entropy.dump

Author: Fabio Erculiani <lxnay@sabayonlinux.org>

Contact: lxnay@sabayonlinux.org

Copyright: Fabio Erculiani

License: GPL-2

Entropy Framework object disk serializer module.

This module contains Entropy Python object serialization functions and disk dumpers.

Serialized objects are stored to disk with proper permissions by default into path given by entropy.const's etpConst['dumpstoragedir'].

Permissions are set using entropy.const's const_setup_perms and const_setup_file functions.

Objects are serialized using Python's cPickle/pickle modules, thus they must be "pickable". Please read Python Library reference for more information.

dumpobj(name, my_object, complete_path=False, ignore_exceptions=True)

source code

Dump pickable object to file

Parameters:

name (string) - name of the object
my_object (any Python "pickable" object) - object to dump
complete_path (bool) - consider "name" argument as a complete path (this overrides the default dump path given by etpConst['dumpstoragedir'])
ignore_exceptions (bool) - ignore any possible exception (EOFError, IOError, OSError,)

Returns: None

None

Raises:

EOFError - could be caused by pickle.dump, ignored if ignore_exceptions is True
IOError - could be caused by pickle.dump, ignored if ignore_exceptions is True
OSError - could be caused by pickle.dump, ignored if ignore_exceptions is True

serialize(myobj, ser_f, do_seek=True)

source code

Serialize object to ser_f (file)

Parameters:

myobj (any Python picklable object) - Python object to serialize
ser_f (file object) - file object to write to
do_seek (bool) - move file cursor back to the beginning of ser_f

Returns: file object

file object where data has been written

Raises:

RuntimeError - caused by pickle.dump in case of system errors
EOFError - caused by pickle.dump in case of race conditions on multi-processing or multi-threading
IOError - caused by pickle.dump in case of race conditions on multi-processing or multi-threading
pickle.PicklingError - when object cannot be recreated

unserialize(serial_f)

source code

Unserialize file to object (file)

Parameters:

serial_f (file object) - file object which data will be read from

Returns: any Python pickable object

rebuilt object

Raises:

pickle.UnpicklingError - when object cannot be recreated

unserialize_string(mystring)

source code

Unserialize pickle string to object

Parameters:

mystring (string) - data stream in string form to reconstruct

Returns: any Python pickable object

reconstructed object

Raises:

pickle.UnpicklingError - when object cannot be recreated

serialize_string(myobj)

source code

Serialize object to string

Parameters:

myobj (any Python picklable object) - object to serialize

Returns: string

serialized string

Raises:

pickle.PicklingError - when object cannot be recreated

loadobj(name, complete_path=False)

source code

Load object from a file

Parameters:

name (string) - name of the object to load
complete_path (bool) - determine whether name argument is a complete disk path to serialized object

Returns: any Python pickable object or None

object or None

getobjmtime(name)

source code

Get dumped object mtime

Parameters:

name (string) - object name

Returns: int

mtime of the file containing the serialized object or 0 if not found

removeobj(name)

source code

Remove cached object referenced by its object name

Parameters:

name (string) - object name

Returns: bool

bool representing whether object has been removed or not

Raises:

OSError - in case of troubles with os.remove()

Variables
	D_EXT = `'.dmp'`
	D_DIR = `'/var/lib/entropy/caches/'`
	E_GID = `1000`

Module dump

dumpobj(name, my_object, complete_path=False, ignore_exceptions=True)

serialize(myobj, ser_f, do_seek=True)

unserialize(serial_f)

unserialize_string(mystring)

serialize_string(myobj)

loadobj(name, complete_path=False)

getobjmtime(name)

removeobj(name)