freeipa/ipalib/config.py
John Dennis 94d457e83c Use DN objects instead of strings
* Convert every string specifying a DN into a DN object

* Every place a dn was manipulated in some fashion it was replaced by
  the use of DN operators

* Add new DNParam parameter type for parameters which are DN's

* DN objects are used 100% of the time throughout the entire data
  pipeline whenever something is logically a dn.

* Many classes now enforce DN usage for their attributes which are
  dn's. This is implmented via ipautil.dn_attribute_property(). The
  only permitted types for a class attribute specified to be a DN are
  either None or a DN object.

* Require that every place a dn is used it must be a DN object.
  This translates into lot of::

    assert isinstance(dn, DN)

  sprinkled through out the code. Maintaining these asserts is
  valuable to preserve DN type enforcement. The asserts can be
  disabled in production.

  The goal of 100% DN usage 100% of the time has been realized, these
  asserts are meant to preserve that.

  The asserts also proved valuable in detecting functions which did
  not obey their function signatures, such as the baseldap pre and
  post callbacks.

* Moved ipalib.dn to ipapython.dn because DN class is shared with all
  components, not just the server which uses ipalib.

* All API's now accept DN's natively, no need to convert to str (or
  unicode).

* Removed ipalib.encoder and encode/decode decorators. Type conversion
  is now explicitly performed in each IPASimpleLDAPObject method which
  emulates a ldap.SimpleLDAPObject method.

* Entity & Entry classes now utilize DN's

* Removed __getattr__ in Entity & Entity clases. There were two
  problems with it. It presented synthetic Python object attributes
  based on the current LDAP data it contained. There is no way to
  validate synthetic attributes using code checkers, you can't search
  the code to find LDAP attribute accesses (because synthetic
  attriutes look like Python attributes instead of LDAP data) and
  error handling is circumscribed. Secondly __getattr__ was hiding
  Python internal methods which broke class semantics.

* Replace use of methods inherited from ldap.SimpleLDAPObject via
  IPAdmin class with IPAdmin methods. Directly using inherited methods
  was causing us to bypass IPA logic. Mostly this meant replacing the
  use of search_s() with getEntry() or getList(). Similarly direct
  access of the LDAP data in classes using IPAdmin were replaced with
  calls to getValue() or getValues().

* Objects returned by ldap2.find_entries() are now compatible with
  either the python-ldap access methodology or the Entity/Entry access
  methodology.

* All ldap operations now funnel through the common
  IPASimpleLDAPObject giving us a single location where we interface
  to python-ldap and perform conversions.

* The above 4 modifications means we've greatly reduced the
  proliferation of multiple inconsistent ways to perform LDAP
  operations. We are well on the way to having a single API in IPA for
  doing LDAP (a long range goal).

* All certificate subject bases are now DN's

* DN objects were enhanced thusly:
  - find, rfind, index, rindex, replace and insert methods were added
  - AVA, RDN and DN classes were refactored in immutable and mutable
    variants, the mutable variants are EditableAVA, EditableRDN and
    EditableDN. By default we use the immutable variants preserving
    important semantics. To edit a DN cast it to an EditableDN and
    cast it back to DN when done editing. These issues are fully
    described in other documentation.
  - first_key_match was removed
  - DN equalty comparison permits comparison to a basestring

* Fixed ldapupdate to work with DN's. This work included:
  - Enhance test_updates.py to do more checking after applying
    update. Add test for update_from_dict(). Convert code to use
    unittest classes.
  - Consolidated duplicate code.
  - Moved code which should have been in the class into the class.
  - Fix the handling of the 'deleteentry' update action. It's no longer
    necessary to supply fake attributes to make it work. Detect case
    where subsequent update applies a change to entry previously marked
    for deletetion. General clean-up and simplification of the
    'deleteentry' logic.
  - Rewrote a couple of functions to be clearer and more Pythonic.
  - Added documentation on the data structure being used.
  - Simplfy the use of update_from_dict()

* Removed all usage of get_schema() which was being called prior to
  accessing the .schema attribute of an object. If a class is using
  internal lazy loading as an optimization it's not right to require
  users of the interface to be aware of internal
  optimization's. schema is now a property and when the schema
  property is accessed it calls a private internal method to perform
  the lazy loading.

* Added SchemaCache class to cache the schema's from individual
  servers. This was done because of the observation we talk to
  different LDAP servers, each of which may have it's own
  schema. Previously we globally cached the schema from the first
  server we connected to and returned that schema in all contexts. The
  cache includes controls to invalidate it thus forcing a schema
  refresh.

* Schema caching is now senstive to the run time context. During
  install and upgrade the schema can change leading to errors due to
  out-of-date cached schema. The schema cache is refreshed in these
  contexts.

* We are aware of the LDAP syntax of all LDAP attributes. Every
  attribute returned from an LDAP operation is passed through a
  central table look-up based on it's LDAP syntax. The table key is
  the LDAP syntax it's value is a Python callable that returns a
  Python object matching the LDAP syntax. There are a handful of LDAP
  attributes whose syntax is historically incorrect
  (e.g. DistguishedNames that are defined as DirectoryStrings). The
  table driven conversion mechanism is augmented with a table of
  hard coded exceptions.

  Currently only the following conversions occur via the table:

  - dn's are converted to DN objects

  - binary objects are converted to Python str objects (IPA
    convention).

  - everything else is converted to unicode using UTF-8 decoding (IPA
    convention).

  However, now that the table driven conversion mechanism is in place
  it would be trivial to do things such as converting attributes
  which have LDAP integer syntax into a Python integer, etc.

* Expected values in the unit tests which are a DN no longer need to
  use lambda expressions to promote the returned value to a DN for
  equality comparison. The return value is automatically promoted to
  a DN. The lambda expressions have been removed making the code much
  simpler and easier to read.

* Add class level logging to a number of classes which did not support
  logging, less need for use of root_logger.

* Remove ipaserver/conn.py, it was unused.

* Consolidated duplicate code wherever it was found.

* Fixed many places that used string concatenation to form a new
  string rather than string formatting operators. This is necessary
  because string formatting converts it's arguments to a string prior
  to building the result string. You can't concatenate a string and a
  non-string.

* Simplify logic in rename_managed plugin. Use DN operators to edit
  dn's.

* The live version of ipa-ldap-updater did not generate a log file.
  The offline version did, now both do.

https://fedorahosted.org/freeipa/ticket/1670
https://fedorahosted.org/freeipa/ticket/1671
https://fedorahosted.org/freeipa/ticket/1672
https://fedorahosted.org/freeipa/ticket/1673
https://fedorahosted.org/freeipa/ticket/1674
https://fedorahosted.org/freeipa/ticket/1392
https://fedorahosted.org/freeipa/ticket/2872
2012-08-12 16:23:24 -04:00

587 lines
20 KiB
Python

# Authors:
# Martin Nagy <mnagy@redhat.com>
# Jason Gerard DeRose <jderose@redhat.com>
#
# Copyright (C) 2008 Red Hat
# see file 'COPYING' for use and warranty information
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
"""
Process-wide static configuration and environment.
The standard run-time instance of the `Env` class is initialized early in the
`ipalib` process and is then locked into a read-only state, after which no
further changes can be made to the environment throughout the remaining life
of the process.
For the per-request thread-local information, see `ipalib.request`.
"""
from ConfigParser import RawConfigParser, ParsingError
from types import NoneType
import os
from os import path
import sys
from socket import getfqdn
from ipapython.dn import DN
from base import check_name
from constants import CONFIG_SECTION
from constants import TYPE_ERROR, OVERRIDE_ERROR, SET_ERROR, DEL_ERROR
class Env(object):
"""
Store and retrieve environment variables.
First an foremost, the `Env` class provides a handy container for
environment variables. These variables can be both set *and* retrieved
either as attributes *or* as dictionary items.
For example, you can set a variable as an attribute:
>>> env = Env()
>>> env.attr = 'I was set as an attribute.'
>>> env.attr
u'I was set as an attribute.'
>>> env['attr'] # Also retrieve as a dictionary item
u'I was set as an attribute.'
Or you can set a variable as a dictionary item:
>>> env['item'] = 'I was set as a dictionary item.'
>>> env['item']
u'I was set as a dictionary item.'
>>> env.item # Also retrieve as an attribute
u'I was set as a dictionary item.'
The variable names must be valid lower-case Python identifiers that neither
start nor end with an underscore. If your variable name doesn't meet these
criteria, a ``ValueError`` will be raised when you try to set the variable
(compliments of the `base.check_name()` function). For example:
>>> env.BadName = 'Wont work as an attribute'
Traceback (most recent call last):
...
ValueError: name must match '^[a-z][_a-z0-9]*[a-z0-9]$|^[a-z]$'; got 'BadName'
>>> env['BadName'] = 'Also wont work as a dictionary item'
Traceback (most recent call last):
...
ValueError: name must match '^[a-z][_a-z0-9]*[a-z0-9]$|^[a-z]$'; got 'BadName'
The variable values can be ``str``, ``int``, or ``float`` instances, or the
``True``, ``False``, or ``None`` constants. When the value provided is an
``str`` instance, some limited automatic type conversion is performed, which
allows values of specific types to be set easily from configuration files or
command-line options.
So in addition to their actual values, the ``True``, ``False``, and ``None``
constants can be specified with an ``str`` equal to what ``repr()`` would
return. For example:
>>> env.true = True
>>> env.also_true = 'True' # Equal to repr(True)
>>> env.true
True
>>> env.also_true
True
Note that the automatic type conversion is case sensitive. For example:
>>> env.not_false = 'false' # Not equal to repr(False)!
>>> env.not_false
u'false'
If an ``str`` value looks like an integer, it's automatically converted to
the ``int`` type. Likewise, if an ``str`` value looks like a floating-point
number, it's automatically converted to the ``float`` type. For example:
>>> env.lucky = '7'
>>> env.lucky
7
>>> env.three_halves = '1.5'
>>> env.three_halves
1.5
Leading and trailing white-space is automatically stripped from ``str``
values. For example:
>>> env.message = ' Hello! ' # Surrounded by double spaces
>>> env.message
u'Hello!'
>>> env.number = ' 42 ' # Still converted to an int
>>> env.number
42
>>> env.false = ' False ' # Still equal to repr(False)
>>> env.false
False
Also, empty ``str`` instances are converted to ``None``. For example:
>>> env.empty = ''
>>> env.empty is None
True
`Env` variables are all set-once (first-one-wins). Once a variable has been
set, trying to override it will raise an ``AttributeError``. For example:
>>> env.date = 'First'
>>> env.date = 'Second'
Traceback (most recent call last):
...
AttributeError: cannot override Env.date value u'First' with 'Second'
An `Env` instance can be *locked*, after which no further variables can be
set. Trying to set variables on a locked `Env` instance will also raise
an ``AttributeError``. For example:
>>> env = Env()
>>> env.okay = 'This will work.'
>>> env.__lock__()
>>> env.nope = 'This wont work!'
Traceback (most recent call last):
...
AttributeError: locked: cannot set Env.nope to 'This wont work!'
`Env` instances also provide standard container emulation for membership
testing, counting, and iteration. For example:
>>> env = Env()
>>> 'key1' in env # Has key1 been set?
False
>>> env.key1 = 'value 1'
>>> 'key1' in env
True
>>> env.key2 = 'value 2'
>>> len(env) # How many variables have been set?
2
>>> list(env) # What variables have been set?
['key1', 'key2']
Lastly, in addition to all the handy container functionality, the `Env`
class provides high-level methods for bootstraping a fresh `Env` instance
into one containing all the run-time and configuration information needed
by the built-in freeIPA plugins.
These are the `Env` bootstraping methods, in the order they must be called:
1. `Env._bootstrap()` - initialize the run-time variables and then
merge-in variables specified on the command-line.
2. `Env._finalize_core()` - merge-in variables from the configuration
files and then merge-in variables from the internal defaults, after
which at least all the standard variables will be set. After this
method is called, the plugins will be loaded, during which
third-party plugins can merge-in defaults for additional variables
they use (likely using the `Env._merge()` method).
3. `Env._finalize()` - one last chance to merge-in variables and then
the instance is locked. After this method is called, no more
environment variables can be set during the remaining life of the
process.
However, normally none of these three bootstraping methods are called
directly and instead only `plugable.API.bootstrap()` is called, which itself
takes care of correctly calling the `Env` bootstrapping methods.
"""
__locked = False
def __init__(self, **initialize):
object.__setattr__(self, '_Env__d', {})
object.__setattr__(self, '_Env__done', set())
if initialize:
self._merge(**initialize)
def __lock__(self):
"""
Prevent further changes to environment.
"""
if self.__locked is True:
raise StandardError(
'%s.__lock__() already called' % self.__class__.__name__
)
object.__setattr__(self, '_Env__locked', True)
def __islocked__(self):
"""
Return ``True`` if locked.
"""
return self.__locked
def __setattr__(self, name, value):
"""
Set the attribute named ``name`` to ``value``.
This just calls `Env.__setitem__()`.
"""
self[name] = value
def __setitem__(self, key, value):
"""
Set ``key`` to ``value``.
"""
if self.__locked:
raise AttributeError(
SET_ERROR % (self.__class__.__name__, key, value)
)
check_name(key)
if key in self.__d:
raise AttributeError(OVERRIDE_ERROR %
(self.__class__.__name__, key, self.__d[key], value)
)
assert not hasattr(self, key)
if isinstance(value, basestring):
value = value.strip()
if isinstance(value, str):
value = value.decode('utf-8')
m = {
'True': True,
'False': False,
'None': None,
'': None,
}
if value in m:
value = m[value]
elif value.isdigit():
value = int(value)
elif key in ('basedn'):
value = DN(value)
else:
try:
value = float(value)
except (TypeError, ValueError):
pass
assert type(value) in (unicode, int, float, bool, NoneType, DN)
object.__setattr__(self, key, value)
self.__d[key] = value
def __getitem__(self, key):
"""
Return the value corresponding to ``key``.
"""
return self.__d[key]
def __delattr__(self, name):
"""
Raise an ``AttributeError`` (deletion is never allowed).
For example:
>>> env = Env()
>>> env.name = 'A value'
>>> del env.name
Traceback (most recent call last):
...
AttributeError: locked: cannot delete Env.name
"""
raise AttributeError(
DEL_ERROR % (self.__class__.__name__, name)
)
def __contains__(self, key):
"""
Return True if instance contains ``key``; otherwise return False.
"""
return key in self.__d
def __len__(self):
"""
Return number of variables currently set.
"""
return len(self.__d)
def __iter__(self):
"""
Iterate through keys in ascending order.
"""
for key in sorted(self.__d):
yield key
def _merge(self, **kw):
"""
Merge variables from ``kw`` into the environment.
Any variables in ``kw`` that have already been set will be ignored
(meaning this method will *not* try to override them, which would raise
an exception).
This method returns a ``(num_set, num_total)`` tuple containing first
the number of variables that were actually set, and second the total
number of variables that were provided.
For example:
>>> env = Env()
>>> env._merge(one=1, two=2)
(2, 2)
>>> env._merge(one=1, three=3)
(1, 2)
>>> env._merge(one=1, two=2, three=3)
(0, 3)
Also see `Env._merge_from_file()`.
:param kw: Variables provides as keyword arguments.
"""
i = 0
for (key, value) in kw.iteritems():
if key not in self:
self[key] = value
i += 1
return (i, len(kw))
def _merge_from_file(self, config_file):
"""
Merge variables from ``config_file`` into the environment.
Any variables in ``config_file`` that have already been set will be
ignored (meaning this method will *not* try to override them, which
would raise an exception).
If ``config_file`` does not exist or is not a regular file, or if there
is an error parsing ``config_file``, ``None`` is returned.
Otherwise this method returns a ``(num_set, num_total)`` tuple
containing first the number of variables that were actually set, and
second the total number of variables found in ``config_file``.
This method will raise a ``ValueError`` if ``config_file`` is not an
absolute path. For example:
>>> env = Env()
>>> env._merge_from_file('my/config.conf')
Traceback (most recent call last):
...
ValueError: config_file must be an absolute path; got 'my/config.conf'
Also see `Env._merge()`.
:param config_file: Absolute path of the configuration file to load.
"""
if path.abspath(config_file) != config_file:
raise ValueError(
'config_file must be an absolute path; got %r' % config_file
)
if not path.isfile(config_file):
return
parser = RawConfigParser()
try:
parser.read(config_file)
except ParsingError:
return
if not parser.has_section(CONFIG_SECTION):
parser.add_section(CONFIG_SECTION)
items = parser.items(CONFIG_SECTION)
if len(items) == 0:
return (0, 0)
i = 0
for (key, value) in items:
if key not in self:
self[key] = value
i += 1
if 'config_loaded' not in self: # we loaded at least 1 file
self['config_loaded'] = True
return (i, len(items))
def _join(self, key, *parts):
"""
Append path components in ``parts`` to base path ``self[key]``.
For example:
>>> env = Env()
>>> env.home = '/people/joe'
>>> env._join('home', 'Music', 'favourites')
u'/people/joe/Music/favourites'
"""
if key in self and self[key] is not None:
return path.join(self[key], *parts)
def __doing(self, name):
if name in self.__done:
raise StandardError(
'%s.%s() already called' % (self.__class__.__name__, name)
)
self.__done.add(name)
def __do_if_not_done(self, name):
if name not in self.__done:
getattr(self, name)()
def _isdone(self, name):
return name in self.__done
def _bootstrap(self, **overrides):
"""
Initialize basic environment.
This method will perform the following steps:
1. Initialize certain run-time variables. These run-time variables
are strictly determined by the external environment the process
is running in; they cannot be specified on the command-line nor
in the configuration files.
2. Merge-in the variables in ``overrides`` by calling
`Env._merge()`. The intended use of ``overrides`` is to merge-in
variables specified on the command-line.
3. Intelligently fill-in the *in_tree*, *context*, *conf*, and
*conf_default* variables if they haven't been set already.
Also see `Env._finalize_core()`, the next method in the bootstrap
sequence.
:param overrides: Variables specified via command-line options.
"""
self.__doing('_bootstrap')
# Set run-time variables (cannot be overridden):
self.ipalib = path.dirname(path.abspath(__file__))
self.site_packages = path.dirname(self.ipalib)
self.script = path.abspath(sys.argv[0])
self.bin = path.dirname(self.script)
self.home = os.environ.get('HOME', None)
# Merge in overrides:
self._merge(**overrides)
# Determine if running in source tree:
if 'in_tree' not in self:
if (
self.bin == self.site_packages
and path.isfile(path.join(self.bin, 'setup.py'))
):
self.in_tree = True
else:
self.in_tree = False
if self.in_tree and 'mode' not in self:
self.mode = 'developer'
# Set dot_ipa:
if 'dot_ipa' not in self:
self.dot_ipa = self._join('home', '.ipa')
# Set context
if 'context' not in self:
self.context = 'default'
# Set confdir:
if 'confdir' not in self:
if self.in_tree:
self.confdir = self.dot_ipa
else:
self.confdir = path.join('/', 'etc', 'ipa')
# Set conf (config file for this context):
if 'conf' not in self:
self.conf = self._join('confdir', '%s.conf' % self.context)
# Set conf_default (default base config used in all contexts):
if 'conf_default' not in self:
self.conf_default = self._join('confdir', 'default.conf')
# Set plugins_on_demand:
if 'plugins_on_demand' not in self:
self.plugins_on_demand = (self.context == 'cli')
def _finalize_core(self, **defaults):
"""
Complete initialization of standard IPA environment.
This method will perform the following steps:
1. Call `Env._bootstrap()` if it hasn't already been called.
2. Merge-in variables from the configuration file ``self.conf``
(if it exists) by calling `Env._merge_from_file()`.
3. Merge-in variables from the defaults configuration file
``self.conf_default`` (if it exists) by calling
`Env._merge_from_file()`.
4. Intelligently fill-in the *in_server* , *logdir*, and *log*
variables if they haven't already been set.
5. Merge-in the variables in ``defaults`` by calling `Env._merge()`.
In normal circumstances ``defaults`` will simply be those
specified in `constants.DEFAULT_CONFIG`.
After this method is called, all the environment variables used by all
the built-in plugins will be available. As such, this method should be
called *before* any plugins are loaded.
After this method has finished, the `Env` instance is still writable
so that 3rd-party plugins can set variables they may require as the
plugins are registered.
Also see `Env._finalize()`, the final method in the bootstrap sequence.
:param defaults: Internal defaults for all built-in variables.
"""
self.__doing('_finalize_core')
self.__do_if_not_done('_bootstrap')
# Merge in context config file and then default config file:
if self.__d.get('mode', None) != 'dummy':
self._merge_from_file(self.conf)
self._merge_from_file(self.conf_default)
# Determine if in_server:
if 'in_server' not in self:
self.in_server = (self.context == 'server')
# Set logdir:
if 'logdir' not in self:
if self.in_tree or not self.in_server:
self.logdir = self._join('dot_ipa', 'log')
else:
self.logdir = path.join('/', 'var', 'log', 'ipa')
# Set log file:
if 'log' not in self:
self.log = self._join('logdir', '%s.log' % self.context)
self._merge(**defaults)
def _finalize(self, **lastchance):
"""
Finalize and lock environment.
This method will perform the following steps:
1. Call `Env._finalize_core()` if it hasn't already been called.
2. Merge-in the variables in ``lastchance`` by calling
`Env._merge()`.
3. Lock this `Env` instance, after which no more environment
variables can be set on this instance. Aside from unit-tests
and example code, normally only one `Env` instance is created,
which means that after this step, no more variables can be set
during the remaining life of the process.
This method should be called after all plugins have been loaded and
after `plugable.API.finalize()` has been called.
:param lastchance: Any final variables to merge-in before locking.
"""
self.__doing('_finalize')
self.__do_if_not_done('_finalize_core')
self._merge(**lastchance)
self.__lock__()