feat: finalize GUI simulation and branch-aware analysis upgrades

Author: SakuraMathcraft

mock_survey/_check_scale_counts.py

@@ -0,0 +1,32 @@
+import re
+from pathlib import Path
+
+import pandas as pd
+
+html = Path(r"E:\QueScript\mock_survey\index.html").read_text(encoding="utf-8", errors="replace")
+scale_qids = re.findall(r'id="q(\d+)"[^>]*data-type="scale_radio"', html)
+matrix_qids = re.findall(r'id="q(\d+)"[^>]*data-type="matrix"', html)
+print("scale_radio_count", len(scale_qids), scale_qids)
+print("matrix_count", len(matrix_qids), matrix_qids)
+
+df = pd.read_csv(r"E:\QueScript\mock_survey\survey_data_collected.csv")
+qcols = [c for c in df.columns if c.startswith("Q")]
+
+expanded = []
+for c in qcols:
+    s = df[c].astype(str)
+    if s.str.contains("|", regex=False).mean() >= 0.2:
+        t = s.str.split("|", expand=True, regex=False)
+        t.columns = [f"{c}_r{i+1}" for i in range(t.shape[1])]
+        expanded.append(t)
+    else:
+        expanded.append(pd.DataFrame({c: df[c]}))
+
+num = pd.concat(expanded, axis=1)
+num = num.replace({"": None, "None": None, "nan": None}).apply(pd.to_numeric, errors="coerce")
+ratio = num.notna().mean()
+common = ratio[ratio >= 0.999].index.tolist()
+print("common_count", len(common))
+print("common_items", common)
+print("top_answer_ratio", ratio.sort_values(ascending=False).head(30).to_dict())
+

mock_survey/_mapping_check_runner.py

@@ -0,0 +1,26 @@
+import json
+import re
+from pathlib import Path
+
+import survey_generator as sg
+
+base = Path(__file__).resolve().parent
+txt = base / "问卷.txt"
+html_path = base / "index.html"
+
+data = sg.parse_survey(str(txt))
+sg.generate_html(data, str(html_path), survey_title=sg.extract_survey_title(str(txt)))
+html = html_path.read_text(encoding="utf-8", errors="replace") if html_path.exists() else ""
+titles = re.findall(r'<div class="question-title">\s*\d+\.\s*(.*?)</div>', html)
+
+out = {
+    "parsed_count": len(data),
+    "html_count": len(titles),
+    "parsed_empty_text": [i + 1 for i, q in enumerate(data) if not str(q.get("text", "")).strip()],
+    "html_empty_titles": [i + 1 for i, t in enumerate(titles) if not str(t).strip()],
+    "parsed_preview": [{"id": q.get("id"), "type": q.get("type"), "text": q.get("text")} for q in data[:20]],
+    "html_preview": titles[:20],
+}
+
+(base / "_mapping_check.json").write_text(json.dumps(out, ensure_ascii=False, indent=2), encoding="utf-8")
+print("done")

mock_survey/analysis_meta.json

@@ -0,0 +1,160 @@
+{
+  "run_id": "RUN_20260306_133858",
+  "seed": 1375906085,
+  "analysis_settings": {
+    "scope": "coverage",
+    "coverage_threshold": 0.6,
+    "branch_min_sample": 10,
+    "branch_min_items": 3
+  },
+  "sample_counts": {
+    "planned": 100,
+    "completed": 100,
+    "included": 100
+  },
+  "item_counts": {
+    "numericizable": 41,
+    "true_scale": 18,
+    "strict_public": 4,
+    "selected": 4,
+    "detected": 41
+  },
+  "selected_items": [
+    "Q8",
+    "Q30",
+    "Q32",
+    "Q33"
+  ],
+  "strict_public_items": [
+    "Q8",
+    "Q30",
+    "Q32",
+    "Q33"
+  ],
+  "coverage_ratio": {
+    "Q1": 0.0,
+    "Q2": 0.0,
+    "Q3": 0.0,
+    "Q4": 0.0,
+    "Q5": 0.0,
+    "Q6": 0.0,
+    "Q7": 0.0,
+    "Q8": 1.0,
+    "Q9": 0.0,
+    "Q10": 0.0,
+    "Q11": 0.0,
+    "Q12": 0.0,
+    "Q13": 0.0,
+    "Q14": 0.0,
+    "Q15": 0.0,
+    "Q16": 0.56,
+    "Q17": 0.56,
+    "Q18": 0.56,
+    "Q19": 0.0,
+    "Q20": 0.0,
+    "Q21": 0.0,
+    "Q22": 0.0,
+    "Q23": 0.56,
+    "Q24_r1": 0.56,
+    "Q24_r2": 0.56,
+    "Q24_r3": 0.56,
+    "Q24_r4": 0.56,
+    "Q24_r5": 0.56,
+    "Q24_r6": 0.56,
+    "Q24_r7": 0.56,
+    "Q24_r8": 0.56,
+    "Q25": 0.56,
+    "Q26": 0.56,
+    "Q27": 0.0,
+    "Q28": 0.0,
+    "Q29": 0.0,
+    "Q30": 1.0,
+    "Q31": 0.0,
+    "Q32": 1.0,
+    "Q33": 1.0,
+    "Q34": 0.0
+  },
+  "branch_sections": [
+    {
+      "branch_index": 1,
+      "trace": "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",
+      "sample_count": 56,
+      "item_count": 18,
+      "strict_item_count": 18,
+      "selected_item_count": 18,
+      "selected_items": [
+        "Q8",
+        "Q16",
+        "Q17",
+        "Q18",
+        "Q23",
+        "Q24_r1",
+        "Q24_r2",
+        "Q24_r3",
+        "Q24_r4",
+        "Q24_r5",
+        "Q24_r6",
+        "Q24_r7",
+        "Q24_r8",
+        "Q25",
+        "Q26",
+        "Q30",
+        "Q32",
+        "Q33"
+      ],
+      "alpha": 0.9776974225885819,
+      "kmo": 3.882258564807101e-15,
+      "cfa_available": true,
+      "cfa_cfi": 0.9417804353593766,
+      "cfa_rmsea": 0.09349905848262634,
+      "n_per_item": 3.111111111111111,
+      "exploratory_only": true,
+      "excluded_from_overall": true
+    },
+    {
+      "branch_index": 2,
+      "trace": "1>2>3>4>5>6>7>8>9>29>30>31>32>33>34",
+      "sample_count": 44,
+      "item_count": 4,
+      "strict_item_count": 4,
+      "selected_item_count": 4,
+      "selected_items": [
+        "Q8",
+        "Q30",
+        "Q32",
+        "Q33"
+      ],
+      "alpha": 0.9568490060248958,
+      "kmo": 0.8353419890103603,
+      "cfa_available": true,
+      "cfa_cfi": 0.9934091102222444,
+      "cfa_rmsea": 0.12153564274746322,
+      "n_per_item": 11.0,
+      "exploratory_only": false,
+      "excluded_from_overall": false
+    }
+  ],
+  "consistency": {
+    "grade": "A",
+    "reliability": "达标",
+    "validity": "达标",
+    "structural_grade": "B"
+  },
+  "target_consistency": {
+    "grade": "A",
+    "reliability": "达标",
+    "validity": "达标",
+    "structural_grade": "B"
+  },
+  "structural_risk": {
+    "grade": "B",
+    "score": 5,
+    "reasons": [
+      "当前全样本仅纳入4题，难代表整份量表结构。",
+      "全样本仅覆盖真量表题的22%，分支互斥较强。",
+      "有1个分支因 n/p<5 仅作探索性参考，不纳入总体结论。"
+    ]
+  },
+  "generated_at": "2026-03-06T13:42:54",
+  "analysis_signature": "6c4f154467d581ffff62bf9daf7b8932733e8837ad9b3868dcb89ac81001ba52"
+}