SET_RECNO READ_UNCOMMITTED MULTIPLE MULTIPLE_KEY RMW
240 db_c_pget (DBC *dbc, SV *key, SV *pkey, SV *data, U32 flags = 0, SV *callback = 0)
241 db_c_del (DBC *dbc, U32 flags = 0, SV *callback = 0)
242
243 db_sequence_open (DB_SEQUENCE *seq, DB_TXN_ornull *txnid, SV *key, U32 flags = 0, SV *callback = 0)
244 flags: CREATE EXCL
245 db_sequence_close (DB_SEQUENCE *seq, U32 flags = 0, SV *callback = 0)
246 db_sequence_get (DB_SEQUENCE *seq, DB_TXN_ornull *txnid, int delta, SV *seq_value, U32 flags = DB_TXN_NOSYNC, SV *callback = 0)
247 flags: TXN_NOSYNC
248 db_sequence_remove (DB_SEQUENCE *seq, DB_TXN_ornull *txnid = 0, U32 flags = 0, SV *callback = 0)
249 flags: TXN_NOSYNC
250
251=head3 db_txn_finish (DB_TXN *txn, U32 flags = 0, SV *callback = 0)
252
253This is not actually a Berkeley DB function but a BDB module
254extension. The background for this exytension is: It is very annoying to
255have to check every single BDB function for error returns and provide a
256codepath out of your transaction. While the BDB module still makes this
257possible, it contains the following extensions:
258
259When a transaction-protected function returns any operating system
260error (errno > 0), BDB will set the C<TXN_DEADLOCK> flag on the
261transaction. This flag is also set by Berkeley DB functions themselves
262when an operation fails with LOCK_DEADLOCK, and it causes all further
263operations on that transaction (including C<db_txn_commit>) to fail.
264
265The C<db_txn_finish> request will look at this flag, and, if it is set,
266will automatically call C<db_txn_abort> (setting errno to C<LOCK_DEADLOCK>
267if it isn't set to something else yet). If it isn't set, it will call
268C<db_txn_commit> and return the error normally.
269
270How to use this? Easy: just write your transaction normally:
271
272 my $txn = $db_env->txn_begin;
273 db_get $db, $txn, "key", my $data;
274 db_put $db, $txn, "key", $data + 1 unless $! == BDB::NOTFOUND;
275 db_txn_finish $txn;
276 die "transaction failed" if $!;
277
278That is, handle only the expected errors. If something unexpected happens
279(EIO, LOCK_NOTGRANTED or a deadlock in either db_get or db_put), then the remaining
280requests (db_put in this case) will simply be skipped (they will fail with
281LOCK_DEADLOCK) and the transaction will be aborted.
282
283You can use the C<< $txn->failed >> method to check wether a transaction
284has failed in this way and abort further processing (excluding
285C<db_txn_finish>).
286
287
288=head2 DB_ENV/database environment methods
289
290Methods available on DB_ENV/$env handles:
291
292 DESTROY (DB_ENV_ornull *env)
293 CODE:
294 if (env)
295 env->close (env, 0);
296
297 $int = $env->set_data_dir (const char *dir)
298 $int = $env->set_tmp_dir (const char *dir)
299 $int = $env->set_lg_dir (const char *dir)
300 $int = $env->set_shm_key (long shm_key)
301 $int = $env->set_cachesize (U32 gbytes, U32 bytes, int ncache = 0)
302 $int = $env->set_flags (U32 flags, int onoff = 1)
303 $int = $env->log_set_config (U32 flags, int onoff = 1) (v4.7)
304 $int = $env->set_intermediate_dir_mode (const char *modestring) (v4.7)
305 $env->set_errfile (FILE *errfile = 0)
306 $env->set_msgfile (FILE *msgfile = 0)
307 $int = $env->set_verbose (U32 which, int onoff = 1)
308 $int = $env->set_encrypt (const char *password, U32 flags = 0)
309 $int = $env->set_timeout (NV timeout_seconds, U32 flags = SET_TXN_TIMEOUT)
310 $int = $env->set_mp_max_openfd (int maxopenfd);
