src/app_logger.py

"""
Centralized logger for Job Automation Agent.
Writes to data/logs/run_YYYY-MM-DD_HH-MM-SS.log
Also keeps last N lines in memory for the UI.
"""
import logging
import os
import sys
from datetime import datetime
from pathlib import Path
from collections import deque

LOG_DIR = Path("data/logs")
_memory_buffer: deque = deque(maxlen=500)   # last 500 lines in memory
_current_log_file: str = ""


def setup(run_id: str = "") -> str:
    """
    Set up file + console logging. Returns path to log file.
    Call once at the start of each pipeline run.
    """
    global _current_log_file, _memory_buffer
    _memory_buffer.clear()

    LOG_DIR.mkdir(parents=True, exist_ok=True)
    tag = run_id or datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
    log_path = str(LOG_DIR / f"run_{tag}.log")
    _current_log_file = log_path

    # Root logger
    root = logging.getLogger()
    root.setLevel(logging.DEBUG)

    # Remove old handlers
    for h in root.handlers[:]:
        root.removeHandler(h)

    fmt = logging.Formatter("%(asctime)s [%(levelname)s] %(name)s — %(message)s",
                            datefmt="%H:%M:%S")

    # File handler
    fh = logging.FileHandler(log_path, encoding="utf-8")
    fh.setLevel(logging.DEBUG)
    fh.setFormatter(fmt)
    root.addHandler(fh)

    # Memory handler (for UI display)
    mh = _MemoryHandler()
    mh.setLevel(logging.DEBUG)
    mh.setFormatter(fmt)
    root.addHandler(mh)

    # Force UTF-8 on stdout/stderr so unicode chars (✓ → etc.) don't crash on Windows cp1252
    try:
        sys.stdout.reconfigure(encoding="utf-8", errors="replace")
        sys.stderr.reconfigure(encoding="utf-8", errors="replace")
    except AttributeError:
        pass  # already replaced by Streamlit or not a TextIOWrapper

    # Redirect stdout/stderr so print() and Playwright output are captured in log file
    sys.stdout = _TeeStream(sys.stdout, log_path)
    sys.stderr = _TeeStream(sys.stderr, log_path)

    return log_path


def get_log_file() -> str:
    return _current_log_file


def get_memory_lines() -> list[str]:
    return list(_memory_buffer)


def read_log_file(tail: int = 200) -> str:
    """Read last N lines of current log file."""
    if not _current_log_file or not os.path.exists(_current_log_file):
        return ""
    try:
        with open(_current_log_file, "r", encoding="utf-8", errors="replace") as f:
            lines = f.readlines()
        return "".join(lines[-tail:])
    except Exception:
        return ""


def list_log_files() -> list[str]:
    """Return all log files, newest first."""
    if not LOG_DIR.exists():
        return []
    return sorted(
        [str(p) for p in LOG_DIR.glob("run_*.log")],
        reverse=True,
    )


class _MemoryHandler(logging.Handler):
    def emit(self, record):
        try:
            _memory_buffer.append(self.format(record))
        except Exception:
            pass


class _TeeStream:
    """Writes to both original stream and log file."""
    def __init__(self, original, log_path: str):
        self._orig = original
        self._log  = log_path

    def write(self, text):
        try:
            self._orig.write(text)
        except (UnicodeEncodeError, UnicodeDecodeError):
            # Windows cp1252 console can't handle unicode — write safe version
            try:
                self._orig.write(text.encode("utf-8", errors="replace").decode("ascii", errors="replace"))
            except Exception:
                pass
        except Exception:
            pass
        if text.strip():
            try:
                with open(self._log, "a", encoding="utf-8", errors="replace") as f:
                    f.write(text)
            except Exception:
                pass

    def flush(self):
        try:
            self._orig.flush()
        except Exception:
            pass

    def fileno(self):
        return self._orig.fileno()

    def isatty(self):
        return False