namedlist 1.6

Specifying Fields

Fields in namedlist.namedlist or namedlist.namedtuple can be specified as in collections.namedtuple: as either a string specifing the field names, or as a iterable of field names. These two uses are equivalent:

>>> Point = namedlist('Point', 'x y')
>>> Point = namedlist('Point', ['x', 'y'])

As are these:

>>> Point = namedtuple('Point', 'x y')
>>> Point = namedtuple('Point', ['x', 'y'])

If using a string, commas are first converted to spaces. So these are equivalent:

>>> Point = namedlist('Point', 'x y')
>>> Point = namedlist('Point', 'x,y')

Specifying Defaults

Per-field defaults can be specified by supplying a 2-tuple (name, default_value) instead of just a string for the field name. This is only supported when you specify a list of field names:

>>> Point = namedlist('Point', [('x', 0), ('y', 0)])
>>> p = Point(3)
>>> assert p.x == 3
>>> assert p.y == 0

Or, using namedtuple:

>>> Point = namedtuple('Point', [('x', 0), ('y', 0)])
>>> p = Point(3)
>>> assert p.x == 3
>>> assert p.y == 0

In addition to, or instead of, these per-field defaults, you can also specify a default value which is used when no per-field default value is specified:

>>> Point = namedlist('Point', 'x y z', default=0)
>>> p = Point(y=3)
>>> assert p.x == 0
>>> assert p.y == 3
>>> assert p.z == 0

>>> Point = namedlist('Point', [('x', 0), 'y', ('z', 0)], default=4)
>>> p = Point(z=2)
>>> assert p.x == 0
>>> assert p.y == 4
>>> assert p.z == 2

In addition to supplying the field names as an iterable of 2-tuples, you can also specify a mapping. The keys will be the field names, and the values will be the per-field default values. This is most useful with an OrderedDict, as the order of the fields will then be deterministic. The module variable NO_DEFAULT can be specified if you want a field to use the per-type default value instead of specifying it with a field:

>>> Point = namedlist('Point', OrderedDict((('x', 0),
...                                         ('y', NO_DEFAULT),
...                                         ('z', 0),
...                                         )),
...                            default=4)
>>> p = Point(z=2)
>>> assert p.x == 0
>>> assert p.y == 4
>>> assert p.z == 2

Writing to values

Instances of the classes generated by namedlist.namedlist are fully writable, unlike the tuple-derived classes returned by collections.namedtuple or namedlist.namedtuple:

>>> Point = namedlist('Point', 'x y')
>>> p = Point(1, 2)
>>> p.y = 4
>>> assert p.x == 1
>>> assert p.y == 4

Specifying __slots__

For namedlist.namedlist, by default, the returned class sets __slots__ which is initialized to the field names. While this decreases memory usage by eliminating the instance dict, it also means that you cannot create new instance members.

To change this behavior, specify use_slots=False when creating the namedlist:

>>> Point = namedlist('Point', 'x y', use_slots=False)
>>> p = Point(0, 1)
>>> p.z = 2
>>> assert p.x == 0
>>> assert p.y == 1
>>> assert p.z == 2

However, note that this method does not add the new variable to _fields, so it is not recognized when iterating over the instance:

>>> list(p)
[0, 1]
>>> sorted(p._asdict().items())
[('x', 0), ('y', 1)]

Additional class members

namedlist.namedlist and namedlist.namedtuple classes contain these members:

• _asdict(): Returns a dict which maps field names to their corresponding values.

• _fields: Tuple of strings listing the field names. Useful for introspection.

Renaming invalid field names

This functionality is identical to collections.namedtuple. If you specify rename=True, then any invalid field names are changed to _0, _1, etc. Reasons for a field name to be invalid are:

• Zero length strings.

• Containing characters other than alphanumerics and underscores.

• A conflict with a Python reserved identifier.

• Beginning with a digit.

• Beginning with an underscore.

• Using the same field name more than once.

For example:

>>> Point = namedlist('Point', 'x x for', rename=True)
>>> assert Point._fields == ('x', '_1', '_2')

Mutable default values

For namedlist.namelist, be aware of specifying mutable default values. Due to the way Python handles default values, each instance of a namedlist will share the default. This is especially problematic with default values that are lists. For example:

>>> A = namedlist('A', [('x', [])])
>>> a = A()
>>> a.x.append(4)
>>> b = A()
>>> assert b.x == [4]

This is probably not the desired behavior, so see the next section.

Specifying a factory function for default values

For namedlist.namedlist, you can supply a zero-argument callable for a default, by wrapping it in a FACTORY call. The only change in this example is to change the default from [] to FACTORY(list). But note that b.x is a new list object, not shared with a.x:

>>> from namedlist import FACTORY
>>> A = namedlist('A', [('x', FACTORY(list))])
>>> a = A()
>>> a.x.append(4)
>>> b = A()
>>> assert b.x == []

Every time a new instance is created, your callable (in this case, list), will be called to produce a new instance for the default value.

Iterating over instances

Because instances are iterable (like lists or tuples), iteration works the same way. Values are returned in definition order:

>>> Point = namedlist('Point', 'x y z t')
>>> p = Point(1.0, 42.0, 3.14, 2.71828)
>>> for value in p:
...    print(value)
1.0
42.0
3.14
2.71828

1.6 2014-12-23 Eric V. Smith

• Add namedlist._update(), similar to dict.update(). Thanks to Arthur Skowronek (issue #23).

• Add namedlist._replace(), similar to namedtuple._replace (issue #24).

1.5 2014-05-20 Eric V. Smith

• Support slices in namedlist.__getattr__ (issue #22).

1.4 2014-03-14 Eric V. Smith

• Add MANIFEST.in to MANIFEST.in, so it will be included in sdists (issue #21).

1.3 2014-03-12 Eric V. Smith

• Support unicode type and field names in Python 2.x (issue #19). The identifiers still must be ASCII only, but you can pass them as unicode. This is useful for code that needs to run under both Python 2 and Python 3.

1.2 2014-02-13 Eric V. Smith

• Produce an RPM named python-namedlist (issue #17).

• Add namedtuple (issue #10). Passes all of the collections.namedtuple tests, except those related to _source. Those tests don’t apply given our different approach to dynamic class creation. All other collections.namedtuple tests have been copied to our test suite.

1.1 2014-02-07 Eric V. Smith

• Added __dict__ so vars() will be supported.

• Fixed pickling from another module (issue #14).

• Moved tests to a separate file (issue #15).

1.0 2014-02-04 Eric V. Smith

• Declare the API stable and release version 1.0.

• Support python 2.6 (issue #8). The doctests don’t pass because OrderedDict isn’t available until 2.7.

0.4 2014-02-04 Eric V. Smith

• Add docstring (issue #7).

• Fixed README.txt typos (thanks pombredanne on bitbucket).

0.3 2014-01-29 Eric V. Smith

• Removed documentation left over from recordtype.

• Make instances unhashable (issue #2).

• For python3, use str.isidentifier (issue #1).

• Reorganize code for name checking. No functional changes.

• Make instances iterable (issue #3).

• Add collections.Sequence ABC (issue #4).

• Have “python setup.py test” also run doctests (issue #5).

0.2 2014-01-28 Eric V. Smith

• Added MANIFEST.in.

• Hopefully fixed a problem with .rst formatting in CHANGES.txt.

0.1 2014-01-28 Eric V. Smith

• Initial release.

• Based off my recordtype project, but uses ast generation instead of building up a string and exec-ing it. This has a number of advantages:

  • Supporting both python2 and python3 is easier. exec has the anti-feature of having different syntax in the two languages.

  • Adding additional features is easier, because I can write in real Python instead of having to write the string version, and deal with all of the escaping and syntax errors.

• Added FACTORY, to allow namedlist to work even with mutable defaults.