More medium & hard solutions

Author: StBogdan

chapter_16/p17_20.py

@@ -0,0 +1,13 @@
+def zeroes_in_fact(n: int):
+    five_pow = 1
+    nr_z = 0
+    while (5**five_pow) <= n:
+        nr_z += (n // 5 ** five_pow)
+        five_pow += 1
+    return nr_z
+
+
+if __name__ == "__main__":
+    exs = [1, 10, 34, 55, 642, 1000]
+    for ex in exs:
+        print(f"Number of zeroes in {ex}! is {zeroes_in_fact(ex)}")

chapter_17/p17_26.py

@@ -0,0 +1,41 @@
+from typing import List, Tuple
+from collections import defaultdict
+
+
+def sparse_similarity(docs_raw: List[Tuple[int, List[int]]]) -> None:
+    doc_dict = dict(docs_raw)
+    rev_index = defaultdict(set)
+
+    for doc_id, doc_words in doc_dict.items():
+        for word in doc_words:
+            rev_index[word].add(doc_id)
+
+    intersect_of_pairs = defaultdict(int)
+
+    for doc_id, doc_words in doc_dict.items():
+        for word in doc_words:
+            for other_doc in rev_index[word]:
+                # We assume an ordering of the doc ids,
+                # as to prevent duplicate counting
+                if other_doc <= doc_id:
+                    continue
+
+                intersect_of_pairs[(doc_id, other_doc)] += 1
+
+    print(f"ID1, ID2:\tSIMILARITY")
+    for (doc1, doc2), in_common in intersect_of_pairs.items():
+        union_size = len(doc_dict[doc1]) + len(doc_dict[doc2]) - in_common
+        docs_simil = in_common / union_size
+
+        print(f"{doc1:3}, {doc2:3}:\t{docs_simil}")
+
+
+if __name__ == "__main__":
+    ex = [
+        (13, [14, 15, 100, 9, 3]),
+        (16, [32, 1, 9, 3, 5]),
+        (19, [15, 29, 2, 6, 8, 7]),
+        (24, [7, 10])
+    ]
+
+    sparse_similarity(ex)

chapter_17/p17_5.py

@@ -0,0 +1,39 @@
+
+def longest_subarr_number_letter_equal(arr: str) -> int:
+    more_letters_now = 0
+    seen_offsets = {0: -1}
+    max_arr_size = 0
+    max_arr_indexes = (-1, -1)
+
+    for i in range(len(arr)):
+        x = arr[i]
+
+        if x.isdigit():
+            more_letters_now -= 1
+        else:
+            more_letters_now += 1
+
+        # print(f"Offsets: {seen_offsets}, letters more now: {more_letters_now}, elem now {x}")
+
+        if more_letters_now in seen_offsets:
+            max_arr_now = i - seen_offsets[more_letters_now]
+            if max_arr_now > max_arr_size:
+                max_arr_size = max_arr_now
+                max_arr_indexes = (seen_offsets[more_letters_now]+1, i)
+        else:
+            seen_offsets[more_letters_now] = i
+
+    return max_arr_size, max_arr_indexes
+
+
+if __name__ == "__main__":
+    exs = [
+        "aaa2222aa22aa222aaaa",
+        "aaa222aa",
+        "e4ee2a2a2a",
+        "z0123456789a"
+    ]
+    for ex_arr in exs:
+        max_arr, (start, end) = longest_subarr_number_letter_equal(ex_arr)
+        print(f"Max arr with eq is of len {max_arr}, "
+              f"namely: {ex_arr[start:end+1]}, from {ex_arr}")