如何在C#中对大缓冲区分区，获取起始索引、结束索引及长度？

Let's walk through how to split your buffer into the required number of partitions, focusing on getting the start index, end index, and length for each segment—perfect for writing data to a single large buffer like a custom buffer pool.

Core Calculation Logic #

The goal is to split the buffer as evenly as possible. When the total length isn't perfectly divisible by the number of partitions, a small number of partitions will be 1 byte longer than the rest. Here's the step-by-step approach:

1. Calculate the base length for most partitions: base_length = total_length // num_partitions (integer division)
2. Find the remainder: remainder = total_length % num_partitions (this tells us how many partitions need to be 1 byte longer)
3. The first remainder partitions will have a length of base_length + 1; the rest will use base_length
4. Iterate through each partition to compute start/end indices:
   • The first partition starts at 0
   • For each subsequent partition, start index = previous partition's end index + 1
   • End index = start index + partition length - 1

Example Walkthrough (Your Total Length = 14, 5 Partitions) #

Let's verify this with your sample case to confirm it works:

• Total length: 14, partitions: 5
• base_length = 14 // 5 = 2
• remainder = 14 % 5 = 4 → first 4 partitions are length 3, last is 2

Which matches your expected output exactly:

• Partition 1: Start=0 || End=2 || Length=3
• Partition 2: Start=3 || End=5 || Length=3
• Partition 3: Start=6 || End=8 || Length=3
• Partition 4: Start=9 || End=11 || Length=3
• Partition 5: Start=12 || End=13 || Length=2

Applying to Your Buffer (Size = 1048576, 5 Partitions) #

Now let's compute this for your actual buffer:

• Total length: 1048576, partitions: 5
• base_length = 1048576 // 5 = 209715
• remainder = 1048576 % 5 = 1 → only the first partition is 1 byte longer, others use the base length

Here are your exact partition details:

• Partition 1: Start=0 || End=209715 || Length=209716
• Partition 2: Start=209716 || End=419430 || Length=209715
• Partition 3: Start=419431 || End=629145 || Length=209715
• Partition 4: Start=629146 || End=838860 || Length=209715
• Partition 5: Start=838861 || End=1048575 || Length=209715

Quick Automation Script (Python Example) #

If you want to generate these partitions programmatically, here's a simple script:

def split_buffer(total_length, num_partitions):
    base_len = total_length // num_partitions
    remainder = total_length % num_partitions
    partitions = []
    current_start = 0
    for i in range(num_partitions):
        # Determine length for current partition
        length = base_len + 1 if i < remainder else base_len
        current_end = current_start + length - 1
        # Store partition details
        partitions.append({
            "start": current_start,
            "end": current_end,
            "length": length
        })
        # Move to next partition's start
        current_start = current_end + 1
    return partitions

# Test with your buffer
buffer_size = 1048576
num_parts = 5
result = split_buffer(buffer_size, num_parts)
for idx, part in enumerate(result, 1):
    print(f"Partition {idx}: Start={part['start']} || End={part['end']} || Length={part['length']}")

内容的提问来源于stack exchange，提问作者Robert Miller