311 $int = $env->set_mp_max_write (int maxwrite, int maxwrite_sleep);
312 $int = $env->set_mp_mmapsize (int mmapsize_mb)
313 $int = $env->set_lk_detect (U32 detect = DB_LOCK_DEFAULT)
314 $int = $env->set_lk_max_lockers (U32 max)
315 $int = $env->set_lk_max_locks (U32 max)
316 $int = $env->set_lk_max_objects (U32 max)
317 $int = $env->set_lg_bsize (U32 max)
318 $int = $env->set_lg_max (U32 max)
319 $int = $env->mutex_set_increment (U32 increment)
320 $int = $env->mutex_set_tas_spins (U32 tas_spins)
321 $int = $env->mutex_set_max (U32 max)
322 $int = $env->mutex_set_align (U32 align)
323
324 $txn = $env->txn_begin (DB_TXN_ornull *parent = 0, U32 flags = 0)
325 flags: READ_COMMITTED READ_UNCOMMITTED TXN_NOSYNC TXN_NOWAIT TXN_SNAPSHOT TXN_SYNC TXN_WAIT TXN_WRITE_NOSYNC
326 $txn = $env->cdsgroup_begin; (v4.5)
327
328=head3 Example:
329
330 use AnyEvent;
331 use BDB;
332
333 our $FH; open $FH, "<&=" . BDB::poll_fileno;
334 our $WATCHER = AnyEvent->io (fh => $FH, poll => 'r', cb => \&BDB::poll_cb);
335
336 BDB::min_parallel 8;
337
338 my $env = db_env_create;
339
340 mkdir "bdtest", 0700;
341 db_env_open
342 $env,
343 "bdtest",
344 BDB::INIT_LOCK | BDB::INIT_LOG | BDB::INIT_MPOOL | BDB::INIT_TXN | BDB::RECOVER | BDB::USE_ENVIRON | BDB::CREATE,
345 0600;
346
347 $env->set_flags (BDB::AUTO_COMMIT | BDB::TXN_NOSYNC, 1);
348
349
350=head2 DB/database methods
351
352Methods available on DB/$db handles:
353
354 DESTROY (DB_ornull *db)
355 CODE:
356 if (db)
357 {
358 SV *env = (SV *)db->app_private;
359 db->close (db, 0);
360 SvREFCNT_dec (env);
361 }
362
363 $int = $db->set_cachesize (U32 gbytes, U32 bytes, int ncache = 0)
364 $int = $db->set_flags (U32 flags)
365 flags: CHKSUM ENCRYPT TXN_NOT_DURABLE
366 Btree: DUP DUPSORT RECNUM REVSPLITOFF
367 Hash: DUP DUPSORT
368 Queue: INORDER
369 Recno: RENUMBER SNAPSHOT
370
371 $int = $db->set_encrypt (const char *password, U32 flags)
372 $int = $db->set_lorder (int lorder)
373 $int = $db->set_bt_minkey (U32 minkey)
374 $int = $db->set_re_delim (int delim)
375 $int = $db->set_re_pad (int re_pad)
376 $int = $db->set_re_source (char *source)
377 $int = $db->set_re_len (U32 re_len)
378 $int = $db->set_h_ffactor (U32 h_ffactor)
379 $int = $db->set_h_nelem (U32 h_nelem)
380 $int = $db->set_q_extentsize (U32 extentsize)
381
382 $dbc = $db->cursor (DB_TXN_ornull *txn = 0, U32 flags = 0)
383 flags: READ_COMMITTED READ_UNCOMMITTED WRITECURSOR TXN_SNAPSHOT
384 $seq = $db->sequence (U32 flags = 0)
385
386=head3 Example:
387
388 my $db = db_create $env;
389 db_open $db, undef, "table", undef, BDB::BTREE, BDB::AUTO_COMMIT | BDB::CREATE | BDB::READ_UNCOMMITTED, 0600;
390
391 for (1..1000) {
392 db_put $db, undef, "key $_", "data $_";
393
394 db_key_range $db, undef, "key $_", my $keyrange;
395 my ($lt, $eq, $gt) = @$keyrange;
396 }
397
398 db_del $db, undef, "key $_" for 1..1000;
399
400 db_sync $db;
401
402
403=head2 DB_TXN/transaction methods
404
405Methods available on DB_TXN/$txn handles:
406
407 DESTROY (DB_TXN_ornull *txn)
408 CODE:
409 if (txn)
410 txn->abort (txn);
411
412 $int = $txn->set_timeout (NV timeout_seconds, U32 flags = SET_TXN_TIMEOUT)
413 flags: SET_LOCK_TIMEOUT SET_TXN_TIMEOUT
414
415 $bool = $txn->failed
416 # see db_txn_finish documentation, above
417
418
419=head2 DBC/cursor methods
420
421Methods available on DBC/$dbc handles:
422
423 DESTROY (DBC_ornull *dbc)
424 CODE:
425 if (dbc)
426 dbc->c_close (dbc);
427
428 $int = $cursor->set_priority ($priority = PRIORITY_*) (v4.6)
429
430=head3 Example:
431
432 my $c = $db->cursor;
433
434 for (;;) {
435 db_c_get $c, my $key, my $data, BDB::NEXT;
436 warn "<$!,$key,$data>";
437 last if $!;
438 }
439
440 db_c_close $c;
441
442
443=head2 DB_SEQUENCE/sequence methods
444
445Methods available on DB_SEQUENCE/$seq handles:
446
447 DESTROY (DB_SEQUENCE_ornull *seq)
448 CODE:
449 if (seq)
450 seq->close (seq, 0);
451
452 $int = $seq->initial_value (db_seq_t value)
453 $int = $seq->set_cachesize (U32 size)
454 $int = $seq->set_flags (U32 flags)
455 flags: SEQ_DEC SEQ_INC SEQ_WRAP
456 $int = $seq->set_range (db_seq_t min, db_seq_t max)
457
458=head3 Example:
459
460 my $seq = $db->sequence;
461
462 db_sequence_open $seq, undef, "seq", BDB::CREATE;
463 db_sequence_get $seq, undef, 1, my $value;
464
465
95=head2 SUPPORT FUNCTIONS 466=head1 SUPPORT FUNCTIONS
96 467
97=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION 468=head2 EVENT PROCESSING AND EVENT LOOP INTEGRATION
98 469
99=over 4 470=over 4
471
472=item $msg = BDB::strerror [$errno]
473
474Returns the string corresponding to the given errno value. If no argument
475is given, use C<$!>.
476
477Note that the BDB module also patches the C<$!> variable directly, so you
478should be able to get a bdb error string by simply stringifying C<$!>.
100 479
101=item $fileno = BDB::poll_fileno 480=item $fileno = BDB::poll_fileno
102 481
103Return the I<request result pipe file descriptor>. This filehandle must be 482Return the I<request result pipe file descriptor>. This filehandle must be
104polled for reading by some mechanism outside this module (e.g. Event or 483polled for reading by some mechanism outside this module (e.g. Event or
143interactiveness when perl is not fast enough to process all requests in 522interactiveness when perl is not fast enough to process all requests in
144time. 523time.
145 524
146For interactive programs, values such as C<0.01> to C<0.1> should be fine. 525For interactive programs, values such as C<0.01> to C<0.1> should be fine.
147 526
148Example: Install an Event watcher that automatically calls 527Example: Install an EV watcher that automatically calls
149BDB::poll_cb with low priority, to ensure that other parts of the 528BDB::poll_cb with low priority, to ensure that other parts of the
150program get the CPU sometimes even under high AIO load. 529program get the CPU sometimes even under high load.
151 530
152 # try not to spend much more than 0.1s in poll_cb 531 # try not to spend much more than 0.1s in poll_cb
153 BDB::max_poll_time 0.1; 532 BDB::max_poll_time 0.1;
154 533
155 # use a low priority so other tasks have priority 534 my $bdb_poll = EV::io BDB::poll_fileno, EV::READ, \&BDB::poll_cb);
156 Event->io (fd => BDB::poll_fileno,
157 poll => 'r', nice => 1,
158 cb => &BDB::poll_cb);
159 535
160=item BDB::poll_wait 536=item BDB::poll_wait
161 537
162If there are any outstanding requests and none of them in the result 538If there are any outstanding requests and none of them in the result
163phase, wait till the result filehandle becomes ready for reading (simply 539phase, wait till the result filehandle becomes ready for reading (simply
175 551
176 BDB::poll_wait, BDB::poll_cb 552 BDB::poll_wait, BDB::poll_cb
177 553
178=item BDB::flush 554=item BDB::flush
179 555
180Wait till all outstanding AIO requests have been handled. 556Wait till all outstanding BDB requests have been handled.
