Coverage for ipapython/dn : 87%

# Authors: 

#   John Dennis <jdennis@redhat.com> 

# 

# Copyright (C) 2011  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/>. 

''' 

Goal 

---- 

To allow a Python programmer the ability to operate on DN's 

(Distinguished Names) in a simple intuitive manner supporting all the 

Pythonic mechanisms for manipulating objects such that the simple 

majority case remains simple with simple code, yet the corner cases 

are fully supported. With the result both simple and complex cases are 

100% correct. 

This is achieved with a fair of amount of syntax sugar which is best 

described as "Do What I Mean" (i.e. DWIM). The class implementations 

take simple expressions and internally convert them to their more 

complex full definitions hiding much of the complexity from the 

programmer. 

Anatomy of a DN 

--------------- 

Some definitions: 

AVA 

    An AVA is an Attribute Value Assertion. In more simple terms it's 

    an attribute value pair typically expressed as attr=value 

    (e.g. cn=Bob). Both the attr and value in an AVA when expressed in 

    a string representation are subject to encoding rules. 

RDN 

    A RDN is a Relative Distinguished Name. A RDN is a non-empty set of 

    AVA's. In the common case a RDN is single valued consisting of 1 

    AVA (e.g. cn=Bob). But a RDN may be multi-valued consisting of 

    more than one AVA. Because the RDN is a set of AVA's the AVA's are 

    unordered when they appear in a multi-valued RDN. In the string 

    representation of a RDN AVA's are separated by the plus sign (+). 

DN 

    A DN is a ordered sequence of 1 or more RDN's. In the string 

    representation of a DN each RDN is separated by a comma (,) 

Thus a DN is: 

Sequence of set of <encoded attr, encoded value> pairs 

The following are valid DN's 

# 1 RDN with 1 AVA (e.g. cn=Bob) 

RDN(AVA) 

# 2 RDN's each with 1 AVA (e.g. cn=Bob,dc=redhat.com) 

RDN(AVA),RDN(AVA) 

# 2 RDN's the first RDN is multi-valued with 2 AVA's 

# the second RDN is singled valued with 1 AVA 

# (e.g. cn=Bob+ou=people,dc=redhat.com 

RDN({AVA,AVA}),RDN(AVA) 

Common programming mistakes 

--------------------------- 

DN's present a pernicious problem for programmers. They appear to have 

a very simple string format in the majority case, a sequence of 

attr=value pairs separated by commas. For example: 

dn='cn=Bob,ou=people,dc=redhat,dc=com' 

As such there is a tendency to believe you can form DN's by simple 

string manipulations such as: 

dn='%s=%s' % ('cn','Bob') + ',ou=people,dc=redhat,dc=com' 

Or to extract a attr & value by searching the string, for example: 

attr=dn[0 : dn.find('=')] 

value=dn[dn.find('=')+1 : dn.find(',')] 

Or compare a value returned by an LDAP query to a known value: 

if value == 'Bob' 

All of these simple coding assumptions are WRONG and will FAIL when a 

DN is not one of the simple DN's (simple DN's are probably the 95% of 

all DN's). This is what makes DN handling pernicious. What works in 

95% of the cases and is simple, fails for the 5% of DN's which are not 

simple. 

Examples of where the simple assumptions fail are: 

* A RDN may be multi-valued 

* A multi-valued RDN has no ordering on it's components 

* Attr's and values must be UTF-8 encoded 

* String representations of AVA's, RDN's and DN's must be completely UTF-8 

* An attr or value may have reserved characters which must be escaped. 

* Whitespace needs special handling 

To complicate matters a bit more the RFC for the string representation 

of DN's (RFC 4514) permits a variety of different syntax's each of 

which can evaluate to exactly the same DN but have different string 

representations. For example, the attr "r,w" which contains a reserved 

character (the comma) can be encoded as a string in these different 

ways: 

'r\,w'          # backslash escape 

'r\2cw'         # hexadecimal ascii escape 

'#722C77'       # binary encoded 

It should be clear a DN string may NOT be a simple string, rather a DN 

string is ENCODED. For simple strings the encoding of the DN is 

identical to the simple string value (this common case leads to 

erroneous assumptions and bugs because it does not account for 

encodings). 

The openldap library we use at the client level uses the backslash 

escape form. The LDAP server we use uses the hexadecimal ascii escape 

form. Thus 'r,w' appears as 'r\,w' when sent from the client to the 

LDAP server as part of a DN. But when it's returned as a DN from the 

server in an LDAP search it's returned as 'r\2cw'. Any attempt to 

compare 'r\,w' to 'r\2cw' for equality will fail despite the fact they 

are indeed equal once decoded. Such a test fails because you're 

comparing two different encodings of the same value. In MIME you 

wouldn't expect the base64 encoding of a string to be equal to the 

same string encoded as quoted-printable would you? 

When you are comparing attrs or values which are part of a DN and 

other string you MUST: 

* Know if either of the strings have been encoded and make sure you're 

  comparing only decoded components component-wise. 

* Extract the component from the DN and decode it. You CANNOT decode 

  the entire DN as a string and operate on it. Why? Consider a value 

  with a comma embedded in it. For example: 

  cn=r\2cw,cn=privilege 

  Is a DN with 2 RDN components: cn=r,w followed by "cn=privilege" 

  But if you decode the entire DN string as a whole you would get: 

  cn=r,w,cn=privilege 

  Which is a malformed DN with 3 RDN's, the 2nd RDN is invalid. 

* Determine if a RDN is multi-valued, if so you must account 

  for the fact each AVA component in the multi-valued RDN can appear 

  in any order and still be equivalent. For example the following two 

  RDN's are equal: 

  cn=Bob+ou=people 

  ou=people+cn=Bob 

  In addition each AVA (cn=Bob & ou=people) needs to be 

  INDEPENDENTLY decoded prior to comparing the unordered set of AVA's 

  in the multi-valued RDN. 

If you are trying to form a new DN or RDN from a raw string you cannot 

simply do string concatenation or string formatting unless you ESCAPE 

the components independently prior to concatenation, for example: 

  base = 'dc=redhat,dc=com' 

  value = 'r,w' 

  dn = 'cn=%s,%s' % (value, base) 

Will result in the malformed DN 'cn=r,w,dc=redhat,dc=com' 

Syntax Sugar 

------------ 

The majority of DN's have a simple string form: 

attr=value,attr=value 

We want the programmer to be able to create DN's, compare them, and 

operate on their components as simply and concisely as possible so 

the classes are implemented to provide a lot of syntax sugar. 

The classes automatically handle UTF-8 <-> Unicode conversions. Every 

attr and value which is returned from a class will be Unicode. Every 

attr and value assigned into an object will be promoted to 

Unicode. All string representations in RFC 4514 format will be UTF-8 

and properly escaped. Thus at the "user" or "API" level every string 

is Unicode with the single exception that the str() method returns RFC 

compliant escaped UTF-8. 

RDN's are assumed to be single-valued. If you need a multi-valued RDN 

(an exception) you must explicitly create a multi-valued RDN. 

Thus DN's are assumed to be a sequence of attr, value pairs, which is 

equivalent to a sequence of RDN's. The attr and value in the pair MUST 

be strings. 

The DN and RDN constructors take a sequence, the constructor parses 

the sequence to find items it knows about. 

The DN constructor will accept in it's sequence: 

  * tuple of 2 strings, converting it to an RDN 

  * list of 2 strings, converting it to an RDN 

  * a RDN object 

  * a DN syntax string (e.g. 'cn=Bob,dc=redhat.com') 

Note DN syntax strings should be avoided if possible when passing to a 

constructor because they run afoul of the problems outlined above 

which the DN, RDN & AVA classes are meant to overcome. But sometimes a 

DN syntax string is all you have to work with. DN strings which come 

from a LDAP library or server will be properly formed and it's safe to 

use those. However DN strings provided via user input should be 

treated suspiciously as they may be improperly formed. You can test 

for this by passing the string to the DN constructor and see if it 

throws an exception. 

The sequence passed to the DN constructor takes each item in order, 

produces one or more RDN's from it and appends those RDN in order to 

its internal RDN sequence. 

For example: 

   DN(('cn', 'Bob'), ('dc', 'redhat.com')) 

This is equivalent to the DN string: 

    cn=Bob,dc=redhat.com 

And is exactly equal to: 

    DN(RDN(AVA('cn','Bob')),RDN(AVA('dc','redhat.com'))) 

The following are alternative syntax's which are all exactly 

equivalent to the above example. 

   DN(['cn', 'Bob'], ['dc', 'redhat.com']) 

   DN(RDN('cn', 'Bob'), RDN('dc', 'redhat.com')) 

You can provide a properly escaped string representation. 

   DN('cn=Bob,dc=redhat.com') 

You can mix and match any of the forms in the constructor parameter 

list. 

   DN(('cn', 'Bob'), 'dc=redhat.com') 

   DN(('cn', 'Bob'), RDN('dc', 'redhat.com')) 

AVA's have an attr and value property, thus if you have an AVA 

# Get the attr and value 

ava.attr  -> u'cn' 

ava.value -> u'Bob' 

# Set the attr and value 

ava.attr  = 'cn' 

ava.value = 'Bob' 

Since RDN's are assumed to be single valued, exactly the same 

behavior applies to an RDN. If the RDN is multi-valued then the attr 

property returns the attr of the first AVA, likewise for the value. 

# Get the attr and value 

rdn.attr  -> u'cn' 

rdn.value -> u'Bob' 

# Set the attr and value 

rdn.attr  = 'cn' 

rdn.value = 'Bob' 

Also RDN's can be indexed by name or position (see the RDN class doc 

for details). 

rdn['cn'] -> u'Bob' 

rdn[0] -> AVA('cn', 'Bob') 

A DN is a sequence of RDN's, as such any of Python's container 

operators can be applied to a DN in a intuitive way. 

# How many RDN's in a DN? 

len(dn) 

# WARNING, this a count of RDN's not how characters there are in the 

# string representation the dn, instead that would be: 

len(str(dn)) 

# Iterate over each RDN in a DN 

for rdn in dn: 

# Get the first RDN in a DN 

dn[0] -> RDN('cn', 'Bob') 

# Get the value of the first RDN in a DN 

dn[0].value -> u'Bob' 

# Get the value of the first RDN by indexing by attr name 

dn['cn'] -> u'Bob' 

# WARNING, when a string is used as an index key the FIRST RDN's value 

# in the sequence whose attr matches the key is returned. Thus if you 

# have a DN like this "cn=foo,cn=bar" then dn['cn'] will always return 

# 'foo' even though there is another attr with the name 'cn'. This is 

# almost always what the programmer wants. See the class doc for how 

# you can override this default behavior and get a list of every value 

# whose attr matches the key. 

# Set the first RDN in the DN (all are equivalent) 

dn[0] = ('cn', 'Bob') 

dn[0] = ['cn', 'Bob'] 

dn[0] = RDN('cn', 'Bob') 

dn[0].attr = 'cn' 

dn[0].value = 'Bob' 

# Get the first two RDN's using slices 

dn[0:2] 

# Get the last two RDN's using slices 

dn[-2:] 

# Get a list of all RDN's using slices 

dn[:] 

# Set the 2nd and 3rd RDN using slices (all are equivalent) 

dn[1:3] = ('cn', 'Bob), ('dc', 'redhat.com') 

dn[1:3] = RDN('cn', 'Bob), RDN('dc', 'redhat.com') 

String representations and escapes: 

# To get an RFC compliant string representation of a DN, RDN or AVA 

# simply call str() on it or evaluate it in a string context. 

str(dn) -> 'cn=Bob,dc=redhat.com' 

# When working with attr's and values you do not have to worry about 

# escapes, simply use the raw unescaped string in a natural fashion. 

rdn = RDN('cn', 'r,w') 

# Thus: 

rdn.value == 'r,w' -> True 

# But: 

str(rdn) == 'cn=r,w' -> False 

# Because: 

str(rdn) -> 'cn=r\2cw' or 'cn='r\,w' # depending on the underlying LDAP library 

Equality and Comparing: 

# All DN's, RDN's and AVA's support equality testing in an intuitive 

# manner. 

dn1 = DN(('cn', 'Bob')) 

dn2 = DN(RDN('cn', 'Bob')) 

dn1 == dn2 -> True 

dn1[0] == dn2[0] -> True 

dn1[0].value = 'Bobby' 

dn1 == dn2 -> False 

DN objects implement startswith(), endswith() and the "in" membership 

operator. You may pass a DN or RDN object to these. Examples: 

if dn.endswith(base_dn): 

if dn.startswith(rdn1): 

if container_dn in dn: 

# See the class doc for how DN's, RDN's and AVA's compare 

# (e.g. cmp()). The general rule is for objects supporting multiple 

# values first their lengths are compared, then if the lengths match 

# the respective components of each are pair-wise compared until one 

# is discovered to be  non-equal. The comparison is case insensitive. 

Cloning (Object Copy): 

All the class types are capable of cloning by passing an object of the 

same type (or subclass) to the constructor. The new object is a copy 

of the object passed as input to the constructor. One place this is 

useful is when you want to coerce between immutable and mutable 

versions in order to modify an object. 

Concatenation, In-Place Addition, Insertion: 

# DN's and RDN's can be concatenated. 

# Return a new DN by appending the RDN's of dn2 to dn1 

dn3 = dn1 + dn2 

# Append a RDN to DN's RDN sequence (all are equivalent) 

dn += ('cn', 'Bob') 

dn += RDN('cn', 'Bob') 

# Append a DN to an existing DN 

dn1 += dn2 

# Prepend a RDN to an existing DN 

dn1.insert(0, RDN('cn', 'Bob')) 

Finally see the unittest for a more complete set of ways you can 

manipulate these objects. 

Mutability 

---------- 

Python makes a clear distinction between mutable and immutable 

objects. Examples of immutable Python objects are strings, integers 

and floats. Examples of mutable Python objects are lists, dicts, and 

sets. Immutable objects cannot be modified, mutable objects can be 

modified. An object's mutability affects how the object behaves when 

passed to a function or method, this is because it's the object's 

reference which is always passed, thus immutable objects behave as if 

it were "call by value" and mutable objects behave as if it were "call 

by reference" (mutable objects can be modifed inside the 

function/method and that modification will be visible to the 

caller. On object's mutability also affects how an object will behave 

when used as a key in a dict or as a member of a set. 

The following discussion applies equally to AVA, RDN and DN object 

class variants. 

The AVA, RDN and DN classes have both immutable and mutable 

variants. The base classes (AVA, RDN, DN) are immutable. Each of the 

immutable base classes have a mutable subclass whose name begins with 

'Editable'. Thus the DN class is immutable, instances of that class 

cannot be modified, there is a mutable class EditableDN derived from 

DN whose instances can be modified. The primary difference between the 

immutable and mutable variants is: 

* Immutable variants are preferred. 

* Mutable variants are exactly identical in behavior to their 

  immutable parent class (except for supporting assignment, etc.) 

* Immutable objects that test as equal will be the same as dict keys 

  and set members even if they are different objects. Mutable variants 

  are not hashable and thus cannot be used as a dict key nor inserted 

  into a set. 

* Only mutable variants support modification via assignment, insert or 

  in-place addition (e.g. +=). 

* In-place addtion (e.g. +=) works for both immutable and mutable 

  variants. The distinction is for immutable objects the lhs is 

  replaced with a new immutable result while a mutable object will be 

  modfied in place and lhs object remains the same object. 

It is trival to coerce between an mutable and immutable AVA, RDN and 

DN types. These classes can clone their objects by passing an object 

of the same type to the constructor. For example: 

  dn1 = DN(('cn', 'Bob')) # dn1 is immutable 

  dn2 = EditableDN(dn1)   # dn2 is mutable copy of dn1, 

                          # equal to dn1 until it's modified 

  and visa-versa 

  dn1 = EditableDN(('cn', 'Bob')) # dn1 is mutable 

  dn2 = DN(dn1)                   # dn2 is immutable copy of dn1, equal to dn1 

''' 

from ldap.dn import str2dn, dn2str 

from ldap import DECODING_ERROR 

import sys 

__all__ = ['AVA', 'EditableAVA', 'RDN', 'EditableRDN', 'DN', 'EditableDN'] 

def _adjust_indices(start, end, length): 

    'helper to fixup start/end slice values' 

    if end > length: 

        end = length 

    elif end < 0: 

        end += length 

        if end < 0: 

            end = 0 

    if start < 0: 

        start += length 

        if start < 0: 

            start = 0 

    return start, end 

class AVA(object): 

    ''' 

    AVA(arg0, ...) 

    An AVA is an LDAP Attribute Value Assertion. It is convenient to think of 

    AVA's as a <attr,value> pair. AVA's are members of RDN's (Relative 

    Distinguished Name). 

    The AVA constructor is passed a sequence of args and a set of 

    keyword parameters used for configuration. 

    The arg sequence may be: 

    1) With 2 arguments, the first argument will be the attr, the 2nd 

    the value. Each argument must be scalar convertable to unicode. 

    2) With a sigle list or tuple argument containing exactly 2 items. 

    Each item must be scalar convertable to unicode. 

    3) With a single string (or unicode) argument, in this case the string will 

    be interpretted using the DN syntax described in RFC 4514 to yield a AVA 

    <attr,value> pair. The parsing recognizes the DN syntax escaping rules. 

    For example: 

    ava = AVA('cn', 'Bob')      # case 1: two strings 

    ava = AVA(('cn', 'Bob'))    # case 2: 2-valued tuple 

    ava = AVA(['cn', 'Bob'])    # case 2: 2-valued list 

    ava = AVA('cn=Bob')         # case 3: DN syntax 

    AVA object have two properties for accessing their data: 

    attr:  the attribute name, cn in our exmaple 

    value: the attribute's value, Bob in our example 

    When attr and value are returned they will always be unicode. When 

    attr or value are set they will be promoted to unicode. 

    AVA objects support indexing by name, e.g. 

    ava['cn'] 

    returns the value (Bob in our example). If the index does key does not match 

    the attr then a KeyError will be raised. 

    AVA objects support equality testing and comparsion (e.g. cmp()). When they 

    are compared the attr is compared first, if the 2 attr's are equal then the 

    values are compared. The comparison is case insensitive (because attr's map 

    to numeric OID's and their values derive from from the 'name' atribute type 

    (OID 2.5.4.41) whose EQUALITY MATCH RULE is caseIgnoreMatch. 

    The str method of an AVA returns the string representation in RFC 4514 DN 

    syntax with proper escaping. 

    ''' 

    is_mutable = False 

    flags = 0 

    def __init__(self, *args, **kwds): 

        if len(args) == 1: 

            arg = args[0] 

            if isinstance(arg, AVA): 

                ava = (arg.attr, arg.value) 

            elif isinstance(arg, basestring): 

                try: 

                    rdns = str2dn(arg.encode('utf-8')) 

                except DECODING_ERROR: 

                    raise ValueError("malformed AVA string = \"%s\"" % arg) 

                if len(rdns) != 1: 

                    raise ValueError("multiple RDN's specified by \"%s\"" % (arg)) 

                rdn = rdns[0] 

                if len(rdn) != 1: 

                    raise ValueError("multiple AVA's specified by \"%s\"" % (arg)) 

                ava = rdn[0] 

            elif isinstance(arg, (tuple, list)): 

                ava = arg 

                if len(ava) != 2: 

                    raise ValueError("tuple or list must be 2-valued, not \"%s\"" % (ava)) 

            else: 

                raise TypeError("with 1 argument, argument must be str,unicode,tuple or list, got %s instead" % \ 

                                arg.__class__.__name__) 

            attr  = ava[0] 

            value = ava[1] 

        elif len(args) == 2: 

            attr  = args[0] 

            value = args[1] 

        else: 

            raise TypeError("takes 1 or 2 arguments (%d given)" % (len(args))) 

        self._set_attr(attr) 

        self._set_value(value) 

    def _get_attr(self): 

        return self._attr_unicode 

    def _set_attr(self, new_attr): 

        # Scalars only 

        if isinstance(new_attr, (tuple, list)): 

            raise TypeError("attr must be scalar, got %s" % type(new_attr)) 

        try: 

            if isinstance(new_attr, unicode): 

                self._attr_unicode = new_attr 

            elif isinstance(new_attr, str): 

                self._attr_unicode = new_attr.decode('utf-8') 

            else: 

                self._attr_unicode = unicode(new_attr) 

        except Exception, e: 

            raise ValueError('unable to convert attr "%s" to unicode: %s' % (new_attr, e)) 

    attr  = property(_get_attr) 

    def _get_value(self): 

        return self._value_unicode 

    def _set_value(self, new_value): 

        # Scalars only 

        if isinstance(new_value, (tuple, list)): 

            raise TypeError("value must be scalar, got %s" % type(new_value)) 

        try: 

            if isinstance(new_value, unicode): 

                self._value_unicode  = new_value 

            elif isinstance(new_value, str): 

                self._value_unicode  = new_value.decode('utf-8') 

            else: 

                self._value_unicode  = unicode(new_value) 

        except Exception, e: 

            raise ValueError('unable to convert value "%s" to unicode: %s' % (new_value, e)) 

    value = property(_get_value) 

    def _to_openldap(self): 

        return [[(self._attr_unicode.encode('utf-8'), self._value_unicode.encode('utf-8'), self.flags)]] 

    def __str__(self): 

        return dn2str(self._to_openldap()) 

    def __repr__(self): 

        return "%s.%s('%s')" % (self.__module__, self.__class__.__name__, self.__str__()) 

    def __getitem__(self, key): 

        if isinstance(key, basestring): 

            if key == self._attr_unicode: 

                return self._value_unicode 

            raise KeyError("\"%s\" not found in %s" % (key, self.__str__())) 

        else: 

            raise TypeError("unsupported type for AVA indexing, must be basestring; not %s" % \ 

                                (key.__class__.__name__)) 

    def __hash__(self): 

        # Hash is computed from AVA's string representation because it's immutable. 

        # 

        # Because attrs & values are comparison case-insensitive the 

        # hash value between two objects which compare as equal but 

        # differ in case must yield the same hash value. 

        return hash(str(self).lower()) 

    def __eq__(self, other): 

        ''' 

        The attr comparison is case insensitive because attr is 

        really an LDAP attribute type which means it's specified with 

        an OID (dotted number) and not a string. Since OID's are 

        numeric the human readable name which maps to the OID is not 

        significant in case. 

        The value comparison is also case insensitive because the all 

        attribute types used in a DN are derived from the 'name' 

        atribute type (OID 2.5.4.41) whose EQUALITY MATCH RULE is 

        caseIgnoreMatch. 

        ''' 

        # Try coercing string to AVA, if successful compare to coerced object 

        if isinstance(other, basestring): 

            try: 

                other_ava = AVA(other) 

                return self.__eq__(other_ava) 

            except Exception: 

                return False 

        # If it's not an AVA it can't be equal 

        if not isinstance(other, AVA): 

            return False 

        # Perform comparison between objects of same type 

        return self._attr_unicode.lower() == other.attr.lower() and \ 

            self._value_unicode.lower() == other.value.lower() 

    def __ne__(self, other): 

        return not self.__eq__(other) 

    def __cmp__(self, other): 

        'comparison is case insensitive, see __eq__ doc for explanation' 

        if not isinstance(other, AVA): 

            raise TypeError("expected AVA but got %s" % (other.__class__.__name__)) 

        result = cmp(self._attr_unicode.lower(), other.attr.lower()) 

        if result != 0: 

            return result 

        result = cmp(self._value_unicode.lower(), other.value.lower()) 

        return result 

class EditableAVA(AVA): 

    ''' 

    Exactly identical to the AVA class except 

    * Hash value is based on object identity, not object 

      value. Objects that test as equal will be non-unique when 

      used as a dict key or member of a set. 

    * The attr and value properties may be modified after object creation. 

    ''' 

    is_mutable = True 

    __hash__ = None 

    attr  = property(AVA._get_attr, AVA._set_attr) 

    value = property(AVA._get_value, AVA._set_value) 

class RDN(object): 

    ''' 

    RDN(arg0, ...) 

    An RDN is a LDAP Relative Distinguished Name. RDN's are members of DN's 

    (Distinguished Name). An RDN contains 1 or more AVA's. If the RDN contains 

    more than one AVA it is said to be a multi-valued RDN. When an RDN is 

    multi-valued the AVA's are unorderd comprising a set. However this 

    implementation orders the AVA's according to the AVA comparison function to 

    make equality and comparison testing easier. Think of this a canonical 

    normalization (however LDAP does not impose any ordering on multiple AVA's 

    within an RDN). Single valued RDN's are the norm and thus the RDN 

    constructor has simple syntax for them. 

    The RDN constructor is passed a sequence of args and a set of 

    keyword parameters used for configuration. 

    The constructor iterates though the sequence and adds AVA's to the RDN. 

    The arg sequence may be: 

    * A 2-valued tuple or list denotes the <attr,value> pair of an AVA. The 

    first member is the attr and the second member is the value, both members 

    must be strings (or unicode). The tuple or list is passed to the AVA 

    constructor and the resulting AVA is added to the RDN. Multiple tuples or 

    lists may appear in the argument list, each adds one additional AVA to the 

    RDN. 

    * A single string (or unicode) argument, in this case the string will 

    be interpretted using the DN syntax described in RFC 4514 to yield one or 

    more AVA <attr,value> pairs. The parsing recognizes the DN syntax escaping 

    rules. 

    * A AVA object, the AVA will be copied into the new RDN respecting 

      the constructors keyword configuration parameters. 

    * A RDN object, the AVA's in the RDN are copied into the new RDN 

      respecting the constructors keyword configuration parameters. 

    Single AVA Examples: 

    RDN(('cn', 'Bob'))                  # tuple yields 1 AVA 

    RDN('cn=Bob')                       # DN syntax with 1 AVA 

    RDN(AVA('cn', 'Bob'))               # AVA object adds 1 AVA 

    Multiple AVA Examples: 

    RDN(('cn', 'Bob'),('ou', 'people')) # 2 tuples yields 2 AVA's 

    RDN('cn=Bob+ou=people')             # DN syntax with 2 AVA's 

    RDN(AVA('cn', 'Bob'),AVA('ou', 'people')) # 2 AVA objects adds 2 AVA's 

    RDN(('cn', 'Bob'), 'ou=people')     # 2 args, 1st tuple forms 1 AVA, 

                                        # 2nd DN syntax string adds 1 AVA, 

                                        # 2 AVA's in total 

    Note: The RHS of a slice assignment is interpreted exactly in the 

    same manner as the constructor argument list (see above examples). 

    RDN objects support iteration over their AVA members. You can iterate all 

    AVA members via any Python iteration syntax. RDN objects support full Python 

    indexing using bracket [] notation. Examples: 

    len(rdn)            # return the number of AVA's 

    rdn[0]              # indexing the first AVA 

    rdn['cn']           # index by AVA attr, returns AVA value 

    for ava in rdn:     # iterate over each AVA 

    rdn[:]              # a slice, in this case a copy of each AVA 

    WARNING: When indexing by attr (e.g. rdn['cn']) there is a possibility more 

    than one AVA has the same attr name as the index key. The default behavior 

    is to return the value of the first AVA whose attr matches the index 

    key. 

    RDN objects support the AVA attr and value properties as another programmer 

    convenience because the vast majority of RDN's are single valued. The attr 

    and value properties return the attr and value properties of the first AVA 

    in the RDN, for example: 

    rdn = RDN(('cn', 'Bob')) # rdn has 1 AVA whose attr == 'cn' and value == 'Bob' 

    len(rdn) -> 1 

    rdn.attr -> u'cn'      # exactly equivalent to rdn[0].attr 

    rdn.value -> u'Bob'    # exactly equivalent to rdn[0].value 

    When attr and value are returned they will always be unicode. When 

    attr or value are set they will be promoted to unicode. 

    If an RDN is multi-valued the attr and value properties still return only 

    the first AVA's properties, programmer beware! Recall the AVA's in the RDN 

    are sorted according the to AVA collating semantics. 

    RDN objects support equality testing and comparison. See AVA for the 

    definition of the comparison method. 

    RDN objects support concatenation and addition with other RDN's or AVA's 

    rdn1 + rdn2 # yields a new RDN object with the contents of each RDN. 

    rdn1 + ava1 # yields a new RDN object with the contents of rdn1 and ava1 

    RDN objects can add AVA's objects via in-place addition. 

    rdn1 += rdn2 # rdn1 now contains the sum of rdn1 and rdn2 

    rdn1 += ava1 # rdn1 has ava1 added to it. 

    The str method of an RDN returns the string representation in RFC 4514 DN 

    syntax with proper escaping. 

    ''' 

    is_mutable = False 

    flags = 0 

    AVA_type = AVA 

    def __init__(self, *args, **kwds): 

        self.avas = self._avas_from_sequence(args) 

        self.avas.sort() 

    def _ava_from_value(self, value): 

        if isinstance(value, AVA): 

            return self.AVA_type(value.attr, value.value) 

        elif isinstance(value, RDN): 

            avas = [] 

            for ava in value.avas: 

                avas.append(self.AVA_type(ava.attr, ava.value)) 

            if len(avas) == 1: 

                return avas[0] 

            else: 

                return avas 

        elif isinstance(value, basestring): 

            try: 

                rdns = str2dn(value.encode('utf-8')) 

                if len(rdns) != 1: 

                    raise ValueError("multiple RDN's specified by \"%s\"" % (value)) 

                rdn = rdns[0] 

                if len(rdn) == 1: 

                    return self.AVA_type(rdn[0][0], rdn[0][1]) 

                else: 

                    avas = [] 

                    for ava_tuple in rdn: 

                        avas.append(self.AVA_type(ava_tuple[0], ava_tuple[1])) 

                    return avas 

            except DECODING_ERROR: 

                raise ValueError("malformed RDN string = \"%s\"" % value) 

        elif isinstance(value, (tuple, list)): 

            if len(value) != 2: 

                raise ValueError("tuple or list must be 2-valued, not \"%s\"" % (value)) 

            return self.AVA_type(value) 

        else: 

            raise TypeError("must be str,unicode,tuple, or AVA, got %s instead" % \ 

                            value.__class__.__name__) 

    def _avas_from_sequence(self, seq): 

        avas = [] 

        for item in seq: 

            ava = self._ava_from_value(item) 

            if isinstance(ava, list): 

                avas.extend(ava) 

            else: 

                avas.append(ava) 

        return avas 

    def _to_openldap(self): 

        return [[(ava.attr.encode('utf-8'), ava.value.encode('utf-8'), self.flags) for ava in self.avas]] 

    def __str__(self): 

        return dn2str(self._to_openldap()) 

    def __repr__(self): 

        return "%s.%s('%s')" % (self.__module__, self.__class__.__name__, self.__str__()) 

    def _next(self): 

        for ava in self.avas: 

            yield ava 

    def __iter__(self): 

        return self._next() 

    def __len__(self): 

        return len(self.avas) 

    def __getitem__(self, key): 

        if isinstance(key, (int, long, slice)): 

            return self.avas[key] 

        elif isinstance(key, basestring): 

            for ava in self.avas: 

                if key == ava.attr: 

                    return ava.value 

            raise KeyError("\"%s\" not found in %s" % (key, self.__str__())) 

        else: 

            raise TypeError("unsupported type for RDN indexing, must be int, basestring or slice; not %s" % \ 

                                (key.__class__.__name__)) 

    def _get_attr(self): 

        if len(self.avas) == 0: 

            raise IndexError("No AVA's in this RDN") 

        return self.avas[0].attr 

    def _set_attr(self, new_attr): 

        if len(self.avas) == 0: 

            raise IndexError("No AVA's in this RDN") 

        self.avas[0].attr = new_attr 

    attr  = property(_get_attr) 

    def _get_value(self): 

        if len(self.avas) == 0: 

            raise IndexError("No AVA's in this RDN") 

        return self.avas[0].value 

    def _set_value(self, new_value): 

        if len(self.avas) == 0: 

            raise IndexError("No AVA's in this RDN") 

        self.avas[0].value = new_value 

    value = property(_get_value) 

    def __hash__(self): 

        # Hash is computed from RDN's string representation because it's immutable 

        # 

        # Because attrs & values are comparison case-insensitive the 

        # hash value between two objects which compare as equal but 

        # differ in case must yield the same hash value. 

        return hash(str(self).lower()) 

    def __eq__(self, other): 

        # Try coercing string to RDN, if successful compare to coerced object 

        if isinstance(other, basestring): 

            try: 

                other_rdn = RDN(other) 

                return self.__eq__(other_rdn) 

            except Exception: 

                return False 

        # If it's not an RDN it can't be equal 

        if not isinstance(other, RDN): 

            return False 

        # Perform comparison between objects of same type 

        return self.avas == other.avas 

    def __ne__(self, other): 

        return not self.__eq__(other) 

    def __cmp__(self, other): 

        if not isinstance(other, RDN): 

            raise TypeError("expected RDN but got %s" % (other.__class__.__name__)) 

        result = cmp(len(self), len(other)) 

        if result != 0: 

            return result 

        i = 0 

        while i < len(self): 

            result = cmp(self[i], other[i]) 

            if result != 0: 

                return result 

            i += 1 

        return 0 

    def __add__(self, other): 

        result = self.__class__(self) 

        if isinstance(other, RDN): 

            for ava in other.avas: 

                result.avas.append(self.AVA_type(ava.attr, ava.value)) 

        elif isinstance(other, AVA): 

            result.avas.append(self.AVA_type(other.attr, other.value)) 

        elif isinstance(other, basestring): 

            rdn = self.__class__(other) 

            for ava in rdn.avas: 

                result.avas.append(self.AVA_type(ava.attr, ava.value)) 

        else: 

            raise TypeError("expected RDN, AVA or basestring but got %s" % (other.__class__.__name__)) 

        result.avas.sort() 

        return result 

class EditableRDN(RDN): 

    ''' 

    Exactly identical to the RDN class except 

    * Hash value is based on object identity, not object 

      value. Objects that test as equal will be non-unique when 

      used as a dict key or member of a set. 

    * AVA components may be assigned via assignment statements. 

    * In-place addition modifes the lhs object. 

    * The attr and value properties may be modified after object creation. 

    ''' 

    is_mutable = True 

    __hash__ = None 

    AVA_type = EditableAVA 

    def __setitem__(self, key, value): 

        if isinstance(key, (int, long)): 

            new_ava = self._ava_from_value(value) 

            if isinstance(new_ava, list): 

                raise TypeError("cannot assign multiple AVA's to single entry") 

            self.avas[key] = new_ava 

        elif isinstance(key, slice): 

            avas = self._avas_from_sequence(value) 

            self.avas[key] = avas 

        elif isinstance(key, basestring): 

            new_ava = self._ava_from_value(value) 

            if isinstance(new_ava, list): 

                raise TypeError("cannot assign multiple AVA's to single entry") 

            found = False 

            i = 0 

            while i < len(self.avas): 

                if key == self.avas[i].attr: 

                    found = True 

                    self.avas[i] = new_ava 

                    break 

                i += 1 

            if not found: