问题

我有以下配置我的日志记录器实例的代码行

```python
logger.add(sys.stderr, level=log_level, format="<green>{time:YYYY-MM-DD HH:mm:ss.SSS zz}</green> | <level>{level: <8}</level> | <yellow>Line {line: >4} ({file}):</yellow> <b>{message}</b>", colorize=True, backtrace=True, diagnose=True)

我想配置我的日志记录器，使用上面的配置，并将日志输出到文件，这样当我调用logger.whatever()时，它会在终端和日志文件中输出

这样，如果我在进行开发工作并直接运行文件时，我可以在终端中看到输出，而当代码由cronjob在服务器上运行时，它可以记录到文件中

我不太理解“sinks”的整个概念，所以如果这是一个简单的问题，我很抱歉

<details>
<summary>英文:</summary>

I have the following line that configures my logger instance

logger.add(sys.stderr, level=log_level, format="<green>{time:YYYY-MM-DD HH:mm:ss.SSS zz}</green> | <level>{level: <8}</level> | <yellow>Line {line: >4} ({file}):</yellow> <b>{message}</b>", colorize=True, backtrace=True, diagnose=True)

I want to configure my logger to use the above, and also output to a file, so that when I call logger.whatever() it outputs to the terminal and to a log file

This way, if I'm doing dev work and running the file directly, I can see the output in my terminal, and when the code is being run on a server by a cronjob, it can log to a file

I don't understand the whole concept of the "sinks" so sorry if this is an easy question

</details>


# 答案1
**得分**: 2

在日志记录的上下文中，“sink”指的是日志记录的输出目标。它可以是流（如`sys.stderr`或`sys.stdout`）、文件路径或接受日志消息作为输入的任何自定义函数。详见[文档](https://loguru.readthedocs.io/en/stable/api/logger.html#sink)以了解更多详情。

使用Loguru，你可以添加尽可能多的sinks。这意味着在你的情况下，你可以有两个sinks：一个写入到`sys.stderr`，另一个写入到`file.log`。

每当使用`logger.info()`记录消息时，比如，它将被发送到已添加的每个sink，并根据相应的配置进行处理。

<details>
<summary>英文:</summary>

In the context of logging, a &quot;sink&quot; refers to an output destination for log records. It can be a stream (such as `sys.stderr` or `sys.stdout`), a file path, or any custom function that accepts the logged message an input. See [documentation](https://loguru.readthedocs.io/en/stable/api/logger.html#sink) for more details.

Using Loguru, you can add as many sinks as you like. This means that in your case, you can have two sinks: one writing to `sys.stderr` and another one writing to `file.log`.

```python
log_level = &quot;DEBUG&quot;
log_format = &quot;&lt;green&gt;{time:YYYY-MM-DD HH:mm:ss.SSS zz}&lt;/green&gt; | &lt;level&gt;{level: &lt;8}&lt;/level&gt; | &lt;yellow&gt;Line {line: &gt;4} ({file}):&lt;/yellow&gt; &lt;b&gt;{message}&lt;/b&gt;&quot;
logger.add(sys.stderr, level=log_level, format=log_format, colorize=True, backtrace=True, diagnose=True)
logger.add(&quot;file.log&quot;, level=log_level, format=log_format, colorize=False, backtrace=True, diagnose=True)

Each time a message is logged with logger.info() for example, it will be sent to each of the added sinks and processed based on the the corresponding configuration.