181 557
182Strictly equivalent to: 558Strictly equivalent to:
183 559
184 BDB::poll_wait, BDB::poll_cb 560 BDB::poll_wait, BDB::poll_cb
185 while BDB::nreqs; 561 while BDB::nreqs;
186 562
563=back
564
565=head2 VERSION CHECKING
566
567BerkeleyDB comes in various versions, many of them have minor
568incompatibilities. This means that traditional "at least version x.x"
569checks are often not sufficient.
570
571Example: set the log_autoremove option in a way compatible with <v4.7 and
572v4.7. Note the use of & on the constants to avoid triggering a compiletime
573bug when the symbol isn't available.
574
575 $DB_ENV->set_flags (&BDB::LOG_AUTOREMOVE ) if BDB::VERSION v0, v4.7;
576 $DB_ENV->log_set_config (&BDB::LOG_AUTO_REMOVE) if BDB::VERSION v4.7;
577
578=over 4
579
580=item BDB::VERSION
581
582The C<BDB::VERSION> function, when called without arguments, returns the
583Berkeley DB version as a v-string (usually with 3 components). You should
584use C<lt> and C<ge> operators exclusively to make comparisons.
585
586Example: check for at least version 4.7.
587
588 BDB::VERSION ge v4.7 or die;
589
590=item BDB::VERSION min-version
591
592Returns true if the BDB version is at least the given version (specified
593as a v-string), false otherwise.
594
595Example: check for at least version 4.5.
596
597 BDB::VERSION v4.7 or die;
598
599=item BDB::VERSION min-version, max-version
600
601Returns true of the BDB version is at least version C<min-version> (specify C<undef> or C<v0> for any minimum version)
602and less then C<max-version>.
603
604Example: check wether version is strictly less then v4.7.
605
606 BDB::VERSION v0, v4.7
607 or die "version 4.7 is not yet supported";
608
609=back
610
611=cut
612
613sub VERSION {
614 # I was dumb enough to override the VERSION method here, so let's try
615 # to fix it up.
616
617 if ($_[0] eq __PACKAGE__) {
618 $VERSION
619 } else {
620 if (@_ > 0) {
621 return undef if VERSION_v lt $_[0];
622 if (@_ > 1) {
623 return undef if VERSION_v ge $_[1];
624 }
625 }
626
627 VERSION_v
628 }
629}
630
187=head3 CONTROLLING THE NUMBER OF THREADS 631=head2 CONTROLLING THE NUMBER OF THREADS
632
633=over 4
188 634
189=item BDB::min_parallel $nthreads 635=item BDB::min_parallel $nthreads
190 636
191Set the minimum number of AIO threads to C<$nthreads>. The current 637Set the minimum number of BDB threads to C<$nthreads>. The current
192default is C<8>, which means eight asynchronous operations can execute 638default is C<8>, which means eight asynchronous operations can execute
193concurrently at any one time (the number of outstanding requests, 639concurrently at any one time (the number of outstanding requests,
194however, is unlimited). 640however, is unlimited).
195 641
196BDB starts threads only on demand, when an AIO request is queued and 642BDB starts threads only on demand, when an BDB request is queued and
197no free thread exists. Please note that queueing up a hundred requests can 643no free thread exists. Please note that queueing up a hundred requests can
198create demand for a hundred threads, even if it turns out that everything 644create demand for a hundred threads, even if it turns out that everything
199is in the cache and could have been processed faster by a single thread. 645is in the cache and could have been processed faster by a single thread.
