2005-01-19 14:14:38 +01:00
|
|
|
###
|
2005-01-19 14:33:05 +01:00
|
|
|
# Copyright (c) 2002-2005, Jeremiah Fincher
|
2021-08-01 21:47:37 +02:00
|
|
|
# Copyright (c) 2010-2021, The Limnoria Contributors
|
2005-01-19 14:14:38 +01:00
|
|
|
# All rights reserved.
|
|
|
|
#
|
|
|
|
# Redistribution and use in source and binary forms, with or without
|
|
|
|
# modification, are permitted provided that the following conditions are met:
|
|
|
|
#
|
|
|
|
# * Redistributions of source code must retain the above copyright notice,
|
|
|
|
# this list of conditions, and the following disclaimer.
|
|
|
|
# * 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.
|
|
|
|
# * Neither the name of the author of this software nor the name of
|
|
|
|
# contributors to this software may be used to endorse or promote products
|
|
|
|
# derived from this software without specific prior written consent.
|
|
|
|
#
|
|
|
|
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS 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 THE COPYRIGHT OWNER 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.
|
|
|
|
###
|
|
|
|
|
|
|
|
"""
|
|
|
|
Module for some slight database-independence for simple databases.
|
|
|
|
"""
|
|
|
|
|
2012-05-02 20:48:49 +02:00
|
|
|
import os
|
2005-01-19 14:14:38 +01:00
|
|
|
import csv
|
|
|
|
import math
|
|
|
|
|
2015-08-11 16:50:23 +02:00
|
|
|
from . import cdb, utils
|
|
|
|
from .utils import minisix
|
2012-09-18 04:12:11 +02:00
|
|
|
from .utils.iter import ilen
|
2005-01-19 14:14:38 +01:00
|
|
|
|
|
|
|
class Error(Exception):
|
|
|
|
"""General error for this module."""
|
|
|
|
|
|
|
|
class NoRecordError(KeyError):
|
|
|
|
pass
|
|
|
|
|
|
|
|
class InvalidDBError(Exception):
|
|
|
|
pass
|
|
|
|
|
|
|
|
class MappingInterface(object):
|
|
|
|
"""This is a class to represent the underlying representation of a map
|
|
|
|
from integer keys to strings."""
|
|
|
|
def __init__(self, filename, **kwargs):
|
|
|
|
"""Feel free to ignore the filename."""
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def get(id):
|
|
|
|
"""Gets the record matching id. Raises NoRecordError otherwise."""
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def set(id, s):
|
|
|
|
"""Sets the record matching id to s."""
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def add(self, s):
|
|
|
|
"""Adds a new record, returning a new id for it."""
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def remove(self, id):
|
|
|
|
"Returns and removes the record with the given id from the database."
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def __iter__(self):
|
|
|
|
"Return an iterator over (id, s) pairs. Not required to be ordered."
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def flush(self):
|
|
|
|
"""Flushes current state to disk."""
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def close(self):
|
|
|
|
"""Flushes current state to disk and invalidates the Mapping."""
|
|
|
|
raise NotImplementedError
|
|
|
|
|
|
|
|
def vacuum(self):
|
|
|
|
"Cleans up in the database, if possible. Not required to do anything."
