import sys, os, re
from pathlib import Path
from dotenv import load_dotenv

load_dotenv(Path("D:/code/crm-需求梳理/lanhu-mcp/.env"), override=True)
sys.path.insert(0, "D:/code/crm-需求梳理/lanhu-mcp")
os.chdir("D:/code/crm-需求梳理/lanhu-mcp")
from lanhu_mcp_server import DATA_DIR

resource_dir = Path(DATA_DIR) / "axure_extract_bade4454"
files_dir = resource_dir / "files"

BACKSLASH = chr(92)

def read_data_js(subdir_name):
    data_js = files_dir / subdir_name / "data.js"
    if not data_js.exists():
        return f"[data.js not found: {data_js}]"
    with open(data_js, encoding='utf-8', errors='ignore') as f:
        js = f.read()
    texts = []
    for m in re.finditer(r'"([^"\\]{2,})"', js):
        t = m.group(1).strip()
        if (t and len(t) >= 2
                and not re.match(r'^[0-9a-f\-]{8,}$', t)
                and not t.startswith('http')
                and not t.startswith('#')
                and not t.startswith('rgb')
                and BACKSLASH not in t
                and not re.match(r'^[A-Za-z][A-Za-z0-9_.]*$', t)
           ):
            texts.append(t)
    seen = set()
    uniq = []
    for t in texts:
        if t not in seen:
            seen.add(t)
            uniq.append(t)
    return '\n'.join(uniq[:300])

targets = {
    "A7-3-1 线索规则":       "a7-3-1_线索规则",
    "A2-1-1 线索看板":        "a2-1-1线索管理（看板视图）",
    "A2-1-1 线索列表":        "a2-1-1线索管理（列表视图）",
    "A2-1-1-1 新建线索":      "a2-1-1-1新建线索",
    "A2-1-2 我的线索":        "a2-1-2我的线索",
    "A2-1-2-1 线索详情":      "a2-1-2-1线索详情",
    "A2-1-2-2 领取线索":      "a2-1-2-2领取线索",
    "A2-1-2-3 编辑线索":      "a2-1-2-3编辑线索",
    "A2-1-4 公共池":          "a2-1-4公共池",
    "A2-1-4-1 公共池领取线索": "a2-1-4-1领取线索",
    "A2-1-5 全部线索":        "a2-1-5全部线索",
}

all_dirs = {d.name: d for d in files_dir.iterdir() if d.is_dir()}

out_path = Path("D:/code/crm-backend-matt/tmp/prototype_text.txt")
with open(out_path, 'w', encoding='utf-8') as out:
    for label, key in targets.items():
        matched_dir = all_dirs.get(key)
        if not matched_dir:
            key_nsp = key.replace(' ', '')
            candidates = [name for name in all_dirs if key_nsp in name.replace(' ', '')]
            if candidates:
                matched_dir = all_dirs[candidates[0]]

        out.write(f"\n{'='*60}\n## {label}  [{matched_dir.name if matched_dir else 'NOT FOUND'}]\n{'='*60}\n")
        if matched_dir:
            text = read_data_js(matched_dir.name)
            out.write(text[:5000])
            print(f"  OK   {label}: {len(text)} chars")
        else:
            out.write("[NOT FOUND]\n")
            print(f"  MISS {label}")

print(f"\nWritten: {out_path}")