200 646
201It is recommended to keep the number of threads relatively low, as some 647It is recommended to keep the number of threads relatively low, as some
206Under most circumstances you don't need to call this function, as the 652Under most circumstances you don't need to call this function, as the
207module selects a default that is suitable for low to moderate load. 653module selects a default that is suitable for low to moderate load.
208 654
209=item BDB::max_parallel $nthreads 655=item BDB::max_parallel $nthreads
210 656
211Sets the maximum number of AIO threads to C<$nthreads>. If more than the 657Sets the maximum number of BDB threads to C<$nthreads>. If more than the
212specified number of threads are currently running, this function kills 658specified number of threads are currently running, this function kills
213them. This function blocks until the limit is reached. 659them. This function blocks until the limit is reached.
214 660
215While C<$nthreads> are zero, aio requests get queued but not executed 661While C<$nthreads> are zero, aio requests get queued but not executed
216until the number of threads has been increased again. 662until the number of threads has been increased again.
251 697
252You can still queue as many requests as you want. Therefore, 698You can still queue as many requests as you want. Therefore,
253C<max_oustsanding> is mainly useful in simple scripts (with low values) or 699C<max_oustsanding> is mainly useful in simple scripts (with low values) or
254as a stop gap to shield against fatal memory overflow (with large values). 700as a stop gap to shield against fatal memory overflow (with large values).
255 701
256=item BDB::set_sync_prepare $cb 702=item $old_cb = BDB::set_sync_prepare $cb
257 703
258Sets a callback that is called whenever a request is created without an 704Sets a callback that is called whenever a request is created without an
259explicit callback. It has to return two code references. The first is used 705explicit callback. It has to return two code references. The first is used
260as the request callback, and the second is called to wait until the first 706as the request callback (it should save the return status), and the second
261callback has been called. The default implementation works like this: 707is called to wait until the first callback has been called (it must set
708C<$!> to the return status).
709
710This mechanism can be used to include BDB into other event mechanisms,
711such as L<Coro::BDB>.
712
713To allow other, callback-based, events to be executed while callback-less
714ones are run, you could use this sync prepare function:
262 715
263 sub { 716 sub {
264 my $status; 717 my $status;
265 ( 718 (
266 sub { $status = $! }, 719 sub { $status = $! },
267 sub { BDB::poll while !defined $status; $! = $status }, 720 sub { BDB::poll while !defined $status; $! = $status },
268 ) 721 )
269 } 722 }
270 723
724It works by polling for results till the request has finished and then
725sets C<$!> to the return value. This means that if you don't use a
726callback, BDB would simply fall back to synchronous operations.
727
728By default, or if the sync prepare function is set to C<undef>, is to
729execute callback-less BDB requests in the foreground thread, setting C<$!>
730to the return value, without polling for other events.
731
271=back 732=back
272 733
273=head3 STATISTICAL INFORMATION 734=head2 STATISTICAL INFORMATION
274 735
275=over 4 736=over 4
276 737
277=item BDB::nreqs 738=item BDB::nreqs
278 739
296 757
297=back 758=back
298 759
299=cut 760=cut
300 761
301set_sync_prepare { 762set_sync_prepare (undef);
302 my $status;
303 (
304 sub {
305 $status = $!;
306 },
307 sub {
308 BDB::poll while !defined $status;
309 $! = $status;
310 },
311 )
312};
313 763
314min_parallel 8; 764min_parallel 8;
315 765
316END { flush } 766END { flush }
317 767
3181; 7681;
319 769
770=head1 COMMON PITFALLS
771
772=head2 Unexpected Crashes
773
774Remember that, by default, BDB will execute requests in parallel, in
775somewhat random order. That means that it is easy to run a C<db_get>
776request on thesa me database as a concurrent C<db_close> request, leading
777to a crash, silent data corruption, eventually the next world war on
778terrorism.
779
780If you only ever use foreground requests (without a callback), this will
781not be an issue.
782
783=head2 Unexpected Freezes or Deadlocks
784
785Remember that, by default, BDB will execute requests in parallel, which
786easily leads to deadlocks (even concurrent put's on the same database can
787deadlock).
