英文:
Pack array of int16_t into multiple uint64_t
问题
以下是您提供的代码的翻译部分：
我有一个包含N个6个条目的数组```cppconst int16_t xzyabc[][6] PROGMEM = {{1, 2, 3, 4, 5, 6},{7, 8, 9, 10, 11, 12},};数组中的每个值都是int16_t数字。对于每四个int16_t，我想将它们打包到一个uint64_t中。我尝试了下面的代码。
void Telemetry::loop(){const size_t numArrays = sizeof(xzyabc) / sizeof(xzyabc[0]);const size_t elementsPerArray = sizeof(xzyabc[0]) / sizeof(xzyabc[0][0]);const size_t elementsPerUint64 = 4;const size_t numUint64 = (numArrays * elementsPerArray + elementsPerUint64 - 1) / elementsPerUint64;uint64_t convertedArray[numUint64];for (size_t i = 0; i < numUint64; i++){uint64_t value = 0;for (size_t j = 0; j < elementsPerUint64; j++){size_t index = i * elementsPerUint64 + j;if (index < numArrays * elementsPerArray){int16_t element = xzyabc[index / elementsPerArray][index % elementsPerArray];value |= static_cast<uint64_t>(element) << (16 * j);}}uint32_t upperPart = (uint32_t)(value >> 32);uint32_t lowerPart = (uint32_t)(value & 0xFFFFFFFF);Serial.print("Number: ");Serial.print(upperPart);Serial.print(lowerPart);Serial.println();convertedArray[i] = value;}callback(convertedArray, numUint64);}
上面的代码的思想是遍历数组，对于每四个元素，将它们放入一个无符号整数中。
结果是：
Number: 262147131073Number: 524295393221Number: 786443655369
这些结果看起来对我来说是错误的，我尝试在NodeJS中解包它失败了。
我该如何修复这个问题？或者我做错了什么？
<details><summary>英文:</summary>I have an array of N per 6 entries
const int16_t xzyabc[][6] PROGMEM = {
{1, 2, 3, 4, 5, 6},
{7, 8, 9, 10, 11, 12},
};
Each value in the array is an int16_t number.
For each four int16_t, I want to pack it onto a uint64_t.
I tried the code below.
void Telemetry::loop()
{
const size_t numArrays = sizeof(xzyabc) / sizeof(xzyabc[0]);
const size_t elementsPerArray = sizeof(xzyabc[0]) / sizeof(xzyabc[0][0]);
const size_t elementsPerUint64 = 4;
const size_t numUint64 = (numArrays * elementsPerArray + elementsPerUint64 - 1) / elementsPerUint64;
uint64_t convertedArray[numUint64];
for (size_t i = 0; i < numUint64; i++)
{
uint64_t value = 0;
for (size_t j = 0; j < elementsPerUint64; j++)
{
size_t index = i * elementsPerUint64 + j;
if (index < numArrays * elementsPerArray)
{
int16_t element = xzyabc[index / elementsPerArray][index % elementsPerArray];
value |= static_cast<uint64_t>(element) << (16 * j);
}
}
uint32_t parteSuperior = (uint32_t)(value >> 32);uint32_t parteInferior = (uint32_t)(value & 0xFFFFFFFF);Serial.print("Number: ");Serial.print(parteSuperior);Serial.print(parteInferior);Serial.println();convertedArray[i] = value;
}
callback(convertedArray, numUint64);
}
The idea of the code above is to iterate over the array and for each quadrupled, put it on a unsigned integer.Results in:
Number: 262147131073
Number: 524295393221
Number: 786443655369
And they look wrong to me, I tried to unpack it on NodeJS and it failed.How can I fix this? Or what I'm doing wrong?</details># 答案1**得分**: 0根据@Peter的建议，我最终使用了memcpy函数。```cppuint64_t buffer[10]{0};memcpy(buffer, xzyabc, sizeof(xzyabc));
英文:
As sugestion of @Peter, I ended using memcpy
uint64_t buffer[10]{0};memcpy(buffer, xzyabc, sizeof(xzyabc));
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论