Create python 2d array indexed by string

Question

(I'm not too great at python, so forgive me if this is a stupid question)

So I want to create a data structure that represents this -

      word1   word2
word3   1      2
word4   3      4

Right now I've tried doing something like this -

self.table = [][]

but this is giving me an invalid syntax error(I guess because I haven't initialized the arrays?). However, even if I were to use this I wouldn't be able to use it because I don't know how large my x and y dimension is(it seems like there would be an array out of index exception).

Should I be using a double dictionary? What should I be using?

Answer 1

The idea is to create a dictionary, which maps the strings to the indeces. In the following there's a little class which overloads the '[ ]'-operator:

class ArrayStrIdx:
    """ 2D Array with string as Indeces

    Usage Example::

        AA = ArrayStrIdx()
        AA["word1", "word3"] = 99
        print(AA["word2", "word5"])
    """

    cols ="word1 word2"
    rows ="word3 word4 word5"

    dd = [[10,11,12],[20,21,22]]  # data

    def __init__(self):
        """ Constructor builds dicts for indexes"""
        self.ri = dict([(w,i) for i,w in enumerate(self.rows.split())])
        self.ci = dict([(w,i) for i,w in enumerate(self.cols.split())])

    def _parsekey(self, key):
        """ Convert index key to indeces for ``self.dd`` """
        w0, w1 = key[0], key[1]
        return self.ci[w0], self.ri[w1]

    def __getitem__(self, key):
        """ overload [] operator - get() method"""
        i0, i1 = self._parsekey(key)
        return self.dd[i0][i1]

    def __setitem__(self, key, value):
        """ overload [] operator -  set() method """
        i0, i1 = self._parsekey(key)
        self.dd[i0][i1] = value

Update: Expanded answer to allow something like AA["word1", "word3"] = 23 .

Answer 2

Maybe you can try initializing your table with

self.table = {r : { c : 0 for c in ['word1', 'word2']} for r in ['word3', 'word4']}

and then you can access each position by

self.table['word3']['word1'] = 2

Answer 3

Python doesn't have a ready-made instruction to create a bi-dimensional matrix, but it's easy to do that using list comprehensions:

matrix = [[0] * cols for i in range(rows)]

then you can access the matrix with, for example

matrix[row][col] += 42

assuming rows=10 and cols=20 then the row index must go from 0 to 9 and the col index from 0 to 19. You can also use negative indexes to mean counting from the end; for example

matrix[-1][-1] = 99

will set the last cell to 99

Answer 4

If you are not opposed to using an external library, you might check out

Pandas Data Frames