788
789You either need to run deadlock detection (and handle the resulting
790errors), or make sure only one process ever updates the database, ine one
791thread, e.g. by using only foreground requests (without a callback).
792
320=head2 FORK BEHAVIOUR 793=head1 FORK BEHAVIOUR
321 794
322This module should do "the right thing" when the process using it forks: 795This module should do "the right thing" when the process using it forks:
323 796
324Before the fork, IO::AIO enters a quiescent state where no requests 797Before the fork, BDB enters a quiescent state where no requests
325can be added in other threads and no results will be processed. After 798can be added in other threads and no results will be processed. After
326the fork the parent simply leaves the quiescent state and continues 799the fork the parent simply leaves the quiescent state and continues
327request/result processing, while the child frees the request/result queue 800request/result processing, while the child frees the request/result queue
328(so that the requests started before the fork will only be handled in the 801(so that the requests started before the fork will only be handled in the
329parent). Threads will be started on demand until the limit set in the 802parent). Threads will be started on demand until the limit set in the
330parent process has been reached again. 803parent process has been reached again.
331 804
332In short: the parent will, after a short pause, continue as if fork had 805In short: the parent will, after a short pause, continue as if fork had
333not been called, while the child will act as if IO::AIO has not been used 806not been called, while the child will act as if BDB has not been used
334yet. 807yet.
335 808
809Win32 note: there is no fork on win32, and perls emulation of it is too
810broken to be supported, so do not use BDB in a windows pseudo-fork, better
811yet, switch to a more capable platform.
812
336=head2 MEMORY USAGE 813=head1 MEMORY USAGE
337 814
338Per-request usage: 815Per-request usage:
339 816
340Each aio request uses - depending on your architecture - around 100-200 817Each aio request uses - depending on your architecture - around 100-200
341bytes of memory. In addition, stat requests need a stat buffer (possibly 818bytes of memory. In addition, stat requests need a stat buffer (possibly
342a few hundred bytes), readdir requires a result buffer and so on. Perl 819a few hundred bytes), readdir requires a result buffer and so on. Perl
343scalars and other data passed into aio requests will also be locked and 820scalars and other data passed into aio requests will also be locked and
344will consume memory till the request has entered the done state. 821will consume memory till the request has entered the done state.
345 822
346This is now awfully much, so queuing lots of requests is not usually a 823This is not awfully much, so queuing lots of requests is not usually a
347problem. 824problem.
348 825
349Per-thread usage: 826Per-thread usage:
350 827
351In the execution phase, some aio requests require more memory for 828In the execution phase, some aio requests require more memory for
352temporary buffers, and each thread requires a stack and other data 829temporary buffers, and each thread requires a stack and other data
353structures (usually around 16k-128k, depending on the OS). 830structures (usually around 16k-128k, depending on the OS).
354 831
832=head1 WIN32 FILENAMES/DATABASE NAME MESS
833
834Perl on Win32 supports only ASCII filenames (the reason is that it abuses
835an internal flag to store wether a filename is Unicode or ANSI, but that
836flag is used for somethign else in the perl core, so there is no way to
837detect wether a filename is ANSI or Unicode-encoded). The BDB module
838tries to work around this issue by assuming that the filename is an ANSI
839filename and BDB was built for unicode support.
840
355=head1 KNOWN BUGS 841=head1 KNOWN BUGS
356 842
357Known bugs will be fixed in the next release. 843Known bugs will be fixed in the next release, except:
844
845 If you use a transaction in any request, and the request returns
846 with an operating system error or DB_LOCK_NOTGRANTED, the internal
847 TXN_DEADLOCK flag will be set on the transaction. See C<db_txn_finish>,
848 above.
358 849
359=head1 SEE ALSO 850=head1 SEE ALSO
360 851
361L<Coro::AIO>. 852L<AnyEvent::BDB> (event loop integration), L<Coro::BDB> (more natural
853syntax), L<IO::AIO> (nice to have).
362 854
363=head1 AUTHOR 855=head1 AUTHOR
364 856
365 Marc Lehmann <schmorp@schmorp.de> 857 Marc Lehmann <schmorp@schmorp.de>
366 http://home.schmorp.de/ 858 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines