dictionary or map with either string or integer as key in python?

Question

This might be a silly question, but for some reason the solution escapes me at the moment.

I would like to have fast and efficient access to data that is in a list format. So for example a list of questions:

q = {}
q[1] = "my first string"
q[2] = "my second string"
q[3] = "my third string"

I can easily find what question 2's string is by doing q[2]. But I would also like to retrieve the question number by indexing q with the string:

q["my second string"] -> gives 2 as answer

I would like to do this without iterating over the keys (defeats the purpose of a dictionary) and like to avoid defining a second dictionary using the string as the key to avoid wasted memory. Is this possible?

Ultimately the reason for this is I would like to access say q[2] or q["my second string"] and get the data associated with question 2, whether using the number or the string as a key to that data. Is this possible without having to iterating over all the keys while avoiding data duplication?

dicts arn't bidirectional, so you either need a second dict or iterate. there's not other way. — mata, Jun 28 '12 at 07:23
possible duplicate of [Efficient bidirectional hash table in Python?](http://stackoverflow.com/questions/3318625/efficient-bidirectional-hash-table-in-python) — ecatmur, Jun 28 '12 at 07:37

score 2 · Answer 1 · answered Jun 28 '12 at 07:30

2

There's no problem having a mixture of int and str as keys

>>> q = {}
>>> q[1] = "my first string"
>>> q[2] = "my second string"
>>> q[3] = "my third string"
>>> q.update({v:k for k,v in q.items()})
>>> q["my second string"]
2

answered Jun 28 '12 at 07:30

John La Rooy

295,403
53
369
502

I want to avoid data duplication in the dict. – stewart99 Jun 28 '12 at 19:53
@doorfly, there is no data duplication, just the references are duplicated – John La Rooy Jun 28 '12 at 21:22

score 1 · Answer 2 · answered Jun 28 '12 at 07:24

You can use an OrderedDict, but for one of the directions it's not going to be as efficient as a normal dictionary lookup.

from collections import OrderedDict
q = OrderedDict()
q["my first string"] = 1
q["my second string"] = 2
q["my third string"] = 3
# Now you have normal key lookups on your string as a normal dict, and to get the order
q.values()[1]  # To get the second value out
# To get the key, value pair of the second entry
q.items()[1]
# Would return `('my second string', 2)`

score 0 · Answer 3 · edited May 23 '17 at 11:48

0

This was asked at Efficient bidirectional hash table in Python?

The answer remains the same - use bidict from http://pypi.python.org/pypi/bidict

edited May 23 '17 at 11:48

Community

1
1

answered Jun 28 '12 at 07:38

ecatmur

152,476
27
293
366

score 0 · Answer 4 · answered Jun 28 '12 at 07:39

class MyDict(dict):
    def __init__(self, **kwargs):
        super(MyDict, self).__init__(**kwargs)
        for k, v in kwargs.iteritems():
            self[v] = k
    def __setitem__(self, key, val):
        super(MyDict, self).__setitem__(key, val)
        super(MyDict, self).__setitem__(val, key)

d = MyDict(a=1, b=2)
print d[1] # "a"
print d[2] # "b"
d['c'] = 3
print d[3] # "c"

dictionary or map with either string or integer as key in python?

4 Answers4