UserDict clear solution for Cipher Crossword by gyahun_dash - python coding challenges

Enable Javascript in your browser and then refresh this page, for a much enhanced experience.

UserDict solution in Clear category for Cipher Crossword by gyahun_dash

from collections import UserDict
from itertools import chain, permutations

class ConsistentDict(UserDict):
    def __setitem__(self, key, item):
        if not key in self.keys(): self.data[key] = item
        elif self[key] != item: raise ValueError #prohibit updates

def enumerate_numbers_on(grid): #line: rows[0, 2, 4], columns[0, 2, 4]
    for line in chain(grid[::2], list(zip(*grid))[::2]): yield from line

#arrange numbers in a line, and find consistent mapping(brute force)
def checkio(crossword, words):
    line = list(enumerate_numbers_on(crossword))
    for candidate_words in permutations(words):
        try: cdict = ConsistentDict(zip(line, ''.join(candidate_words)))
        except ValueError: continue
        else: return [[cdict.get(n, ' ') for n in row] for row in crossword]

March 16, 2014

Comments:

gyahun_dash on March 16, 2014, 2 p.m.
I've tried to let ConsistentDict inherit dict, but it doesn't work: <pre class='brush: python'>class ConsistentDict(dict): def __setitem__(self, key, item): if not key in self.keys(): dict.__setitem__(self, key, item) elif self[key] != item: raise ValueError #prohibit updates </pre> In this case, ConsistentDict doesn't call overridden __setitem__. I guess: when argument is zip(keys, values), dict calls operator.setitem or anything...

veky on March 30, 2014, 7 p.m.
Well, that's exactly why UserDict exists (precisely, why it still exists in Py3k). It is much closer to a naive model of how dict constructor should work. Native dict constructor just bypasses __setitem__, because dict was written without thought someone would inherit from it initially (for a long time it was impossible). It doesn't call operator.setitem either, it reaches much deeper in the API.

gyahun_dash on April 1, 2014, 2:41 p.m.
<blockquote> dict was written without thought someone would inherit from it initially </blockquote> Wow, I've not known that. Exactly, I've wondered UserDict exists in Py3k. I hope they will be integrated someday. Such situations make me confused... Thank you.

veky on April 1, 2014, 4:11 p.m.
Well, I don't think they will be "integrated", precisely because of such things as your example. It's probably much faster if constructor doesn't add items one by one (calling setitem hook), and a very important use case is every time you call a function with named parameters (** specification). Slowing it down just so you could have your mental model is not very realistic, even more because nobody ever said setitem hook is called in such occasions - precisely, docs say: <pre class='brush: python'>object.__setitem__(self, key, value) </pre> <blockquote> Called to implement assignment to self[key]. [...] This should only be implemented for mappings if the objects support changes to the values for keys, or if new keys can be added, or for sequences if elements can be replaced. </blockquote> So you see, it should really be called addorreplaceitem :-D. There was no intention that constructor calls it. But a more important point is this: even a UserDict (that could probably be more accurately called NaiveDict:)) doesn't do exactly what you want in all cases (specifically, update method): see <a href="http://stackoverflow.com/questions/2060972/subclassing-python-dictionary-to-override-setitem">http://stackoverflow.com/questions/2060972/subclassing-python-dictionary-to-override-setitem</a> That is probably something that should be fixed. Even better, you can do it yourself probably. :-)

gyahun_dash on April 3, 2014, 3:14 p.m.
<blockquote> Well, I don't think they will be "integrated", precisely because of such things as your example. It's probably much faster if constructor doesn't add items one by one (calling setitem hook), and a very important use case is every time you call a function with named parameters (** specification). </blockquote> That makes sense. <blockquote> But a more important point is this: even a UserDict (that could probably be more accurately called NaiveDict:)) doesn't do exactly what you want in all cases (specifically, update method): see <a href="http://stackoverflow.com/questions/2060972/subclassing-python-dictionary-to-override-setitem">http://stackoverflow.com/questions/2060972/subclassing-python-dictionary-to-override-setitem</a> That is probably something that should be fixed. Even better, you can do it yourself probably. :-) </blockquote> I see, I'll take more care of dict's behaviors for next time. Thank you for your advice. (Dictionaries exactly taught me a good lesson)

vinc on Aug. 30, 2014, 3:04 p.m.
Nice usage of UserDict. Thanks for practice.

gyahun_dash on Sept. 2, 2014, 12:02 p.m.
Thank you.

2010min0ru on Oct. 22, 2020, 11:48 a.m.
Awesome solution! That is exactly the same method of a bruteforce that i was using. But your is about 3 times shorter. Great work.

cerankas on Oct. 16, 2023, 4:27 p.m.
Great solution, brief and educational. Thanks.

Pavellver on Nov. 1, 2023, 7:33 p.m.
Very interesting solution, learned a lot for myself. Thank you very much for this!

List, list and list again

0 %

PyCharm The Python IDE for Professional Developers. Enjoy productive Python, web and scientific development with PyCharm. Download it now!