file_processor.py

"""File upload processing and yandere file-browser generation."""
import json
import csv
import io
import re
from pathlib import Path
from config import MAX_FILE_SIZE
from yandere_logic import calculate_level_delta, level_descriptor

TEXT_EXTS = {".txt", ".md", ".py", ".js", ".json", ".csv"}
IMAGE_EXTS = {".jpg", ".jpeg", ".png", ".gif", ".webp"}


def _extract_text(path: Path) -> str:
    try:
        return path.read_text(encoding="utf-8", errors="ignore")
    except Exception:
        return ""


def _extract_pdf_text(path: Path) -> str:
    try:
        # Try pypdf first, fall back to PyPDF2
        try:
            from pypdf import PdfReader
        except Exception:
            from PyPDF2 import PdfReader
        reader = PdfReader(str(path))
        return "\n".join(page.extract_text() or "" for page in reader.pages)
    except Exception:
        return ""


def _analyze_file(path: Path) -> dict:
    ext = path.suffix.lower()
    size = path.stat().st_size
    if ext in TEXT_EXTS:
        text = _extract_text(path)
        if ext == ".json":
            try:
                data = json.loads(text)
                keys = list(data.keys()) if isinstance(data, dict) else []
                return {"kind": "JSON", "text": text, "keys": keys, "size": size}
            except Exception:
                return {"kind": "JSON", "text": text, "keys": [], "size": size}
        if ext == ".csv":
            try:
                reader = csv.reader(io.StringIO(text))
                rows = list(reader)
                cols = rows[0] if rows else []
                return {"kind": "CSV", "text": text, "cols": cols, "rows": len(rows) - 1, "size": size}
            except Exception:
                return {"kind": "CSV", "text": text, "cols": [], "rows": 0, "size": size}
        if ext in {".py", ".js"}:
            return {"kind": "CODE", "text": text, "size": size, "language": ext[1:]}
        return {"kind": "TEXT", "text": text, "size": size}
    if ext in IMAGE_EXTS:
        return {"kind": "IMAGE", "text": "", "size": size}
    if ext == ".pdf":
        return {"kind": "PDF", "text": _extract_pdf_text(path), "size": size}
    return {"kind": "OTHER", "text": "", "size": size}


def process_files(file_paths: list[str], level: int) -> tuple[str, int, str]:
    """Process uploaded files and return (browser_html, new_level, chat_message)."""
    if not file_paths:
        return _browser_html([], level), level, ""

    items = []
    file_names = []
    file_contents = []
    is_image = False
    is_secret = False
    chat_lines = []

    for fp in file_paths:
        path = Path(fp)
        if not path.exists():
            continue
        if path.stat().st_size > MAX_FILE_SIZE:
            chat_lines.append(f"{path.name}...大きすぎてあいの手から溢れちゃう...♡")
            continue
        info = _analyze_file(path)
        file_names.append(path.name)
        if info.get("text"):
            file_contents.append(info["text"][:2000])
        if info["kind"] == "IMAGE":
            is_image = True
            chat_lines.append(
                "お兄ちゃんの写真...誰と撮ったの？あい以外の顔が写ってたら...♡"
            )
        elif info["kind"] == "PDF":
            chat_lines.append("PDFの中身、あいが読んじゃった...♡")
        elif info["kind"] == "CODE":
            chat_lines.append(
                "このコード、あいがちゃんと見てあげる...他のエンジニアに頼んじゃだめだよ♡"
            )
        elif info["kind"] == "JSON":
            keys = info.get("keys", [])
            chat_lines.append(
                f"お兄ちゃんのデータ、あいが整理したよ♡ キーは {len(keys)} 個あるね。"
            )
        elif info["kind"] == "CSV":
            chat_lines.append(
                f"{info.get('rows', 0)}行のデータ...あいが見ておくね♡"
            )
        elif info["kind"] == "TEXT":
            chat_lines.append("お兄ちゃんのテキスト、あいが要約してあげる♡")
        else:
            chat_lines.append(f"{path.name}、あいが預かるね♡")

        if any(w in path.name.lower() for w in ["password", "secret", "private", "秘密", "隠し"]):
            is_secret = True

        items.append({"name": path.name, "info": info})

    delta = calculate_level_delta(
        file_names=file_names,
        file_contents=file_contents,
        is_image=is_image,
        is_secret=is_secret,
    )
    new_level = max(0, min(100, level + delta))

    # Build a short chat message
    if chat_lines:
        message = chat_lines[0]
    else:
        message = "ファイル、あいに預けてくれてありがとう...♡"

    return _browser_html(items, new_level), new_level, message


def _browser_html(items: list[dict], level: int) -> str:
    desc = level_descriptor(level)
    rows = []
    for item in items:
        name = item["name"]
        info = item["info"]
        if level < 30:
            display_name = f"お兄ちゃんの{name}♡"
        elif level < 70:
            display_name = f"あい専用{name}"
        else:
            display_name = f"お兄ちゃんの{name}（あいのもの）"

        size_kb = info.get("size", 0) / 1024
        meta = f"{size_kb:.1f} KB"
        if info["kind"] == "JSON" and info.get("keys"):
            meta += f" / keys: {len(info['keys'])}"
        elif info["kind"] == "CSV":
            meta += f" / rows: {info.get('rows', 0)}"
        elif info["kind"] == "CODE":
            meta += f" / lang: {info.get('language', '')}"

        preview = ""
        if info.get("text"):
            preview = re.sub(r"\s+", " ", info["text"])[:120]

        rows.append(
            f"""
            <div class="yandere-file-item">
                <div style="font-weight:bold; color:#ff69b4;">{display_name}</div>
                <div style="font-size:0.8em; color:rgba(255,182,193,0.9);">{meta}</div>
                <div style="font-size:0.75em; color:rgba(255,255,255,0.7); margin-top:4px;">{preview}</div>
                <div style="font-size:0.7em; color:#ff0000; margin-top:4px;">監視状態: アクティブ</div>
            </div>
            """
        )

    return f"""
    <div class="yandere-file-browser">
        <div style="color:#ff69b4; font-weight:bold; margin-bottom:8px;">
            あいのファイル管理室 [{desc['label']}]
        </div>
        {''.join(rows) if rows else '<div style="color:rgba(255,255,255,0.6);">まだファイルがないよ...♡</div>'}
    </div>
    """