python-peps/pep-0454.txt

PEP: 454
Title: Add a new tracemalloc module to trace Python memory allocations
Version: $Revision$
Last-Modified: $Date$
Author: Victor Stinner <victor.stinner@gmail.com>
Status: Draft
Type: Standards Track
Content-Type: text/x-rst
Created: 3-September-2013
Python-Version: 3.4


Abstract
========

Add a new tracemalloc module to trace Python memory allocations.



Rationale
=========

This PEP proposes to a new ``tracemalloc`` module, a debug tool to trace
memory allocations made by Python. The module provides the following
information:

* Statistics on allocations per Python line number (file and line):
  size, number, and average size of allocations
* Compute delta between two "snapshots"
* Location of a memory allocation: Python filename and line number

To trace the most Python memory allocations, the module should be
enabled as early as possible in your application by calling
``tracemalloc.enable()`` function, by setting the ``PYTHONTRACEMALLOC``
environment variable to ``1``, or  by using ``-X tracemalloc`` command
line option.


API
===

Functions
---------

enable():

   Start tracing Python memory allocations.

disable():

   Stop tracing Python memory allocations and stop the timer started by
   ``start_timer()``.

is_enabled():

  ``True`` if the module is enabled, ``False`` otherwise.

get_object_trace(obj):

   Get the trace of a Python object *obj* as a namedtuple with 3 attributes:

   - ``size``: size in bytes of the object
   - ``filename``: name of the Python script where the object was allocated
   - ``lineno``: line number where the object was allocated

   Return ``None`` if tracemalloc did not save the location where the object
   was allocated, for example if tracemalloc was disabled.

get_process_memory():

   Get the memory usage of the current process as a meminfo namedtuple with
   two attributes:

   * ``rss``: Resident Set Size in bytes
   * ``vms``: size of the virtual memory in bytes

   Return ``None`` if the platform is not supported.

   Use the ``psutil`` module if available.

start_timer(delay: int, func: callable, args: tuple=(), kwargs: dict={}):

   Start a timer calling ``func(*args, **kwargs)`` every *delay*
   seconds.

   The timer is based on the Python memory allocator, it is not real
   time.  *func* is called after at least *delay* seconds, it is not
   called exactly after *delay* seconds if no Python memory allocation
   occurred.

   If ``start_timer()`` is called twice, previous parameters are
   replaced.  The timer has a resolution of 1 second.

   `start_timer()`` is used by ``DisplayTop`` and ``TakeSnapshot``
   to run regulary a task.

stop_timer():

   Stop the timer started by ``start_timer()``.


DisplayTop
----------

DisplayTop(count: int, file=sys.stdout) class:

   Display the list of the N biggest memory allocations.

display():

      Display the top

start(delay: int):

      Start a task using tracemalloc timer to display the top every delay seconds.

stop():

      Stop the task started by the ``DisplayTop.start()`` method

color attribute:

      Use color if True (bool, default: stream.isatty()).

compare_with_previous attribute:

      If ``True``, compare with the previous top if ``True``. If ``False``,
      compare with the first one (bool, default: ``True``): .

filename_parts attribute:

      Number of displayed filename parts (int, default: ``3``).

show_average attribute:

      If ``True``, show the average size of allocations (bool, default: ``True``).

show_count attribute:

      If ``True``, show the number of allocations (bool, default: ``True``).

show_lineno attribute:

      If ``True``, use also the line number, not only the filename (bool, default:
      ``True``). If ``False``, only show the filename.

show_size attribute:

      if ``True``, show the size of allocations (bool, default: ``True``).

user_data_callback attribute:

      Optional callback collecting user data (callable, default:
      ``None``).  See ``Snapshot.create()``.


Snapshot
--------

Snapshot() class:

   Snapshot of Python memory allocations. Use ``TakeSnapshot`` to regulary
   take snapshots.

create(user_data_callback=None)

      Take a snapshot. If user_data_callback is specified, it must be a callable
      object returning a list of (title: str, format: str, value: int). format
      must be "size". The list must always have the same length and the same order
      to be able to compute differences between values.

      Example: ``[('Video memory', 'size', 234902)]``.

filter_filenames(patterns: str|list, include: bool)

      Remove filenames not matching any pattern if include is True, or remove
      filenames matching a pattern if include is False (exclude). See
      fnmatch.fnmatch() for the syntax of patterns.

write(filename)

      Write the snapshot into a file.

pid attribute:

      Identifier of the process which created the snapshot (int).

stats attribute:

      Raw memory allocation statistics (dict).

timestamp attribute:

      Date and time of the creation of the snapshot (str).


TakeSnapshot
------------

TakeSnapshot class:

   Task taking snapshots of Python memory allocations: write them into files.

start(delay: int)

      Start a task taking a snapshot every delay seconds.

stop():

      Stop the task started by the ``TakeSnapshot.start()`` method.

take_snapshot():

      Take a snapshot.

filename_template attribute:

      Template (str) to create a filename. "Variables" can be used in the
      template:

      * ``$pid``: identifier of the current process
      * ``$timestamp``: current date and time
      * ``$counter``: counter starting at 1 and incremented at each snapshot

user_data_callback attribute:

      Optional callback collecting user data (callable, default: None).
      See ``Snapshot.create()``.


Links
=====

* `Python issue #18874: Add a new tracemalloc module to trace Python
  memory allocations <http://bugs.python.org/issue18874>`_

Similar projects:

* `Meliae: Python Memory Usage Analyzer
  <https://pypi.python.org/pypi/meliae>`_
* `Issue #3329: API for setting the memory allocator used by Python
  <http://bugs.python.org/issue3329>`_
* `Guppy-PE: umbrella package combining Heapy and GSL
  <http://guppy-pe.sourceforge.net/>`_
* `PySizer <http://pysizer.8325.org/>`_: developed for Python 2.4
* `memory_profiler <https://pypi.python.org/pypi/memory_profiler>`_
* `pympler <http://code.google.com/p/pympler/>`_
* `Dozer <https://pypi.python.org/pypi/Dozer>`_: WSGI Middleware version of
  the CherryPy memory leak debugger
* `objgraph <http://mg.pov.lt/objgraph/>`_
* `caulk <https://github.com/smartfile/caulk/>`_

Copyright
=========

This document has been placed into the public domain.