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
class DirMapping(MappingInterface):
|
|
|
|
def __init__(self, filename, **kwargs):
|
|
|
|
self.dirname = filename
|
|
|
|
if not os.path.exists(self.dirname):
|
|
|
|
os.mkdir(self.dirname)
|
|
|
|
if not os.path.exists(os.path.join(self.dirname, 'max')):
|
|
|
|
self._setMax(1)
|
|
|
|
|
|
|
|
def _setMax(self, id):
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(os.path.join(self.dirname, 'max'), 'w')
|
2005-01-19 14:14:38 +01:00
|
|
|
try:
|
|
|
|
fd.write(str(id))
|
|
|
|
finally:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def _getMax(self):
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(os.path.join(self.dirname, 'max'))
|
2005-01-19 14:14:38 +01:00
|
|
|
try:
|
|
|
|
i = int(fd.read())
|
|
|
|
return i
|
|
|
|
finally:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def _makeFilename(self, id):
|
|
|
|
return os.path.join(self.dirname, str(id))
|
|
|
|
|
2012-05-02 20:50:06 +02:00
|
|
|
def get(self, id):
|
2005-01-19 14:14:38 +01:00
|
|
|
try:
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self._makeFilename(id))
|
2005-01-19 14:14:38 +01:00
|
|
|
return fd.read()
|
2014-01-20 15:49:15 +01:00
|
|
|
except EnvironmentError as e:
|
2005-01-19 14:14:38 +01:00
|
|
|
exn = NoRecordError(id)
|
|
|
|
exn.realException = e
|
|
|
|
raise exn
|
2014-01-03 17:44:01 +01:00
|
|
|
finally:
|
|
|
|
fd.close()
|
2005-01-19 14:14:38 +01:00
|
|
|
|
2012-05-02 20:50:06 +02:00
|
|
|
def set(self, id, s):
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self._makeFilename(id), 'w')
|
2005-01-19 14:14:38 +01:00
|
|
|
fd.write(s)
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def add(self, s):
|
|
|
|
id = self._getMax()
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self._makeFilename(id), 'w')
|
2005-01-19 14:14:38 +01:00
|
|
|
try:
|
|
|
|
fd.write(s)
|
|
|
|
return id
|
|
|
|
finally:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def remove(self, id):
|
|
|
|
try:
|
|
|
|
os.remove(self._makeFilename(id))
|
2015-08-31 15:38:35 +02:00
|
|
|
except EnvironmentError:
|
2014-01-20 15:43:55 +01:00
|
|
|
raise NoRecordError(id)
|
2005-01-19 14:14:38 +01:00
|
|
|
|
|
|
|
class FlatfileMapping(MappingInterface):
|
|
|
|
def __init__(self, filename, maxSize=10**6):
|
|
|
|
self.filename = filename
|
|
|
|
try:
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename)
|
2005-01-19 14:14:38 +01:00
|
|
|
strId = fd.readline().rstrip()
|
|
|
|
self.maxSize = len(strId)
|
|
|
|
try:
|
|
|
|
self.currentId = int(strId)
|
|
|
|
except ValueError:
|
2014-01-20 15:43:55 +01:00
|
|
|
raise Error('Invalid file for FlatfileMapping: %s' % filename)
|
2014-01-20 15:49:15 +01:00
|
|
|
except EnvironmentError as e:
|
2005-01-19 14:14:38 +01:00
|
|
|
# File couldn't be opened.
|
|
|
|
self.maxSize = int(math.log10(maxSize))
|
|
|
|
self.currentId = 0
|
|
|
|
self._incrementCurrentId()
|
2014-01-03 17:44:01 +01:00
|
|
|
finally:
|
2014-01-03 18:48:30 +01:00
|
|
|
if 'fd' in locals():
|
|
|
|
fd.close()
|
2005-01-19 14:14:38 +01:00
|
|
|
|
|
|
|
def _canonicalId(self, id):
|
|
|
|
if id is not None:
|
|
|
|
return str(id).zfill(self.maxSize)
|
|
|
|
else:
|
|
|
|
return '-'*self.maxSize
|
|
|
|
|
|
|
|
def _incrementCurrentId(self, fd=None):
|
|
|
|
fdWasNone = fd is None
|
|
|
|
if fdWasNone:
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename, 'a')
|
2005-01-19 14:14:38 +01:00
|
|
|
fd.seek(0)
|
|
|
|
self.currentId += 1
|
|
|
|
fd.write(self._canonicalId(self.currentId))
|
|
|
|
fd.write('\n')
|
|
|
|
if fdWasNone:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def _splitLine(self, line):
|
|
|
|
line = line.rstrip('\r\n')
|
|
|
|
(id, s) = line.split(':', 1)
|
|
|
|
return (id, s)
|
|
|
|
|
|
|
|
def _joinLine(self, id, s):
|
|
|
|
return '%s:%s\n' % (self._canonicalId(id), s)
|
|
|
|
|
|
|
|
def add(self, s):
|
|
|
|
line = self._joinLine(self.currentId, s)
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename, 'r+')
|
2005-01-19 14:14:38 +01:00
|
|
|
try:
|
|
|
|
fd.seek(0, 2) # End.
|
|
|
|
fd.write(line)
|
|
|
|
return self.currentId
|
|
|
|
finally:
|
|
|
|
self._incrementCurrentId(fd)
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def get(self, id):
|
|
|
|
strId = self._canonicalId(id)
|
|
|
|
try:
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename)
|
2005-01-19 14:14:38 +01:00
|
|
|
fd.readline() # First line, nextId.
|
|
|
|
for line in fd:
|
|
|
|
(lineId, s) = self._splitLine(line)
|
|
|
|
if lineId == strId:
|
|
|
|
return s
|
2014-01-20 15:43:55 +01:00
|
|
|
raise NoRecordError(id)
|
2005-01-19 14:14:38 +01:00
|
|
|
finally:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
# XXX This assumes it's not been given out. We should make sure that our
|
|
|
|
# maximum id remains accurate if this is some value we've never given
|
|
|
|
# out -- i.e., self.maxid = max(self.maxid, id) or something.
|
|
|
|
def set(self, id, s):
|
|
|
|
strLine = self._joinLine(id, s)
|
|
|
|
try:
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename, 'r+')
|
2005-01-19 14:14:38 +01:00
|
|
|
self.remove(id, fd)
|
|
|
|
fd.seek(0, 2) # End.
|
|
|
|
fd.write(strLine)
|
|
|
|
finally:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def remove(self, id, fd=None):
|
|
|
|
fdWasNone = fd is None
|
|
|
|
strId = self._canonicalId(id)
|
|
|
|
try:
|
|
|
|
if fdWasNone:
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename, 'r+')
|
2005-01-19 14:14:38 +01:00
|
|
|
fd.seek(0)
|
|
|
|
fd.readline() # First line, nextId
|
|
|
|
pos = fd.tell()
|
|
|
|
line = fd.readline()
|
|
|
|
while line:
|
|
|
|
(lineId, _) = self._splitLine(line)
|
|
|
|
if lineId == strId:
|
|
|
|
fd.seek(pos)
|
|
|
|
fd.write(self._canonicalId(None))
|
|
|
|
fd.seek(pos)
|
|
|
|
fd.readline() # Same line we just rewrote the id for.
|
|
|
|
pos = fd.tell()
|
|
|
|
line = fd.readline()
|
|
|
|
# We should be at the end.
|
|
|
|
finally:
|
|
|
|
if fdWasNone:
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def __iter__(self):
|
2012-08-05 16:56:07 +02:00
|
|
|
fd = open(self.filename)
|
2005-01-19 14:14:38 +01:00
|
|
|
fd.readline() # First line, nextId.
|
|
|
|
for line in fd:
|
|
|
|
(id, s) = self._splitLine(line)
|
|
|
|
if not id.startswith('-'):
|
|
|
|
yield (int(id), s)
|
|
|
|
fd.close()
|
|
|
|
|
|
|
|
def vacuum(self):
|
2012-08-05 16:56:07 +02:00
|
|
|
infd = open(self.filename)
|
2005-02-01 11:13:23 +01:00
|
|
|
outfd = utils.file.AtomicFile(self.filename,makeBackupIfSmaller=False)
|
2005-01-19 14:14:38 +01:00
|
|
|
outfd.write(infd.readline()) # First line, nextId.
|
|
|
|
for line in infd:
|
|
|
|
if not line.startswith('-'):
|
|
|
|
outfd.write(line)
|
|
|
|
infd.close()
|
|
|
|
outfd.close()
|
|
|
|
|
|
|
|
def flush(self):
|
|
|
|
pass # No-op, we maintain no open files.
|
|
|
|
|
|
|
|
def close(self):
|
|
|
|
self.vacuum() # Should we do this? It should be fine.
|
|
|
|
|
|
|
|
|
|
|
|
class CdbMapping(MappingInterface):
|
|
|
|
def __init__(self, filename, **kwargs):
|
|
|
|
self.filename = filename
|
|
|
|
self._openCdb() # So it can be overridden later.
|
|
|
|
if 'nextId' not in self.db:
|
|
|
|
self.db['nextId'] = '1'
|
|
|
|
|
|
|
|
def _openCdb(self, *args, **kwargs):
|
2012-08-05 15:47:30 +02:00
|
|
|
self.db = cdb.open_db(self.filename, 'c', **kwargs)
|
2005-01-19 14:14:38 +01:00
|
|
|
|
|
|
|
def _getNextId(self):
|
|
|
|
i = int(self.db['nextId'])
|
|
|
|
self.db['nextId'] = str(i+1)
|
|
|
|
return i
|
|
|
|
|
|
|
|
def get(self, id):
|
|
|
|
try:
|
|
|
|
return self.db[str(id)]
|
|
|
|
except KeyError:
|
2014-01-20 15:43:55 +01:00
|
|
|
raise NoRecordError(id)
|
2005-01-19 14:14:38 +01:00
|
|
|
|
|
|
|
# XXX Same as above.
|
|
|
|
def set(self, id, s):
|
|
|
|
self.db[str(id)] = s
|
|
|
|
|
|
|
|
def add(self, s):
|
|
|
|
id = self._getNextId()
|
|
|
|
self.set(id, s)
|
|
|
|
return id
|
|
|
|
|
|
|
|
def remove(self, id):
|
|
|
|
del self.db[str(id)]
|
|
|
|
|
|
|
|
def __iter__(self):
|
2015-08-08 22:20:14 +02:00
|
|
|
for (id, s) in self.db.items():
|
2005-01-19 14:14:38 +01:00
|
|
|
if id != 'nextId':
|
|
|
|
yield (int(id), s)
|
|
|
|
|
|
|
|
def flush(self):
|
|
|
|
self.db.flush()
|
|
|
|
|
|
|
|
def close(self):
|
|
|
|
self.db.close()
|
|
|
|
|
|
|
|
|
|
|
|
class DB(object):
|
|
|
|
Mapping = 'flat' # This is a good, sane default.
|
|
|
|
Record = None
|
|
|
|
def __init__(self, filename, Mapping=None, Record=None):
|
|
|
|
if Record is not None:
|
|
|
|
self.Record = Record
|
|
|
|
if Mapping is not None:
|
|
|
|
self.Mapping = Mapping
|
2015-08-10 20:24:11 +02:00
|
|
|
if isinstance(self.Mapping, minisix.string_types):
|
2005-01-19 14:14:38 +01:00
|
|
|
self.Mapping = Mappings[self.Mapping]
|
|
|
|
self.map = self.Mapping(filename)
|
|
|
|
|
|
|
|
def _newRecord(self, id, s):
|
|
|
|
record = self.Record(id=id)
|
|
|
|
record.deserialize(s)
|
|
|
|
return record
|
|
|
|
|
|
|
|
def get(self, id):
|
|
|
|
s = self.map.get(id)
|
|
|
|
return self._newRecord(id, s)
|
|
|
|
|
|
|
|
def set(self, id, record):
|
|
|
|
s = record.serialize()
|
|
|
|
self.map.set(id, s)
|
|
|
|
|
|
|
|
def add(self, record):
|
|
|
|
s = record.serialize()
|
|
|
|
id = self.map.add(s)
|
|
|
|
record.id = id
|
|
|
|
return id
|
|
|
|
|
|
|
|
def remove(self, id):
|
|
|
|
self.map.remove(id)
|
|
|
|
|
|
|
|
def __iter__(self):
|
|
|
|
for (id, s) in self.map:
