优化多线程上传视频切片

service/upload_video.py

 import hashlib
 import hmac
 import json
+import math
 import os
 import random
 import threading
@@ -96,32 +97,63 @@ def video_split(file_path, max_workers=MAX_WORKERS):
     :param max_workers:
     :return:
     """
-    file_size = os.path.getsize(file_path)
-    min_chunk_size = 2 * 1024 * 1024  # 2MB
-
-    # 根据线程数和文件大小计算分片大小
-    chunk_size_bytes = max(min_chunk_size, int(file_size / max_workers))
     with open(file_path, 'rb') as file:
         file_data = file.read()
+    file_size = len(file_data)
+
+    # 配置分片大小参数
+    MIN_CHUNK_SIZE = 2 * 1024 * 1024  # 2MB最小分片
+    MAX_CHUNK_SIZE = 10 * 1024 * 1024  # 10MB最大分片
+
+    # 计算基础分片大小
+    base_chunk_size = max(MIN_CHUNK_SIZE, min(MAX_CHUNK_SIZE, file_size // max(1, max_workers)))
+
+    # 计算实际需要的分片数（向上取整）
+    total_slices_needed = (file_size + base_chunk_size - 1) // base_chunk_size
+
+    # 调整为线程数的倍数（但不超过文件大小限制）
+    total_slices = min(
+        max_workers * max(1, math.ceil(total_slices_needed / max_workers)),
+        math.ceil(file_size / MIN_CHUNK_SIZE)  # 最大不超过按最小分片大小计算的数量
+    )
+
+    # 重新计算实际分片大小（向上取整）
+    chunk_size_bytes = (file_size + total_slices - 1) // total_slices
+
+    # 确保分片大小在限制范围内
+    chunk_size_bytes = max(MIN_CHUNK_SIZE, min(MAX_CHUNK_SIZE, chunk_size_bytes))
+
+    # 重新计算实际分片数量（向上取整）
+    total_slices = (file_size + chunk_size_bytes - 1) // chunk_size_bytes
+
+    logger.debug(f"文件大小: {file_size} bytes | 线程数: {max_workers} | "
+                 f"分片大小: {chunk_size_bytes} bytes | 分片数量: {total_slices}")
 
-    total_slices = (len(file_data) + chunk_size_bytes - 1) // chunk_size_bytes
     slices = []
     chunks = {}
 
-    for i in tqdm(range(total_slices), desc='切片'):
+    for i in tqdm(range(total_slices), desc='优化切片'):
         start = i * chunk_size_bytes
-        end = min((i + 1) * chunk_size_bytes, len(file_data))
-        data = file_data[start:end]
+        end = min((i + 1) * chunk_size_bytes, file_size)
 
+        # 仅当分片在文件范围内才读取数据
+        if start < file_size:
+            data = file_data[start:end]
             crc32 = hex(zlib.crc32(data) & 0xFFFFFFFF)[2:].zfill(8)
-        # 每一段视频的请求参数信息
+        else:
+            # 超出文件范围的分片跳过
+            continue
+
         slice_info = {
-            'part_number': i + 1,  # 当前视频切片的顺序
-            'part_offset': start,  # 上传文件的写入地址
-            "crc32": crc32
+            'part_number': i + 1,
+            'part_offset': start,
+            "crc32": crc32,
+            "data_size": len(data)
         }
+
         chunks[start] = data
         slices.append(slice_info)
+
     return slices, chunks
 
 
@@ -653,7 +685,6 @@ def threaded_upload(upload_queue, upload, max_workers):
     """多线程执行分片上传"""
     success_count = 0
     total_slices = upload_queue.qsize()
-    logger.info(f"开始上传 {total_slices} 个分片")
     lock = threading.Lock()
     exceptions = []
 
@@ -669,7 +700,6 @@ def threaded_upload(upload_queue, upload, max_workers):
             try:
                 # 单分片上传逻辑
                 success = upload.upload_single_slice(task_item['slice'],task_item['chunk'])
-                logger.debug(f"分片 {task_item['slice']['part_number']}/{total_slices} 上传完成")
                 with lock:
                     if success:
                         success_count += 1