txai-dev/yu-zhi-ran
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: 合规审查卡死修复 + 小红书复制格式 + today-only过滤

Browse Source
This commit is contained in:
Yuzhiran Dev
2026-05-27 18:25:30 +08:00
parent 6585909ffc
commit bd3228806d
14 changed files with 278 additions and 148 deletions
Download Patch File Download Diff File Expand all files Collapse all files
platform/backend/app/core/collector.py
+31 -19
View File
@@ -8,6 +8,7 @@ from pathlib import Path
import logging
import os
import time
import threading
from typing import Optional, Dict
logger = logging.getLogger(__name__)
@@ -17,6 +18,7 @@ if os.getenv('PROJECT_ROOT'):
PROJECT_ROOT = Path(os.getenv('PROJECT_ROOT'))
_running_processes: Dict[str, dict] = {}
_lock = threading.Lock()
def _get_cmd():
script_path = PROJECT_ROOT / "scripts" / "collector.py"
@@ -29,21 +31,26 @@ def _get_cmd():
def run_collector():
"""运行选题收集脚本（非阻塞，后台运行）"""
cmd = _get_cmd()
proc = subprocess.Popen(
cmd,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
cwd=PROJECT_ROOT
)
_running_processes["collector"] = {
"pid": proc.pid,
"started_at": time.time(),
"process": proc
}
with _lock:
existing = _running_processes.get("collector", {})
proc = existing.get("process")
if proc and proc.poll() is None:
raise RuntimeError("已有内容采集任务正在运行")
cmd = _get_cmd()
proc = subprocess.Popen(
cmd,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
cwd=PROJECT_ROOT
)
_running_processes["collector"] = {
"pid": proc.pid,
"started_at": time.time(),
"process": proc,
}
logger.info("Collector started in background (pid=%s)", proc.pid)
return {"ok": True, "pid": proc.pid}
return {"ok": True, "pid": proc.pid, "_proc": proc}
def run_collector_blocking(timeout: int = 300):
"""运行选题收集脚本（阻塞，带超时，给定时任务用）"""
@@ -60,22 +67,27 @@ def run_collector_blocking(timeout: int = 300):
return {"ok": True, "output": result.stdout}
def get_collector_status() -> Optional[Dict]:
"""获取当前采集任务状态"""
"""获取当前采集任务状态（不消费子进程 stdout，留给 _monitor_subprocess）"""
info = _running_processes.get("collector")
if not info:
return None
proc: subprocess.Popen = info["process"]
if "finished" in info:
return {
"status": "completed" if info.get("returncode") == 0 else "failed",
"pid": info["pid"],
"elapsed": round(time.time() - info["started_at"], 1),
"returncode": info.get("returncode"),
}
if proc.poll() is not None:
stdout, stderr = proc.communicate()
info["returncode"] = proc.returncode
info["finished"] = True
elapsed = time.time() - info["started_at"]
del _running_processes["collector"]
return {
"status": "completed" if proc.returncode == 0 else "failed",
"pid": info["pid"],
"elapsed": round(elapsed, 1),
"returncode": proc.returncode,
"stdout": stdout.strip()[-500:],
"stderr": stderr.strip()[-500:],
}
return {
"status": "running",
platform/backend/app/core/generator.py
+19 -7
View File
@@ -2,8 +2,10 @@ import subprocess
from pathlib import Path
import logging
import os
import json
import time
from typing import Optional, Dict
import threading
from typing import Optional, Dict, List
logger = logging.getLogger(__name__)
@@ -12,6 +14,7 @@ if os.getenv('PROJECT_ROOT'):
PROJECT_ROOT = Path(os.getenv('PROJECT_ROOT'))
_running_processes: Dict[str, dict] = {}
_lock = threading.Lock()
def _get_cmd(topic_id: str = None):
script_path = PROJECT_ROOT / "scripts" / "creator.py"
@@ -21,15 +24,22 @@ def _get_cmd(topic_id: str = None):