|
|
|
|
# We don't need to yield the id because it's in the record.
|
|
|
|
yield self._newRecord(id, s)
|
|
|
|
|
|
|
|
def select(self, p):
|
|
|
|
for record in self:
|
|
|
|
if p(record):
|
|
|
|
yield record
|
|
|
|
|
|
|
|
def random(self):
|
|
|
|
try:
|
2005-01-31 16:22:48 +01:00
|
|
|
return self._newRecord(*utils.iter.choice(self.map))
|
2005-01-19 14:14:38 +01:00
|
|
|
except IndexError:
|
|
|
|
return None
|
|
|
|
|
|
|
|
def size(self):
|
|
|
|
return ilen(self.map)
|
|
|
|
|
|
|
|
def flush(self):
|
|
|
|
self.map.flush()
|
|
|
|
|
|
|
|
def vacuum(self):
|
|
|
|
self.map.vacuum()
|
|
|
|
|
|
|
|
def close(self):
|
|
|
|
self.map.close()
|
|
|
|
|
|
|
|
Mappings = {
|
|
|
|
'cdb': CdbMapping,
|
|
|
|
'flat': FlatfileMapping,
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
class Record(object):
|
|
|
|
def __init__(self, id=None, **kwargs):
|
|
|
|
if id is not None:
|
|
|
|
assert isinstance(id, int), 'id must be an integer.'
|
|
|
|
self.id = id
|
|
|
|
self.fields = []
|
|
|
|
self.defaults = {}
|
|
|
|
self.converters = {}
|
|
|
|
for name in self.__fields__:
|
|
|
|
if isinstance(name, tuple):
|
|
|
|
(name, spec) = name
|
|
|
|
else:
|
|
|
|
spec = utils.safeEval
|
|
|
|
assert name != 'id'
|
|
|
|
self.fields.append(name)
|
|
|
|
if isinstance(spec, tuple):
|
|
|
|
(converter, default) = spec
|
|
|
|
else:
|
|
|
|
converter = spec
|
|
|
|
default = None
|
|
|
|
self.defaults[name] = default
|
|
|
|
self.converters[name] = converter
|
2005-01-31 16:24:36 +01:00
|
|
|
seen = set()
|
2015-08-08 22:20:14 +02:00
|
|
|
for (name, value) in kwargs.items():
|
2005-01-19 14:14:38 +01:00
|
|
|
assert name in self.fields, 'name must be a record value.'
|
|
|
|
seen.add(name)
|
|
|
|
setattr(self, name, value)
|
|
|
|
for name in self.fields:
|
|
|
|
if name not in seen:
|
|
|
|
default = self.defaults[name]
|
|
|
|
if callable(default):
|
|
|
|
default = default()
|
|
|
|
setattr(self, name, default)
|
|
|
|
|
|
|
|
def serialize(self):
|
|
|
|
return csv.join([repr(getattr(self, name)) for name in self.fields])
|
|
|
|
|
|
|
|
def deserialize(self, s):
|
2005-01-31 16:24:36 +01:00
|
|
|
unseenRecords = set(self.fields)
|
2005-01-19 14:14:38 +01:00
|
|
|
for (name, strValue) in zip(self.fields, csv.split(s)):
|
|
|
|
setattr(self, name, self.converters[name](strValue))
|
|
|
|
unseenRecords.remove(name)
|
|
|
|
for name in unseenRecords:
|
|
|
|
setattr(self, name, self.defaults[name])
|
|
|
|
|
|
|
|
|
2006-02-11 16:52:51 +01:00
|
|
|
# vim:set shiftwidth=4 softtabstop=4 expandtab textwidth=79:
|