File size: 8,305 Bytes
87b5875
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
from __future__ import annotations

import json
import re
from dataclasses import dataclass
from typing import Any

from cert_study_app.services.text_cleanup_service import clean_inline_text


YES_WORDS = {"yes", "y", "예", "true", "selected"}
NO_WORDS = {"no", "n", "μ•„λ‹ˆμ˜€", "μ•„λ‹ˆμš”", "false", "not selected"}


@dataclass(frozen=True)
class AnswerEvaluation:
    chosen: str
    answer: str
    normalized_chosen: str
    normalized_answer: str
    correct: bool
    ordered: bool = False


def normalize_options(raw: Any) -> list[str]:
    if not raw:
        return []
    try:
        if isinstance(raw, str):
            parsed = json.loads(raw)
            if isinstance(parsed, (list, dict)):
                raw = parsed
        if isinstance(raw, dict):
            keyed = []
            for key, value in sorted(raw.items(), key=lambda kv: str(kv[0])):
                text = clean_inline_text(value)
                if re.match(r"^[A-Za-z1-9][\.\)]\s+", text):
                    keyed.append(text)
                else:
                    keyed.append(f"{key}. {text}")
            return split_embedded_options(keyed)
        if isinstance(raw, list):
            return split_embedded_options(raw)
    except Exception:
        return []
    return []


def split_embedded_options(options: list[str]) -> list[str]:
    split = []
    expected = "A"
    for option in options:
        text = clean_inline_text(option)
        if not text:
            continue
        parts = list(re.finditer(r"(?<![A-Za-z0-9])([A-Z])[\.\)]\s+", text))
        if len(parts) <= 1:
            split.append(text)
            continue
        for index, match in enumerate(parts):
            end = parts[index + 1].start() if index + 1 < len(parts) else len(text)
            part = text[match.start() : end].strip()
            label = match.group(1).upper()
            if label >= expected:
                split.append(part)
                expected = chr(ord(label) + 1)
    return split


def extract_options_from_stem(stem: str) -> list[str]:
    lines = [line.strip() for line in (stem or "").splitlines() if line.strip()]
    options = []
    current_key = None
    current_parts = []
    expected = "A"

    def flush():
        if current_key and current_parts:
            text = " ".join(current_parts).strip()
            if text:
                options.append(f"{current_key}. {text}")

    for line in lines:
        if re.match(r"^(answer|μ •λ‹΅|explanation|reference)\b", line, re.I):
            flush()
            current_key = None
            current_parts = []
            break

        marker = None
        body = ""
        exact = re.match(r"^([A-Z])[\.\)]?$", line, re.I)
        inline = re.match(r"^([A-Z])[\.\)]?\s+(.+)$", line, re.I)
        if exact:
            marker = exact.group(1).upper()
        elif inline:
            marker = inline.group(1).upper()
            body = inline.group(2).strip()

        if marker and marker >= expected:
            flush()
            current_key = marker
            current_parts = [body] if body else []
            expected = chr(ord(marker) + 1)
        elif current_key:
            current_parts.append(line)

    flush()
    return options if len(options) >= 2 else []


def extract_answer_from_stem(stem: str) -> str:
    match = re.search(r"(?:Answer|μ •λ‹΅)\s*:?\s*([A-Z1-9])", stem or "", re.I)
    if not match:
        return ""
    value = match.group(1).upper()
    if value.isdigit():
        return chr(ord("A") + int(value) - 1)
    return value


def option_label(value: Any) -> str:
    text = str(value or "").strip().upper()
    if text.isdigit():
        return chr(ord("A") + int(text) - 1)
    match = re.match(r"^([A-Z])(?:[\s\.,\)]|$)", text)
    return match.group(1) if match else text


def choice_labels(value: Any) -> list[str]:
    text = str(value or "").strip().upper()
    if not text:
        return []
    if re.fullmatch(r"[A-Z]{2,26}", text):
        return list(text)
    tokens = re.findall(r"\b[A-Z]\b|\b[1-9]\b", text)
    return [option_label(token) for token in tokens]


def yes_no_labels(value: Any) -> list[str]:
    raw_value = _parse_jsonish(value)

    if isinstance(raw_value, list):
        values = [
            item.get("value") or item.get("answer") or item.get("selected_answer")
            for item in raw_value
            if isinstance(item, dict)
        ]
        labels = yes_no_labels(",".join(str(item) for item in values if item))
        if labels:
            return labels
    elif isinstance(raw_value, dict):
        labels = yes_no_labels(",".join(str(item) for item in raw_value.values()))
        if labels:
            return labels

    text = str(value or "").strip().lower()
    if not text:
        return []
    if re.fullmatch(r"[yn](?:\s*,\s*[yn])+", text, re.I):
        return [token.upper() for token in re.findall(r"[yn]", text, re.I)]

    tokens = re.findall(
        r"(?<![κ°€-힣])예(?![κ°€-힣])|μ•„λ‹ˆμ˜€|μ•„λ‹ˆμš”|\byes\b|\bno\b|\bselected\b|\bnot selected\b|\btrue\b|\bfalse\b",
        text,
        re.I,
    )
    labels = []
    for token in tokens:
        lowered = token.lower()
        labels.append("Y" if token == "예" or lowered in YES_WORDS else "N")
    return labels if len(labels) >= 2 else []


def structured_answer_values(value: Any) -> list[str]:
    raw_value = _parse_jsonish(value)
    if isinstance(raw_value, list):
        values = [
            item.get("value") or item.get("answer") or item.get("selected_answer")
            for item in raw_value
            if isinstance(item, dict)
        ]
    elif isinstance(raw_value, dict):
        values = list(raw_value.values())
    else:
        return []
    return [str(item).strip() for item in values if str(item).strip()]


def option_text(options: list[str], label: str) -> str:
    label = option_label(label)
    for index, option in enumerate(options, 1):
        text = str(option).strip()
        keys = {str(index), chr(ord("A") + index - 1)}
        match = re.match(r"^([A-Z1-9])[\.\)]\s+(.+)$", text, re.I)
        if match:
            keys.add(option_label(match.group(1)))
            if match.group(1).isdigit():
                keys.add(match.group(1))
        if label in keys:
            return text
    return ""


def option_labels_from_texts(options: list[str], values: Any, allow_duplicates: bool = False) -> list[str]:
    if not values:
        return []
    if isinstance(values, str):
        values = [values]
    labels = []
    for value in values:
        target = re.sub(r"\s+", " ", str(value or "").strip()).lower()
        if not target:
            continue
        for index, option in enumerate(options, 1):
            body = re.sub(r"\s+", " ", _option_body(option)).lower()
            if body and (target == body or target in body or body in target):
                label = chr(ord("A") + index - 1)
                if allow_duplicates or label not in labels:
                    labels.append(label)
                break
    return labels


def normalize_answer(value: Any, *, ordered: bool = False) -> str:
    yn = yes_no_labels(value)
    if yn:
        return ",".join(yn)
    labels = choice_labels(value)
    if not labels:
        return option_label(value)
    if ordered:
        return ",".join(labels)
    return ",".join(sorted(set(labels)))


def evaluate_answer(chosen: Any, answer: Any, *, ordered: bool = False) -> AnswerEvaluation:
    normalized_chosen = normalize_answer(chosen, ordered=ordered)
    normalized_answer = normalize_answer(answer, ordered=ordered)
    return AnswerEvaluation(
        chosen=str(chosen or ""),
        answer=str(answer or ""),
        normalized_chosen=normalized_chosen,
        normalized_answer=normalized_answer,
        correct=normalized_chosen == normalized_answer,
        ordered=ordered,
    )


def _option_body(option: str) -> str:
    text = str(option or "").strip()
    match = re.match(r"^([A-Z1-9])[\.\)]\s+(.+)$", text, re.I)
    return (match.group(2) if match and match.group(2) else text).strip()


def _parse_jsonish(value: Any) -> Any:
    if isinstance(value, (list, dict)):
        return value
    try:
        return json.loads(value) if isinstance(value, str) else None
    except Exception:
        return None