uplook 1.1.1

An opinionated Python module to handle configuration values with support for external data sources.

What?

An opinionated Python module to handle configuration values with support for external data sources.

How?

The UpLook class takes an arbitrary number of key/values. These variables are stored and accessible as regular object attributes. Values of type <<dict>> can be accessed in dotted format.

A value can also be a lookup function by using a special syntax:

    +-> The name of the funtion.  1 tilde means the function is a static lookup.
    |   2 tildes means a lookup is done every time the attribute is accessed.
    |
    |         +------------+------> An optional pair of values.  The first value is the <key>
    |         |         |           to lookup.  The second value is a default value to return
    |         |         |           in case the key lookup fails.
~~function("key", "default value")
                     |
                     |
                     +-> The default value can also be JSON.  In that case you should not use
                         quotes around the value.

An actual example would look like:

>>>> u = UpLook(servers='~~consul("first", [])')

For this to work you first need to register the function responsible for retrieving the data. Depending what is required to lookup the desired value you could feed a <key> value to the function.

An (silly) example lookup function looks like:

def show_members(cluster_name):
    if cluster_name == "first"
        return ["one", "two", "three"]
    else:
        raise Exception("Unknown cluster")

>>>> u.registerLookup("consul", show_members)

Accessing value u.value.server will then trigger the the registered lookup function and return its value:

>>>> u.value.servers
["one", "two", "three"]

When the registered function return an Exception, the default value is returned:

>>>> u = UpLook(servers='~~consul("blahblahblah", [])')
>>>> u.value.servers
[]

Remarks

• When executing the lookup function fails and a default value has been defined, no error is raised and the default value is returned.

• When executing the lookup function fails and no default value has been defined an error is raised.

• When the default value is between single or double quotes, it is returned literally. When the default value is not between quotes, it is considered to be JSON. When The JSON is invalid, an error is returned.

More examples

Provided kwargs are accessible as attributes

>>>> from uplook import UpLook
>>>> u = UpLook(one=1, two=2)
>>>> u
UpLook({'two': 2, 'one': 1})
>>>> u.value.one
1

Dict argument values are recursively mapped to attributes

>>>> from uplook import UpLook
>>>> u = UpLook(levels = {"level1":{"level2":{"level3": "hello"}}})
>>>> u
UpLook({'levels': {'level1': {'level2': {'level3': 'hello'}}}})
>>>> u.value.levels.level1.level2.level3
'hello'

Get the data portion without all helper methods

>>>> from uplook import UpLook
>>>> u = UpLook(one=1, two=2)
>>>> u
UpLook({'two': 2, 'one': 1})
>>>> data = u.get()
>>>> data.one
1

Get a simple dict representation of the data

>>>> from uplook import UpLook
>>>> u = UpLook(one=1, two=2)
>>>> u
UpLook({'two': 2, 'one': 1})
>>>> data = u.dump()
{'two': 2, 'one': 1}

Iterate over key/value pairs of a data container

>>>> from uplook import UpLook
>>>> u = UpLook(one=1, two=2)
>>>> u
UpLook({'two': 2, 'one': 1})
>>>> for key, value in u.get():
....     print "key: %s, value: %s" % (key, value)
....
key: two, value: 2
key: one, value: 1

External lookup values

Some value lookup function

from uplook import UpLook
from uplook.errors import NoSuchValue


def someLookupFunction(key):
    data = {"value.number.one": "hi",
            "value.number.two": "this",
            "value.number.three": "is",
            "value.number.four": "a",
            "value.number.five": "silly",
            "value.number.six": "demo"
            }

    try:
        return data[key]
    except KeyError:
        raise NoSuchValue("%s is an unknown value." % (key))


def randomInt(max):
    return random.randint(0, max)

Initialize an Uplook instance with a dynamic and static lookup

>>> instance = UpLook(static='~fubar("value.number.one", "unknown")',
>>>                   dynamic='~~random(100)',
>>>                   normal='hello')

List all user defined lookup functions

>>> for function in instance.listfunctions():
        print function
fubar
random
>>>

Register lookup functions

>>> instance.registerLookup("fubar", someLookupFunction)
>>> instance.registerLookup("random", randomInt)

Access a static lookup value

>>> print test.value.static
hi
>>> print test.value.static
hi

Access a dynamic lookup value

>>> print test.value.dynamic
>>> 81
>>> print test.value.dynamic
>>> 16