"""Read only decrypted cache schemas and aggregate voice counts; never content."""
import json
import re
import sqlite3
from pathlib import Path


OUT = Path(__file__).resolve().parent
BASES = {"source": Path(r"C:\wechat_rpa"),
         "installed": Path(r"C:\Users\isard\AppData\Local\ZhenYangTangRPA")}
RELATIVES = ("wxwork_reply_cache", "archive_auto_backup/decrypted", "wxwork_decrypted")
paths = []
for base_name, base in BASES.items():
    for relative in RELATIVES:
        root = base / relative
        if not root.is_dir():
            continue
        for folder in sorted(root.iterdir()):
            if not folder.is_dir() or not re.fullmatch(r"\d{1,24}", folder.name):
                continue
            path = folder / "message.db"
            if path.is_file() and path.resolve().is_relative_to(root.resolve()):
                paths.append((base_name, relative, folder.name, path))
labels = {account: "account-" + str(i + 1) for i, account in enumerate(sorted({r[2] for r in paths}))}
result = {"scope": "existing decrypted caches only; no active identity discovery",
          "sqlite": "mode=ro&immutable=1; main-file snapshot only; WAL ignored",
          "databases": []}
for base, relative, account, path in paths:
    item = {"location": base, "cache": relative, "account": labels[account], "tables": {}}
    result["databases"].append(item)
    connection = None
    try:
        connection = sqlite3.connect(path.resolve().as_uri() + "?mode=ro&immutable=1", uri=True, timeout=3)
        connection.execute("PRAGMA query_only=ON")
        connection.execute("BEGIN")
        names = {r[0] for r in connection.execute("SELECT name FROM sqlite_master WHERE type='table'")}
        for table in ("message_table", "msg_voice2text"):
            if table not in names:
                continue
            columns = [{"name": row[1], "type": row[2], "pk": row[5]}
                       for row in connection.execute('PRAGMA table_info("' + table + '")')]
            item["tables"][table] = {"columns": columns,
                "count": connection.execute('SELECT count(*) FROM "' + table + '"').fetchone()[0]}
        if "message_table" not in item["tables"]:
            continue
        columns = {row["name"] for row in item["tables"]["message_table"]["columns"]}
        if "content_type" not in columns:
            continue
        item["voice_types"] = {str(row[0]): row[1] for row in connection.execute(
            "SELECT content_type,count(*) FROM message_table WHERE content_type IN (4,16) GROUP BY content_type")}
        if {"sender_id", "conversation_id"} <= columns:
            row = connection.execute("""SELECT count(*),
                sum(CAST(sender_id AS TEXT) != ?),
                sum(conversation_id LIKE 'M:%' OR conversation_id LIKE 'S:%')
                FROM message_table WHERE content_type IN (4,16)""", (account,)).fetchone()
            item["voice_counts"] = dict(zip(("total", "incoming", "personal"), (int(v or 0) for v in row)))
        if "msg_voice2text" in item["tables"]:
            trans_columns = {row["name"] for row in item["tables"]["msg_voice2text"]["columns"]}
            if {"message_id", "text"} <= trans_columns:
                item["nonempty_transcriptions"] = connection.execute(
                    "SELECT count(*) FROM msg_voice2text WHERE text IS NOT NULL AND trim(text) != ''").fetchone()[0]
                if "server_id" in columns:
                    item["voice_with_cached_transcription"] = connection.execute("""
                        SELECT count(*) FROM message_table m WHERE m.content_type IN (4,16)
                        AND EXISTS (SELECT 1 FROM msg_voice2text v
                            WHERE CAST(v.message_id AS TEXT)=CAST(m.server_id AS TEXT)
                            AND v.text IS NOT NULL AND trim(v.text) != '')""").fetchone()[0]
    except Exception as exc:
        item["error_type"] = type(exc).__name__
    finally:
        if connection is not None:
            connection.close()
OUT.mkdir(parents=True, exist_ok=True)
(OUT / "voice-schema-counts.json").write_text(json.dumps(result, ensure_ascii=False, indent=2), encoding="utf-8")
print(json.dumps({"databases": [{key: value for key, value in item.items() if key != "tables"} |
    {"tables": {key: value["count"] for key, value in item["tables"].items()}}
    for item in result["databases"]]}, ensure_ascii=False))