summaryrefslogtreecommitdiffstats
path: root/Doc/library/pkgutil.rst
blob: 9971145958c01b1d16bc178a5fd16bce19155202 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
:mod:`pkgutil` --- Package extension utility
============================================

.. module:: pkgutil
   :synopsis: Utilities for the import system.

.. versionadded:: 2.3

**Source code:** :source:`Lib/pkgutil.py`

--------------

This module provides utilities for the import system, in particular package
support.


.. function:: extend_path(path, name)

   Extend the search path for the modules which comprise a package.  Intended
   use is to place the following code in a package's :file:`__init__.py`::

      from pkgutil import extend_path
      __path__ = extend_path(__path__, __name__)

   This will add to the package's ``__path__`` all subdirectories of directories
   on ``sys.path`` named after the package.  This is useful if one wants to
   distribute different parts of a single logical package as multiple
   directories.

   It also looks for :file:`\*.pkg` files beginning where ``*`` matches the
   *name* argument.  This feature is similar to :file:`\*.pth` files (see the
   :mod:`site` module for more information), except that it doesn't special-case
   lines starting with ``import``.  A :file:`\*.pkg` file is trusted at face
   value: apart from checking for duplicates, all entries found in a
   :file:`\*.pkg` file are added to the path, regardless of whether they exist
   on the filesystem.  (This is a feature.)

   If the input path is not a list (as is the case for frozen packages) it is
   returned unchanged.  The input path is not modified; an extended copy is
   returned.  Items are only appended to the copy at the end.

   It is assumed that :data:`sys.path` is a sequence.  Items of :data:`sys.path`
   that are not (Unicode or 8-bit) strings referring to existing directories are
   ignored.  Unicode items on :data:`sys.path` that cause errors when used as
   filenames may cause this function to raise an exception (in line with
   :func:`os.path.isdir` behavior).


.. class:: ImpImporter(dirname=None)

   :pep:`302` Importer that wraps Python's "classic" import algorithm.

   If *dirname* is a string, a :pep:`302` importer is created that searches that
   directory.  If *dirname* is ``None``, a :pep:`302` importer is created that
   searches the current :data:`sys.path`, plus any modules that are frozen or
   built-in.

   Note that :class:`ImpImporter` does not currently support being used by
   placement on :data:`sys.meta_path`.


.. class:: ImpLoader(fullname, file, filename, etc)

   :pep:`302` Loader that wraps Python's "classic" import algorithm.


.. function:: find_loader(fullname)

   Find a :pep:`302` "loader" object for *fullname*.

   If *fullname* contains dots, path must be the containing package's
   ``__path__``.  Returns ``None`` if the module cannot be found or imported.
   This function uses :func:`iter_importers`, and is thus subject to the same
   limitations regarding platform-specific special import locations such as the
   Windows registry.


.. function:: get_importer(path_item)

   Retrieve a :pep:`302` importer for the given *path_item*.

   The returned importer is cached in :data:`sys.path_importer_cache` if it was
   newly created by a path hook.

   If there is no importer, a wrapper around the basic import machinery is
   returned.  This wrapper is never inserted into the importer cache (``None``
   is inserted instead).

   The cache (or part of it) can be cleared manually if a rescan of
   :data:`sys.path_hooks` is necessary.


.. function:: get_loader(module_or_name)

   Get a :pep:`302` "loader" object for *module_or_name*.

   If the module or package is accessible via the normal import mechanism, a
   wrapper around the relevant part of that machinery is returned.  Returns
   ``None`` if the module cannot be found or imported.  If the named module is
   not already imported, its containing package (if any) is imported, in order
   to establish the package ``__path__``.

   This function uses :func:`iter_importers`, and is thus subject to the same
   limitations regarding platform-specific special import locations such as the
   Windows registry.


.. function:: iter_importers(fullname='')

   Yield :pep:`302` importers for the given module name.

   If fullname contains a '.', the importers will be for the package containing
   fullname, otherwise they will be importers for :data:`sys.meta_path`,
   :data:`sys.path`, and Python's "classic" import machinery, in that order.  If
   the named module is in a package, that package is imported as a side effect
   of invoking this function.

   Non-:pep:`302` mechanisms (e.g. the Windows registry) used by the standard
   import machinery to find files in alternative locations are partially
   supported, but are searched *after* :data:`sys.path`.  Normally, these
   locations are searched *before* :data:`sys.path`, preventing :data:`sys.path`
   entries from shadowing them.

   For this to cause a visible difference in behaviour, there must be a module
   or package name that is accessible via both :data:`sys.path` and one of the
   non-:pep:`302` file system mechanisms.  In this case, the emulation will find
   the former version, while the builtin import mechanism will find the latter.

   Items of the following types can be affected by this discrepancy:
   ``imp.C_EXTENSION``, ``imp.PY_SOURCE``, ``imp.PY_COMPILED``,
   ``imp.PKG_DIRECTORY``.


.. function:: iter_modules(path=None, prefix='')

   Yields ``(module_loader, name, ispkg)`` for all submodules on *path*, or, if
   path is ``None``, all top-level modules on ``sys.path``.

   *path* should be either ``None`` or a list of paths to look for modules in.

   *prefix* is a string to output on the front of every module name on output.


.. function:: walk_packages(path=None, prefix='', onerror=None)

   Yields ``(module_loader, name, ispkg)`` for all modules recursively on
   *path*, or, if path is ``None``, all accessible modules.

   *path* should be either ``None`` or a list of paths to look for modules in.

   *prefix* is a string to output on the front of every module name on output.

   Note that this function must import all *packages* (*not* all modules!) on
   the given *path*, in order to access the ``__path__`` attribute to find
   submodules.

   *onerror* is a function which gets called with one argument (the name of the
   package which was being imported) if any exception occurs while trying to
   import a package.  If no *onerror* function is supplied, :exc:`ImportError`\s
   are caught and ignored, while all other exceptions are propagated,
   terminating the search.

   Examples::

      # list all modules python can access
      walk_packages()

      # list all submodules of ctypes
      walk_packages(ctypes.__path__, ctypes.__name__ + '.')


.. function:: get_data(package, resource)

   Get a resource from a package.

   This is a wrapper for the :pep:`302` loader :func:`get_data` API.  The
   *package* argument should be the name of a package, in standard module format
   (``foo.bar``).  The *resource* argument should be in the form of a relative
   filename, using ``/`` as the path separator.  The parent directory name
   ``..`` is not allowed, and nor is a rooted name (starting with a ``/``).

   The function returns a binary string that is the contents of the specified
   resource.

   For packages located in the filesystem, which have already been imported,
   this is the rough equivalent of::

      d = os.path.dirname(sys.modules[package].__file__)
      data = open(os.path.join(d, resource), 'rb').read()

   If the package cannot be located or loaded, or it uses a :pep:`302` loader
   which does not support :func:`get_data`, then ``None`` is returned.

   .. versionadded:: 2.6