TaskC05

TaskC05/run.py

@@ -17,9 +17,9 @@ def print_debug(*args, **kwargs):
 
 if len(sys.argv) == 1:
     print("Default arguments parsed\n")
-    sys.argv.append("elem.arg")
-    sys.argv.append("elem.in")
-    sys.argv.append("elem.exp")
+    sys.argv.append("medium.arg")
+    sys.argv.append("medium.in")
+    sys.argv.append("medium.exp")
 
 with open(sys.argv[1], mode="r", newline="", encoding="utf8") as csvfile:
     filereader = csv.reader(csvfile, delimiter="\t", quotechar="|")
@@ -30,19 +30,16 @@ with open(sys.argv[1], mode="r", newline="", encoding="utf8") as csvfile:
         fsa_description = fsa_description[1:]
 
     # get accepting states
-    accepting_state = []
-    for item in reversed(fsa_description):
-        if len(item) == 1:
-            accepting_state.append(item[0])
-        else:
-            break
-
-    fsa_description = fsa_description[: -len(accepting_state)]
+    accepting_states = []
 
     fsa_description_map: dict[tuple[str, str], list[str]] = {}
     fsa_description_direction: dict[str, list[tuple[str, str]]] = {}
     is_error = False
     for num, item in enumerate(fsa_description):
+        if len(item) == 1:
+            accepting_states.append(item[0])
+            continue
+
         for letter in item[2]:
             if letter not in alfabet:
                 print_debug(
@@ -57,7 +54,9 @@ with open(sys.argv[1], mode="r", newline="", encoding="utf8") as csvfile:
             fsa_description_map[tupleItems] = fsa_description_map.get(
                 tupleItems, []
             ) + [item[1]]
-            fsa_description_direction[item[0]].append((item[1], letter))
+            fsa_description_direction[item[0]] = fsa_description_direction.get(
+                item[0], []
+            ) + [(item[1], letter)]
 
     if is_error:
         exit(-1)
@@ -101,60 +100,72 @@ with open(sys.argv[3], mode="r", newline="", encoding="utf8") as file:
 
 
 def is_correct(current_state: str) -> bool:
-    if current_state in accepting_state:
+    if current_state in accepting_states:
         return True
     return False
 
 
-def find_description(current_state: str) -> tuple[str, str]:
+def traverse_to_description(current_state: str) -> list[str]:
     if current_state not in fsa_description_direction:
-        return current_state, "-1"  # unspecified next state
+        return [""]  # the end
 
+    good_end_states: list[str] = []
     for next_state, description in fsa_description_direction[current_state]:
-        if is_correct(next_state):
-            return next_state, description
+        new_good_end_states = traverse_to_description(next_state)
+        for new_good_end_state in new_good_end_states:
+            if is_correct(next_state) or new_good_end_state != "":
+                good_end_states.append(description + new_good_end_state)
 
-        end_state, description = find_description(next_state)
+    return good_end_states
 
 
-def traverse_states(current_state: str, word: str) -> str:
+def find_descriptions(current_state: str, word: str) -> list[str]:
+    state_after_word = traverse_word(current_state, word)
+    if state_after_word == "-1":
+        return [word + ";OOV"]
+
+    words_with_description: list[str] = []
+    for next_state in fsa_description_map[(state_after_word, SEMICOLON)]:
+        good_endings = traverse_to_description(next_state)
+        for good_ending in good_endings:
+            words_with_description.append(word + ";" + good_ending)
+
+    sorted_descriptions = sorted(words_with_description, key=lambda x: x.split(";")[1])
+    return sorted_descriptions
+
+
+def traverse_word(current_state: str, word: str) -> str:
+    if len(word) == 0:
+        return current_state
+
     ending_state = ""
-    for letter in word:
-        if (current_state, letter) in fsa_description_map:
-            next_states = fsa_description_map[(current_state, letter)]
-            for next_state in next_states:
-                ending_state = traverse_states(next_state, word[1:])
-                if is_correct(ending_state):
-                    return ending_state  # found a correct state
-            return "-2"  # no correct state found
-        else:
-            return "-1"  # unspecified next state
+    letter = word[0]
+    if (current_state, letter) in fsa_description_map:
+        next_states = fsa_description_map[(current_state, letter)]
+        for next_state in next_states:
+            ending_state = traverse_word(next_state, word[1:])
+            return ending_state
 
-    if (ending_state, SEMICOLON) not in fsa_description_map:
-        return current_state  # end of the word
-
-    for next_state in fsa_description_map[(ending_state, SEMICOLON)]:
-        ending_state, description = find_description(next_state)
+    return "-1"  # unspecified next state
 
 
 is_difference = []
-for i, word in enumerate(test_in):
+test_out_numerator = 0
+for word in test_in:
     current_state = "0"
+    descriptions = []
     if len(word) != 0:
-        current_state = traverse_states(current_state, word)
+        descriptions = find_descriptions(current_state, word)
 
-    print(str(i + 1) + "\t", end="")
-    if current_state in accepting_state:
-        print("YES\t", end="")
-        if "YES" != test_out[i]:
-            print("ERROR", word, end="")
-            is_difference.append(i + 1)
-    else:
-        print("NO\t", end="")
-        if "NO" != test_out[i]:
-            print("ERROR", word, end="")
-            is_difference.append(i + 1)
-    print()
+    for description in descriptions:
+        print(str(test_out_numerator + 1) + "\t", end="")
+        if description != test_out[test_out_numerator]:
+            print(f"ERROR\t{word} | {description}")
+            is_difference.append(test_out_numerator + 1)
+        else:
+            print(f"OK\t{description}")
+
+        test_out_numerator += 1
 
 if len(is_difference) != 0:
     print(is_difference)