Question

我有一个String对象，现在因为它包含一个包含不同文件类型的文件夹的差异，而不是其中的所有内容都编码在同一个字符集中。

正确的代码在字符串中，但每当我尝试访问字符串时，groovy会尝试提供帮助并解码字符串，这会使事情变得混乱。

现在以下似乎做了我需要的事情

String decoded_diff = "String that contains codes from different character encodings"
patch_file_name = 'changes.patch'
patch_file =  new File(pwd(), patch_file_name)
patch_file.delete()
max_block_size = 1024 * 1024
char[] char_buffer = new char[max_block_size]
block_start = 0
patch_length = decoded_diff.length()
while (true) {
    block_size = Math.min(patch_length - block_start, max_block_size)
    decoded_diff.getChars(block_start, block_start + block_size, char_buffer, 0)
    block_start += block_size
    byte[] byte_buffer = new byte[block_size]
    for (int i = 0; i < block_size; i++) {
        byte_buffer[i] = (int) char_buffer[i]
    }
    patch_file.append(byte_buffer)
    if (block_start == patch_length) break
}

然而，它是sloooow

有没有更快的方法来实现同样的目标？最终的补丁文件必须与原始差异相同才能工作。不幸的是我无法发送文件本身（jenkins目前不支持管道作业中的文件参数）所以我必须将其转义并将其作为json参数列表的一部分发送，因此接收端的这种痛苦的严重性。

Answer 1

为什么不：

String decoded_diff = "String that contains codes from different character encodings"
patch_file_name = 'changes.patch'
patch_file =  new File(pwd(), patch_file_name)
patch_file.delete()

patchFile.withOutputStream { os ->
    os << decoded_diff.bytes
}

在Groovy中将字符串写入磁盘而不进行任何字符转换

1 个答案: