import threading
|
from datetime import datetime
|
from pathlib import Path
|
from auto_media_publisher.modifiers.rewriter.docx_rewriter_gemini import GeminiDocxRewriter # 根据你的路径修改
|
import random
|
import time
|
|
class DocxRewriteWorkerThread(threading.Thread):
|
def __init__(self, input_dir, output_dir, interval_minutes=10, rewrite_style="random"):
|
super().__init__()
|
self.input_dir = Path(input_dir)
|
self.output_dir = Path(output_dir)
|
self.interval = interval_minutes * 60 # 转换成秒
|
self.rewriter = GeminiDocxRewriter(rewrite_style=rewrite_style)
|
self.stop_event = threading.Event()
|
|
def stop(self):
|
self.stop_event.set()
|
|
def run(self):
|
while not self.stop_event.is_set():
|
print(f"[DocxRewriteThread] Scanning for .docx files at {datetime.now().isoformat()}")
|
for docx_path in self.input_dir.rglob("*.docx"):
|
try:
|
rel_path = docx_path.relative_to(self.input_dir)
|
timestamp_prefix = datetime.now().strftime("%Y%m%d%H%M%S")
|
new_filename = f"{timestamp_prefix}_{docx_path.name}"
|
output_path = self.output_dir / rel_path.parent / new_filename
|
|
output_path.parent.mkdir(parents=True, exist_ok=True)
|
|
print(f"[DocxRewriteThread] Rewriting: {docx_path}")
|
self.rewriter.rewrite(str(docx_path), str(output_path))
|
|
sleep_duration = random.uniform(10, 15) # 秒
|
print(f"[DocxRewriteThread] Sleeping for {sleep_duration:.1f}s to avoid rate limit.")
|
time.sleep(sleep_duration)
|
|
except Exception as e:
|
print(f"[DocxRewriteThread] Failed to process {docx_path}: {e}")
|
|
|
print(f"[DocxRewriteThread] Waiting for {self.interval // 60} minutes...")
|
self.stop_event.wait(self.interval)
|
