我有一个fs。指向一个相当大的文件的 ReadStream 对象。现在我想从 ReadStream 读取 8000 字节,但内部缓冲区只有 6000 字节。所以我的方法是读取这 6000 个字节,然后通过使用 while-loop 来检查内部缓冲区长度是否不再为 0,从而等待内部缓冲区再次填满。
像这样:
BinaryObject.prototype.read = function(length) {
var value;
// Check whether we have enough data in the internal buffer
if (this.stream._readableState.length < length) {
// Not enough data - read the full internal buffer to
// force the ReadStream to fill it again.
value = this.read(this.stream._readableState.length);
while (this.stream._readableState.length === 0) {
// Wait...?
}
// We should have some more data in the internal buffer
// here... Read the rest and add it to our `value` buffer
// ... something like this:
//
// value.push(this.stream.read(length - value.length))
// return value
} else {
value = this.stream.read(length);
this.stream.position += length;
return value;
}
};
问题是,缓冲区不再填充 - 脚本只会在 while 循环中空闲。
最好的方法是什么?
这很简单。您无需在自己这边进行任何缓冲:
var fs = require('fs'),
rs = fs.createReadStream('/path/to/file');
var CHUNK_SIZE = 8192;
rs.on('readable', function () {
var chunk;
while (null !== (chunk = rs.read(CHUNK_SIZE))) {
console.log('got %d bytes of data', chunk.length);
}
});
rs.on('end', function () {
console.log('end');
});
如果CHUNK_SIZE
大于内部缓冲区,则节点将返回 null 并在再次发出readable
之前再缓冲一些。您甚至可以通过传递以下内容来配置缓冲区的初始大小:
var rs = fs.createReadStream('/path/to/file', {highWatermark: CHUNK_SIZE});
下面是在流中读取文件的示例。
var fs = require('fs'),
readStream = fs.createReadStream(srcPath);
readStream.on('data', function (chunk) {
console.log('got %d bytes of data', chunk.length);
});
readStream.on('readable', function () {
var chunk;
while (null !== (chunk = readStream.read())) {
console.log('got %d bytes of data', chunk.length);
}
});
readStream.on('end', function () {
console.log('got all bytes of data');
});