cmd = [str(venv_python), str(script_path)] if venv_python.exists() else ["python3", str(script_path)]
if topic_id:
cmd.extend(["--topic-id", topic_id])
else:
cmd.append("--today-only")
return cmd
def run_creator(topic_id: str = None):
"""非阻塞：后台启动创作脚本，立即返回"""
cmd = _get_cmd(topic_id)
proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, cwd=str(PROJECT_ROOT))
_running_processes["generator"] = {"pid": proc.pid, "started_at": time.time(), "process": proc, "topic_id": topic_id}
with _lock:
existing = _running_processes.get("generator", {})
proc = existing.get("process")
if proc and proc.poll() is None:
raise RuntimeError("已有创作任务正在运行")
cmd = _get_cmd(topic_id)
proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, cwd=str(PROJECT_ROOT))
_running_processes["generator"] = {"pid": proc.pid, "started_at": time.time(), "process": proc, "topic_id": topic_id}
logger.info("Creator started in background (pid=%s, topic_id=%s)", proc.pid, topic_id)
return {"ok": True, "pid": proc.pid, "topic_id": topic_id}
return {"ok": True, "pid": proc.pid, "topic_id": topic_id, "_proc": proc}
def run_creator_blocking(topic_id: str = None, timeout: int = 1800):
"""阻塞版：带超时，给定时任务使用"""
@@ -93,9 +103,11 @@ def get_generator_status() -> Optional[Dict]:
if not info:
return None
proc = info["process"]
if "finished" in info:
return {"status": "completed" if info.get("returncode") == 0 else "failed", "pid": info["pid"], "elapsed": round(time.time() - info["started_at"], 1), "returncode": info.get("returncode"), "topic_id": info.get("topic_id")}
if proc.poll() is not None:
stdout, stderr = proc.communicate()
info["returncode"] = proc.returncode
info["finished"] = True
elapsed = time.time() - info["started_at"]
del _running_processes["generator"]
return {"status": "completed" if proc.returncode == 0 else "failed", "pid": info["pid"], "elapsed": round(elapsed, 1), "returncode": proc.returncode, "topic_id": info["topic_id"]}
return {"status": "running", "pid": info["pid"], "elapsed": round(time.time() - info["started_at"], 1)}
platform/backend/app/core/optimizer.py
+18 -7
View File
@@ -4,6 +4,7 @@ import logging
import os
import json
import time
import threading
from datetime import datetime
from typing import List, Optional, Dict
@@ -14,6 +15,7 @@ if os.getenv('PROJECT_ROOT'):
PROJECT_ROOT = Path(os.getenv('PROJECT_ROOT'))
_running_processes: Dict[str, dict] = {}
_lock = threading.Lock()
def _get_cmd(topic_ids: List[str] = None):
script_path = PROJECT_ROOT / "scripts" / "compliance_optimizer.py"
@@ -23,15 +25,22 @@ def _get_cmd(topic_ids: List[str] = None):
cmd = [str(venv_python), str(script_path)] if venv_python.exists() else ["python3", str(script_path)]
if topic_ids:
cmd.extend(["--topic-ids", ','.join(topic_ids)])
else:
cmd.append("--today-only")
return cmd
def run_optimizer(topic_ids: List[str] = None):
"""非阻塞：后台启动合规审查脚本，立即返回"""
cmd = _get_cmd(topic_ids)
proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, cwd=str(PROJECT_ROOT))
_running_processes["optimizer"] = {"pid": proc.pid, "started_at": time.time(), "process": proc, "topic_ids": topic_ids}
with _lock:
existing = _running_processes.get("optimizer", {})
proc = existing.get("process")
if proc and proc.poll() is None:
raise RuntimeError("已有合规审查任务正在运行，请等待完成后再试")
cmd = _get_cmd(topic_ids)
proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, cwd=str(PROJECT_ROOT))
_running_processes["optimizer"] = {"pid": proc.pid, "started_at": time.time(), "process": proc, "topic_ids": topic_ids}
logger.info("Optimizer started in background (pid=%s, topic_ids=%s)", proc.pid, topic_ids)
return {"ok": True, "pid": proc.pid, "topic_ids": topic_ids}
return {"ok": True, "pid": proc.pid, "topic_ids": topic_ids, "_proc": proc}
def run_optimizer_blocking(topic_ids: List[str] = None, timeout: int = 600):
"""阻塞版：带超时，给定时任务使用"""
@@ -54,9 +63,11 @@ def get_optimizer_status() -> Optional[Dict]:
if not info:
return None
proc = info["process"]
if "finished" in info:
return {"status": "completed" if info.get("returncode") == 0 else "failed", "pid": info["pid"], "elapsed": round(time.time() - info["started_at"], 1), "returncode": info.get("returncode")}
if proc.poll() is not None:
stdout, stderr = proc.communicate()
info["returncode"] = proc.returncode
info["finished"] = True
elapsed = time.time() - info["started_at"]
del _running_processes["optimizer"]
return {"status": "completed" if proc.returncode == 0 else "failed", "pid": info["pid"], "elapsed": round(elapsed, 1), "returncode": proc.returncode, "stdout": stdout.strip()[-300:], "stderr": stderr.strip()[-300:]}
return {"status": "completed" if proc.returncode == 0 else "failed", "pid": info["pid"], "elapsed": round(elapsed, 1), "returncode": proc.returncode}
return {"status": "running", "pid": info["pid"], "elapsed": round(time.time() - info["started_at"], 1)}
platform/backend/app/core/scheduler.py
+40 -12
View File
@@ -31,10 +31,38 @@ MODULES = {
"scheduled_task_monitor": {"name": "⏰ 任务监控", "cron": "*"},
}
LOG_FILE_MAP = {
"scheduled_refresh_search_cache": "opencode_search",
"scheduled_fetch_trends": "trends",
"scheduled_collect": "collector",
"scheduled_generate": "creator",
"scheduled_optimize": "optimizer",
"scheduled_optimize_sources": "optimizer_sources",
"scheduled_metrics_sync": "metrics_sync",
"scheduled_reset_search_usage": "reset_search_usage",
"scheduled_task_monitor": "task_monitor",
}
def _log_to_file(module_id: str, status: str, message: str = None, error_trace: str = None):
log_name = LOG_FILE_MAP.get(module_id, module_id)
log_dir = PROJECT_ROOT / "automation" / "logs"
log_dir.mkdir(parents=True, exist_ok=True)
log_file = log_dir / f"{log_name}_{datetime.now().strftime('%Y-%m-%d')}.log"
ts = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
line = f"[{ts}] [{status.upper()}] {message or ''}"
if error_trace:
line += f"\n{error_trace[:500]}"
try:
with open(log_file, 'a', encoding='utf-8') as f:
f.write(line + "\n")
except Exception:
pass
def _log_task(module_id: str, status: str, message: str = None,
error_trace: str = None, result_data: dict = None,
started_at: datetime = None, finished_at: datetime = None,
triggered_by: str = "scheduler", next_run_time: datetime = None):
_log_to_file(module_id, status, message, error_trace)
try:
from ..database import SessionLocal
from ..models import TaskLog
@@ -159,7 +187,7 @@ class TaskScheduler:
logger.info("[Scheduled] Fetching hot trends...")
import subprocess
result = subprocess.run(
[sys.executable, str(Path(__file__).parent.parent.parent.parent / "scripts" / "trends.py")],
[sys.executable, str(PROJECT_ROOT / "scripts" / "trends.py")],
capture_output=True, text=True, timeout=120
)
if result.returncode == 0:
@@ -190,7 +218,7 @@ class TaskScheduler:
logger.info("[Scheduled] Refreshing search cache via opencode...")
import subprocess
result = subprocess.run(
[sys.executable, str(Path(__file__).parent.parent.parent.parent / "scripts" / "opencode_search.py"), "--refresh-cache"],
[sys.executable, str(PROJECT_ROOT / "scripts" / "opencode_search.py"), "--refresh-cache"],
capture_output=True, text=True, timeout=600
)
for line in result.stdout.strip().split("\n"):
@@ -285,10 +313,10 @@ class TaskScheduler:
started_at=started, finished_at=datetime.now(timezone.utc))
logger.exception("[Scheduled] Collection failed: %s", e)
def _run_optimize_sources(self):
def _run_optimize_sources(self, triggered_by="scheduler"):
"""AI自动优化采集类别与信息源：对比市场热点和当前配置，给出调整建议"""
started = datetime.now(timezone.utc)
_log_task("scheduled_optimize_sources", "running", started_at=started)
_log_task("scheduled_optimize_sources", "running", started_at=started, triggered_by=triggered_by)
try:
logger.info("[Scheduled] Starting source optimization with AI...")
from .nvidia_client import call_llm
@@ -337,19 +365,19 @@ class TaskScheduler:
"sources_assessed": len(result.get("source_assessment", [])),
"suggested_cats": len(result.get("suggested_new_categories", [])),
"suggested_srcs": len(result.get("suggested_new_sources", []))},
started_at=started, finished_at=datetime.now(timezone.utc))
started_at=started, finished_at=datetime.now(timezone.utc), triggered_by=triggered_by)
db.close()
except Exception as e:
_log_task("scheduled_optimize_sources", "failed",
message=str(e),
error_trace=traceback.format_exc(),
started_at=started, finished_at=datetime.now(timezone.utc))
started_at=started, finished_at=datetime.now(timezone.utc), triggered_by=triggered_by)
logger.exception("[Scheduled] Source AI optimization failed: %s", e)
def _run_metrics_sync(self):
def _run_metrics_sync(self, triggered_by="scheduler"):
"""定时从各平台公开API获取发布文章的效果数据（当前仅支持知乎）"""
started = datetime.now(timezone.utc)
_log_task("scheduled_metrics_sync", "running", started_at=started)
_log_task("scheduled_metrics_sync", "running", started_at=started, triggered_by=triggered_by)
try:
logger.info("[Scheduled] Starting metrics sync (zhihu auto-fetch)...")
from ..database import SessionLocal
@@ -411,7 +439,7 @@ class TaskScheduler:
_log_task("scheduled_metrics_sync", "success",
message=f"同步完成，{count} 篇知乎文章",
result_data={"articles_synced": count},
started_at=started, finished_at=datetime.now(timezone.utc))
started_at=started, finished_at=datetime.now(timezone.utc), triggered_by=triggered_by)
# 生成指标反馈：按 field 聚合表现，写入 metrics_feedback.json 供 collector 读取
try:
import json as json_mod
@@ -436,7 +464,7 @@ class TaskScheduler:
"top_domains": [(f, s) for f, s, _ in scored[:5]],
"detail": [{"field": f, "score": s, "articles": c} for f, s, c in scored],
}
feedback_file = Path(__file__).parent.parent.parent.parent / "automation" / "data" / "metrics_feedback.json"
feedback_file = PROJECT_ROOT / "automation" / "data" / "metrics_feedback.json"
feedback_file.parent.mkdir(parents=True, exist_ok=True)
feedback_file.write_text(json_mod.dumps(feedback_data, ensure_ascii=False, indent=2), encoding='utf-8')
logger.info("[Scheduled] Metrics feedback written: top domain %s (score %.1f)", scored[0][0], scored[0][1])
@@ -445,14 +473,14 @@ class TaskScheduler:
else:
_log_task("scheduled_metrics_sync", "success",
message="无已发布的知乎文章",
started_at=started, finished_at=datetime.now(timezone.utc))
started_at=started, finished_at=datetime.now(timezone.utc), triggered_by=triggered_by)
logger.info("[Scheduled] Metrics sync: no zhihu articles to sync")
db.close()
except Exception as e:
_log_task("scheduled_metrics_sync", "failed",
message=str(e),
error_trace=traceback.format_exc(),
started_at=started, finished_at=datetime.now(timezone.utc))
started_at=started, finished_at=datetime.now(timezone.utc), triggered_by=triggered_by)
logger.exception("[Scheduled] Metrics sync failed: %s", e)
def _run_reset_search_usage(self):