ref: 3ec84c5d7364c8ed4602561743bf35e3b399e53d
dir: /sys/lib/python/bsddb/__init__.py/
#---------------------------------------------------------------------- # Copyright (c) 1999-2001, Digital Creations, Fredericksburg, VA, USA # and Andrew Kuchling. All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are # met: # # o Redistributions of source code must retain the above copyright # notice, this list of conditions, and the disclaimer that follows. # # o Redistributions in binary form must reproduce the above copyright # notice, this list of conditions, and the following disclaimer in # the documentation and/or other materials provided with the # distribution. # # o Neither the name of Digital Creations nor the names of its # contributors may be used to endorse or promote products derived # from this software without specific prior written permission. # # THIS SOFTWARE IS PROVIDED BY DIGITAL CREATIONS AND CONTRIBUTORS *AS # IS* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED # TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A # PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL DIGITAL # CREATIONS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, # BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS # OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND # ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR # TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE # USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH # DAMAGE. #---------------------------------------------------------------------- """Support for BerkeleyDB 3.3 through 4.4 with a simple interface. For the full featured object oriented interface use the bsddb.db module instead. It mirrors the Sleepycat BerkeleyDB C API. """ try: if __name__ == 'bsddb3': # import _pybsddb binary as it should be the more recent version from # a standalone pybsddb addon package than the version included with # python as bsddb._bsddb. import _pybsddb _bsddb = _pybsddb from bsddb3.dbutils import DeadlockWrap as _DeadlockWrap else: import _bsddb from bsddb.dbutils import DeadlockWrap as _DeadlockWrap except ImportError: # Remove ourselves from sys.modules import sys del sys.modules[__name__] raise # bsddb3 calls it db, but provide _db for backwards compatibility db = _db = _bsddb __version__ = db.__version__ error = db.DBError # So bsddb.error will mean something... #---------------------------------------------------------------------- import sys, os # for backwards compatibility with python versions older than 2.3, the # iterator interface is dynamically defined and added using a mixin # class. old python can't tokenize it due to the yield keyword. if sys.version >= '2.3': import UserDict from weakref import ref exec """ class _iter_mixin(UserDict.DictMixin): def _make_iter_cursor(self): cur = _DeadlockWrap(self.db.cursor) key = id(cur) self._cursor_refs[key] = ref(cur, self._gen_cref_cleaner(key)) return cur def _gen_cref_cleaner(self, key): # use generate the function for the weakref callback here # to ensure that we do not hold a strict reference to cur # in the callback. return lambda ref: self._cursor_refs.pop(key, None) def __iter__(self): try: cur = self._make_iter_cursor() # FIXME-20031102-greg: race condition. cursor could # be closed by another thread before this call. # since we're only returning keys, we call the cursor # methods with flags=0, dlen=0, dofs=0 key = _DeadlockWrap(cur.first, 0,0,0)[0] yield key next = cur.next while 1: try: key = _DeadlockWrap(next, 0,0,0)[0] yield key except _bsddb.DBCursorClosedError: cur = self._make_iter_cursor() # FIXME-20031101-greg: race condition. cursor could # be closed by another thread before this call. _DeadlockWrap(cur.set, key,0,0,0) next = cur.next except _bsddb.DBNotFoundError: return except _bsddb.DBCursorClosedError: # the database was modified during iteration. abort. return def iteritems(self): if not self.db: return try: cur = self._make_iter_cursor() # FIXME-20031102-greg: race condition. cursor could # be closed by another thread before this call. kv = _DeadlockWrap(cur.first) key = kv[0] yield kv next = cur.next while 1: try: kv = _DeadlockWrap(next) key = kv[0] yield kv except _bsddb.DBCursorClosedError: cur = self._make_iter_cursor() # FIXME-20031101-greg: race condition. cursor could # be closed by another thread before this call. _DeadlockWrap(cur.set, key,0,0,0) next = cur.next except _bsddb.DBNotFoundError: return except _bsddb.DBCursorClosedError: # the database was modified during iteration. abort. return """ else: class _iter_mixin: pass class _DBWithCursor(_iter_mixin): """ A simple wrapper around DB that makes it look like the bsddbobject in the old module. It uses a cursor as needed to provide DB traversal. """ def __init__(self, db): self.db = db self.db.set_get_returns_none(0) # FIXME-20031101-greg: I believe there is still the potential # for deadlocks in a multithreaded environment if someone # attempts to use the any of the cursor interfaces in one # thread while doing a put or delete in another thread. The # reason is that _checkCursor and _closeCursors are not atomic # operations. Doing our own locking around self.dbc, # self.saved_dbc_key and self._cursor_refs could prevent this. # TODO: A test case demonstrating the problem needs to be written. # self.dbc is a DBCursor object used to implement the # first/next/previous/last/set_location methods. self.dbc = None self.saved_dbc_key = None # a collection of all DBCursor objects currently allocated # by the _iter_mixin interface. self._cursor_refs = {} def __del__(self): self.close() def _checkCursor(self): if self.dbc is None: self.dbc = _DeadlockWrap(self.db.cursor) if self.saved_dbc_key is not None: _DeadlockWrap(self.dbc.set, self.saved_dbc_key) self.saved_dbc_key = None # This method is needed for all non-cursor DB calls to avoid # BerkeleyDB deadlocks (due to being opened with DB_INIT_LOCK # and DB_THREAD to be thread safe) when intermixing database # operations that use the cursor internally with those that don't. def _closeCursors(self, save=1): if self.dbc: c = self.dbc self.dbc = None if save: try: self.saved_dbc_key = _DeadlockWrap(c.current, 0,0,0)[0] except db.DBError: pass _DeadlockWrap(c.close) del c for cref in self._cursor_refs.values(): c = cref() if c is not None: _DeadlockWrap(c.close) def _checkOpen(self): if self.db is None: raise error, "BSDDB object has already been closed" def isOpen(self): return self.db is not None def __len__(self): self._checkOpen() return _DeadlockWrap(lambda: len(self.db)) # len(self.db) def __getitem__(self, key): self._checkOpen() return _DeadlockWrap(lambda: self.db[key]) # self.db[key] def __setitem__(self, key, value): self._checkOpen() self._closeCursors() def wrapF(): self.db[key] = value _DeadlockWrap(wrapF) # self.db[key] = value def __delitem__(self, key): self._checkOpen() self._closeCursors() def wrapF(): del self.db[key] _DeadlockWrap(wrapF) # del self.db[key] def close(self): self._closeCursors(save=0) if self.dbc is not None: _DeadlockWrap(self.dbc.close) v = 0 if self.db is not None: v = _DeadlockWrap(self.db.close) self.dbc = None self.db = None return v def keys(self): self._checkOpen() return _DeadlockWrap(self.db.keys) def has_key(self, key): self._checkOpen() return _DeadlockWrap(self.db.has_key, key) def set_location(self, key): self._checkOpen() self._checkCursor() return _DeadlockWrap(self.dbc.set_range, key) def next(self): self._checkOpen() self._checkCursor() rv = _DeadlockWrap(self.dbc.next) return rv def previous(self): self._checkOpen() self._checkCursor() rv = _DeadlockWrap(self.dbc.prev) return rv def first(self): self._checkOpen() self._checkCursor() rv = _DeadlockWrap(self.dbc.first) return rv def last(self): self._checkOpen() self._checkCursor() rv = _DeadlockWrap(self.dbc.last) return rv def sync(self): self._checkOpen() return _DeadlockWrap(self.db.sync) #---------------------------------------------------------------------- # Compatibility object factory functions def hashopen(file, flag='c', mode=0666, pgsize=None, ffactor=None, nelem=None, cachesize=None, lorder=None, hflags=0): flags = _checkflag(flag, file) e = _openDBEnv(cachesize) d = db.DB(e) d.set_flags(hflags) if pgsize is not None: d.set_pagesize(pgsize) if lorder is not None: d.set_lorder(lorder) if ffactor is not None: d.set_h_ffactor(ffactor) if nelem is not None: d.set_h_nelem(nelem) d.open(file, db.DB_HASH, flags, mode) return _DBWithCursor(d) #---------------------------------------------------------------------- def btopen(file, flag='c', mode=0666, btflags=0, cachesize=None, maxkeypage=None, minkeypage=None, pgsize=None, lorder=None): flags = _checkflag(flag, file) e = _openDBEnv(cachesize) d = db.DB(e) if pgsize is not None: d.set_pagesize(pgsize) if lorder is not None: d.set_lorder(lorder) d.set_flags(btflags) if minkeypage is not None: d.set_bt_minkey(minkeypage) if maxkeypage is not None: d.set_bt_maxkey(maxkeypage) d.open(file, db.DB_BTREE, flags, mode) return _DBWithCursor(d) #---------------------------------------------------------------------- def rnopen(file, flag='c', mode=0666, rnflags=0, cachesize=None, pgsize=None, lorder=None, rlen=None, delim=None, source=None, pad=None): flags = _checkflag(flag, file) e = _openDBEnv(cachesize) d = db.DB(e) if pgsize is not None: d.set_pagesize(pgsize) if lorder is not None: d.set_lorder(lorder) d.set_flags(rnflags) if delim is not None: d.set_re_delim(delim) if rlen is not None: d.set_re_len(rlen) if source is not None: d.set_re_source(source) if pad is not None: d.set_re_pad(pad) d.open(file, db.DB_RECNO, flags, mode) return _DBWithCursor(d) #---------------------------------------------------------------------- def _openDBEnv(cachesize): e = db.DBEnv() if cachesize is not None: if cachesize >= 20480: e.set_cachesize(0, cachesize) else: raise error, "cachesize must be >= 20480" e.set_lk_detect(db.DB_LOCK_DEFAULT) e.open('.', db.DB_PRIVATE | db.DB_CREATE | db.DB_THREAD | db.DB_INIT_LOCK | db.DB_INIT_MPOOL) return e def _checkflag(flag, file): if flag == 'r': flags = db.DB_RDONLY elif flag == 'rw': flags = 0 elif flag == 'w': flags = db.DB_CREATE elif flag == 'c': flags = db.DB_CREATE elif flag == 'n': flags = db.DB_CREATE #flags = db.DB_CREATE | db.DB_TRUNCATE # we used db.DB_TRUNCATE flag for this before but BerkeleyDB # 4.2.52 changed to disallowed truncate with txn environments. if file is not None and os.path.isfile(file): os.unlink(file) else: raise error, "flags should be one of 'r', 'w', 'c' or 'n'" return flags | db.DB_THREAD #---------------------------------------------------------------------- # This is a silly little hack that allows apps to continue to use the # DB_THREAD flag even on systems without threads without freaking out # BerkeleyDB. # # This assumes that if Python was built with thread support then # BerkeleyDB was too. try: import thread del thread if db.version() < (3, 3, 0): db.DB_THREAD = 0 except ImportError: db.DB_THREAD = 0 #----------------------------------------------------------------------