A default value for function args.
Use this when you need to distinguish between None and no value.
Example:
>>> def foo(arg=NotGiven):
... print arg is NotGiven
...
>>> foo()
True
>>> foo(None)
False
A container for arbitrary attributes.
Usage:
>>> do = DumbObject(a=1, b=2)
>>> do.b
2
Alternatives to this class include collections.namedtuple in Python 2.6, and formencode.declarative.Declarative in Ian Bicking’s FormEncode package. Both alternatives offer more features, but DumbObject shines in its simplicity and lack of dependencies.
I count the number of occurrences of each value registered with me.
Call the instance to register a value. The result is available as the .result attribute. Example:
>>> counter = Counter()
>>> counter("foo")
>>> counter("bar")
>>> counter("foo")
>>> sorted(counter.result.items())
[('bar', 1), ('foo', 2)]
>> counter.result
{'foo': 2, 'bar': 1}
To see the most frequently-occurring items in order:
>>> counter.get_popular(1)
[(2, 'foo')]
>>> counter.get_popular()
[(2, 'foo'), (1, 'bar')]
Or if you prefer the list in item order:
>>> counter.get_sorted_items()
[('bar', 1), ('foo', 2)]
Build a Counter from an iterable in one step.
This is the same as adding each item individually.
>>> counter = Counter.correlate(["A", "B", "A"])
>>> counter.result["A"]
2
>>> counter.result["B"]
1
Return the results as as a list of (count, item) pairs, with the most frequently occurring items first.
If max_items is provided, return no more than that many items.
Return the result as a list of (item, count) pairs sorted by item.
Accumulate a dict of all values for each key.
Call the instance to register a value. The result is available as the .result attribute. Example:
>>> bowling_scores = Accumulator()
>>> bowling_scores("Fred", 0)
>>> bowling_scores("Barney", 10)
>>> bowling_scores("Fred", 1)
>>> bowling_scores("Barney", 9)
>>> sorted(bowling_scores.result.items())
[('Barney', [10, 9]), ('Fred', [0, 1])]
>> bowling_scores.result
{'Fred': [0, 1], 'Barney': [10, 9]}
The values are stored in the order they’re registered.
Alternatives to this class include paste.util. multidict.MultiDict in Ian Bicking’s Paste package.
Create an Accumulator based on several related values.
key is a function to calculate the key for each item, akin to list.sort(key=).
This is the same as adding each item individually.
Accumulate a dict of unique values for each key.
The values are stored in an unordered set.
Call the instance to register a value. The result is available as the .result attribute.
Return a list of unique elements in the iterable, preserving the order.
Usage:
>>> unique([None, "spam", 2, "spam", "A", "spam", "spam", "eggs", "spam"])
[None, 'spam', 2, 'A', 'eggs']
Return a copy of the dict with only the specified keys present.
dic may be any mapping. The return value is always a Python dict.
>> only_some_keys({"A": 1, "B": 2, "C": 3}, ["A", "C"])
>>> sorted(only_some_keys({"A": 1, "B": 2, "C": 3}, ["A", "C"]).items())
[('A', 1), ('C', 3)]
Return a copy of the dict without the specified keys.
>>> except_keys({"A": 1, "B": 2, "C": 3}, ["A", "C"])
{'B': 2}
Return two copies of the dict. The first has only the keys specified. The second has all the other keys from the original dict.
>> extract_keys({"From": "F", "To": "T", "Received", R"}, ["To", "From"])
({"From": "F", "To": "T"}, {"Received": "R"})
>>> regular, extra = extract_keys({"From": "F", "To": "T", "Received": "R"}, ["To", "From"])
>>> sorted(regular.keys())
['From', 'To']
>>> sorted(extra.keys())
['Received']
Like dict.iteritems() but with a specified key order.
Arguments:
This yields the items listed in key_order. If a key does not exist in the dict, yield the default value if specified, otherwise skip the missing key. Afterwards, if other_keys is true, yield the remaining items in an arbitrary order.
Usage:
>>> dic = {"To": "you", "From": "me", "Date": "2008/1/4", "Subject": "X"}
>>> dic["received"] = "..."
>>> order = ["From", "To", "Subject"]
>>> list(ordered_items(dic, order, False))
[('From', 'me'), ('To', 'you'), ('Subject', 'X')]
Delete several keys from a dict, ignoring those that don’t exist.
This modifies the dict in place.
>>> d ={"A": 1, "B": 2, "C": 3}
>>> del_quiet(d, ["A", "C"])
>>> d
{'B': 2}
Correlate several dicts under one superdict.
If you have several dicts each with a ‘name’ key, this puts them in a container dict keyed by name.
>>> d1 = {"name": "Fred", "age": 41}
>>> d2 = {"name": "Barney", "age": 31}
>>> flintstones = correlate_dicts([d1, d2], "name")
>>> sorted(flintstones.keys())
['Barney', 'Fred']
>>> flintstones["Fred"]["age"]
41
If you’re having trouble spelling this method correctly, remember: “relate” has one ‘l’. The ‘r’ is doubled because it occurs after a prefix. Thus “correlate”.
Correlate several objects under one dict.
If you have several objects each with a ‘name’ attribute, this puts them in a dict keyed by name.
>>> class Flintstone(DumbObject):
... pass
...
>>> fred = Flintstone(name="Fred", age=41)
>>> barney = Flintstone(name="Barney", age=31)
>>> flintstones = correlate_objects([fred, barney], "name")
>>> sorted(flintstones.keys())
['Barney', 'Fred']
>>> flintstones["Barney"].age
31
If you’re having trouble spelling this method correctly, remember: “relate” has one ‘l’. The ‘r’ is doubled because it occurs after a prefix. Thus “correlate”.
Distribute a list into a N-column table (list of lists).
lis is a list of values to distribute.
columns is an int greater than 1, specifying the number of columns in the table.
direction is a string beginning with “H” (horizontal) or “V” (vertical), case insensitive. This affects how values are distributed in the table, as described below.
fill is a value that will be placed in any remaining cells if the data runs out before the last row or column is completed. This must be an immutable value such as None , "", 0, “ ”, etc. If you use a mutable value like [] and later change any cell containing the fill value, all other cells containing the fill value will also be changed.
The return value is a list of lists, where each sublist represents a row in the table. table[0] is the first row. table[0][0] is the first column in the first row. table[0][1] is the second column in the first row.
This can be displayed in an HTML table via the following Mako template:
<table>
% for row in table:
<tr>
% for cell in row:
<td>${cell}</td>
% endfor cell
</tr>
% endfor row
</table>
In a horizontal table, each row is filled before going on to the next row. This is the same as dividing the list into chunks:
>>> distribute([1, 2, 3, 4, 5, 6, 7, 8], 3, "H")
[[1, 2, 3], [4, 5, 6], [7, 8, None]]
In a vertical table, the first element of each sublist is filled before going on to the second element. This is useful for displaying an alphabetical list in columns, or when the entire column will be placed in a single <td> with a <br /> between each element:
>>> food = ["apple", "banana", "carrot", "daikon", "egg", "fish", "gelato", "honey"]
>>> table = distribute(food, 3, "V", "")
>>> table
[['apple', 'daikon', 'gelato'], ['banana', 'egg', 'honey'], ['carrot', 'fish', '']]
>>> for row in table:
... for item in row:
... print "%-9s" % item,
... print "." # To show where the line ends.
...
apple daikon gelato .
banana egg honey .
carrot fish .
Alternatives to this function include a NumPy matrix of objects.