Convert elements in sublist to keys in dictionary and append to new list with sublists

Question

I'm trying to create a list with sublists of keys from a dictionary based on elements in another list of sublists (data below).

I have a function attempting to follow this logic:

1. See if the element in the sublist matches any exact value in the dictionary
2. If it does, append the associated dictionary key to a sublist within a new list
3. If not, try to convert the element into an integer and append that to a sublist in the new list
4. If that doesn't work, append the whole sublist to an entirely new list
5. Return a list with sublists of integers converted from the strings in the original file and a list with sublists that can't be converted

I hope that logic makes sense. Thanks to some comments on a previous version of this question (Replacing elements in a sublist with keys from a dictionary based on matches with dictionary values), I've edited the code and dictionary so the re.search function now works. Here's the new code below:

import re

def conversion(d, file):
    key_list = list(d.keys())
    for i in range(0, 21):
        for j in file:
            for k in j:
                good1 = []
                good1a = []
                good2 = []
                good2a = []
                bad1 = []
                bad2 = []
                if k == d[i]:
                    good1a.append(key_list[i])
                    good1.append(good1a)
                    if isinstance(int(k), int):
                        if int(k) <= 20:
                            good2a.append(int(k))
                            good2.append(good2a)
                        else:
                            bad1.append(j)
                    else:
                        bad2.append(j)
    list1 = []
    list2 = []
    list1.append(good1)
    list1.append(good2)
    list2.append(bad1)
    list2.append(bad2)
    return list1, list2

However, this is not outputting what I would expect:

([[], []], [[], []])

This is probably something to do with the placing of my lists or something along those lines but I'm really new to python and I'm not following the logic of where things should go yet.

Any help would be appreciated!

Many thanks,

Carolina

Data:

d = {0: "zero, null",
    1: "one, un, eins",
    2: "two, deux, zwei",
    3: "three, trois, drei",
    4: "four, quatre, vier",
    5: "five, cinq, funf",
    6: "six, sechs",
    7: "seven, sept, sieben",
    8: "eight, huit, acht",
    9: "nine, neuf, neun",
    10: "ten, dix, zehn",
    11: "eleven, onze, elf",
    12: "twelve, douze, zwolf",
    13: "thirteen, treize, dreizehn",
    14: "fourteen, quatorze, vierzehn",
    15: "fifteen, quinze, funfzehn",
    16: "sixteen, seize, sechzehn",
    17: "seventeen, dix-sept, siebzehn",
    18: "eighteen, dix-huit, achtzehn",
    19: "nineteen, dix-neuf, neunzehn",
    20: "twenty, vingt, zwanzig"}

file = [['16', '10', '8', '3', '7'], ['8', '9', '19', '20', '4'], ['sechs', 'acht', 'sechzehn', 'funf', 'null'], ['1', '30', '2', '5', '7'], ['vierzehn', 'eins', 'zwei', 'neun', 'drei'], ['six', 'neuf', 'seize', 'zero'], ['fourteen', 'eleven', 'forteen', 'eight', 'twenty'], ['douze', 'onze', 'huit', 'quinze', 'sept'], ['18', '9', '9', '22', '4'], ['un', 'trois', 'quatorze', 'dix-huit', 'vingt'], ['five', 'three', 'nineteen', 'twenty', 'zero'], ['einundzwanzig', 'vierzehn', 'eins', 'zwei', 'vier']]

Answer 1

Try this:

import re

def conversion(d, file):
    list1 = []
    list2 = []
    for key, value in d.items():
        for i in file:
            list3 = []
            for j in i:
                good = []
                if re.search(fr"\b{re.escape(j)}\b", value):
                    good.append(key)
                    list3.append(good)
                else:
                    try:
                        good.append(int(j))
                        list3.append(good)
                    except:
                        bad = []
                        bad.append(i)
    list1.append(list3)
    list2.append(bad)
    return list1.Append(list2)

Answer 2

Using a for-loop try to match, through the regex word boundary, each number in the sublist with the values (numbers written in full) in the dictionary. If match, add the key to that corresponding number to the matched list and break the loop; If the number is not present in any of the spelled out numbers in the dictionary, then, use try/expect to check if number can be converted to an integer. If the conversion is possible add the integer number to sublist_match, otherwise add to the sublist_no_match.

def conversion(d, file):
    matched = []
    no_match = []
    for sublist in file:
        sublist_match = []
        sublist_no_match = []
        for number in sublist:
            for key, str_values in d.items():
                if re.search(fr"\b{re.escape(number)}\b", str_values):
                    sublist_match.append(key)
                    break
            else:
                try:
                    num_item = int(number)
                except ValueError:
                    sublist_no_match.append(number)
                else:
                    sublist_match.append(num_item)

        matched.append(sublist_match)
        no_match.append(sublist_no_match)

    return [matched, no_match]

result = conversion(d,file)
print(result)

Output from result

[[[16, 10, 8, 3, 7],
  [8, 9, 19, 20, 4],
  [6, 8, 16, 5, 0],
  [1, 30, 2, 5, 7],
  [14, 1, 2, 9, 3],
  [6, 9, 16, 0],
  [14, 11, 8, 20],
  [12, 11, 8, 15, 7],
  [18, 9, 9, 22, 4],
  [1, 3, 14, 18, 20],
  [5, 3, 19, 20, 0],
  [14, 1, 2, 4]],
 [[], [], [], [], [], [], ['forteen'], [], [], [], [], ['einundzwanzig']]]