pyr0ball
50bae0d393
Xander's ghcr.io/hotio/qbittorrent:latest container uses a different format than the classic GUI build: `(N) 2026-04-26T03:32:59 - message` with a single-char level code before an ISO timestamp, not inside parens. Added _HOTIO_RE alongside _CLASSIC_RE; unified via _match_line() helper so parse() loop is unchanged. 28 tests passing, both formats covered.
123 lines
3.9 KiB
Python
123 lines
3.9 KiB
Python
"""qBittorrent log parser.
|
|
|
|
Handles two formats produced by different qBittorrent builds:
|
|
|
|
Classic (pre-5.x GUI builds):
|
|
(YYYY/MM/DD HH:MM:SS) [Level] Message text
|
|
(YYYY/MM/DD HH:MM:SS) Message text (no explicit level)
|
|
|
|
Hotio/headless (5.x container builds, e.g. ghcr.io/hotio/qbittorrent):
|
|
(N) 2026-04-26T03:32:59 - Message text
|
|
(W) 2026-04-26T03:33:00 - Warning message
|
|
|
|
Level codes for hotio format: N=Normal/Info, I=Info, W=Warning, C=Critical.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import re
|
|
from datetime import datetime, timezone
|
|
from typing import Iterator
|
|
|
|
from app.ingest.base import (
|
|
SourceState, apply_patterns, detect_severity, make_entry_id, now_iso,
|
|
)
|
|
from app.services.models import LogPattern, RetrievedEntry
|
|
|
|
# Classic: (2026/05/09 14:23:01) [Warning] Tracker '...' is not working.
|
|
_CLASSIC_RE = re.compile(
|
|
r"^\((?P<ts>\d{4}[/-]\d{2}[/-]\d{2}\s+\d{2}:\d{2}:\d{2})\)"
|
|
r"(?:\s+\[(?P<level>[^\]]+)\])?"
|
|
r"\s+(?P<msg>.*)$"
|
|
)
|
|
|
|
# Hotio/headless: (N) 2026-04-26T03:33:11 - Successfully listening on IP...
|
|
_HOTIO_RE = re.compile(
|
|
r"^\((?P<level>[NIWC])\)\s+"
|
|
r"(?P<ts>\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2})"
|
|
r"\s+-\s+(?P<msg>.*)$"
|
|
)
|
|
|
|
_LEVEL_MAP: dict[str, str] = {
|
|
# Classic bracket labels (lowercased for lookup)
|
|
"normal": "INFO",
|
|
"info": "INFO",
|
|
"warning": "WARN",
|
|
"critical": "CRITICAL",
|
|
# Hotio single-char codes (lowercased for lookup)
|
|
"n": "INFO",
|
|
"i": "INFO",
|
|
"w": "WARN",
|
|
"c": "CRITICAL",
|
|
}
|
|
|
|
|
|
def _parse_ts(ts_str: str) -> tuple[str, str]:
|
|
"""Return (raw, iso). Handles classic (space sep) and hotio (T sep) timestamps."""
|
|
for fmt in ("%Y-%m-%dT%H:%M:%S", "%Y-%m-%d %H:%M:%S", "%Y/%m/%d %H:%M:%S"):
|
|
try:
|
|
dt = datetime.strptime(ts_str, fmt).replace(tzinfo=timezone.utc)
|
|
return ts_str, dt.isoformat()
|
|
except ValueError:
|
|
continue
|
|
return ts_str, ""
|
|
|
|
|
|
def _match_line(line: str) -> dict | None:
|
|
"""Try both log formats. Returns {ts, level, msg} or None."""
|
|
m = _CLASSIC_RE.match(line)
|
|
if m:
|
|
return {"ts": m.group("ts"), "level": m.group("level") or "", "msg": m.group("msg")}
|
|
m = _HOTIO_RE.match(line)
|
|
if m:
|
|
return {"ts": m.group("ts"), "level": m.group("level"), "msg": m.group("msg")}
|
|
return None
|
|
|
|
|
|
def is_qbit_log(first_line: str) -> bool:
|
|
return _match_line(first_line.strip()) is not None
|
|
|
|
|
|
def parse(
|
|
lines: Iterator[str],
|
|
source_id: str,
|
|
compiled_patterns: list[tuple[LogPattern, object]],
|
|
ingest_time: str | None = None,
|
|
) -> Iterator[RetrievedEntry]:
|
|
ingest_time = ingest_time or now_iso()
|
|
state = SourceState()
|
|
pending_text: str | None = None
|
|
pending_meta: dict = {}
|
|
|
|
def _emit(text: str, meta: dict) -> RetrievedEntry:
|
|
repeat, out_of_order = state.observe(text, meta.get("ts_iso"))
|
|
matched = apply_patterns(text, compiled_patterns)
|
|
return RetrievedEntry(
|
|
entry_id=make_entry_id(source_id, state.sequence, text),
|
|
source_id=source_id,
|
|
sequence=state.sequence,
|
|
timestamp_raw=meta.get("ts_raw", ""),
|
|
timestamp_iso=meta.get("ts_iso", ""),
|
|
ingest_time=ingest_time,
|
|
severity=meta.get("severity"),
|
|
repeat_count=repeat,
|
|
out_of_order=out_of_order,
|
|
matched_patterns=matched,
|
|
text=text,
|
|
)
|
|
|
|
for raw_line in lines:
|
|
line = raw_line.rstrip("\n")
|
|
m = _match_line(line)
|
|
if m:
|
|
if pending_text is not None:
|
|
yield _emit(pending_text, pending_meta)
|
|
|
|
ts_raw, ts_iso = _parse_ts(m["ts"])
|
|
severity = _LEVEL_MAP.get(m["level"].lower()) or detect_severity(m["msg"])
|
|
pending_meta = {"ts_raw": ts_raw, "ts_iso": ts_iso, "severity": severity}
|
|
pending_text = m["msg"]
|
|
elif pending_text is not None:
|
|
pending_text += "\n" + line.strip()
|
|
|
|
if pending_text is not None:
|
|
yield _emit(pending_text, pending_meta)
|