Hello there,

I was reading the `filecmp` module and I noticed that it defined the BUFSIZE manually, I think it's better to stick to the io.DEFAULT_BUFFER_SIZE variable for both consistency and easy to maintain in the future.


Cheers,

Is there some reason a priori that these constants should be defined to be the same rather than being the same coincidentally?

Well, I think because the io module provides Python’s main facilities for dealing with various types of I/O, so using io.DEFAULT_BUFFER_SIZE is a kind of standard buffer size. I don't see any scientific reason for choosing the 8 * 1024 inside the filecmp module.

My guess is that guy wanted to use something that is fairly small and would be okay in different circumstances (OSs) and the idea of DEFAULT_BUFFER_SIZE is the same thing.

I think, the chances are high that in the future as computers reinforce, we might change the buffer size as well. So using the io.DEFAULT_BUFFER_SIZE will guarantee that we only change io.DEFAULT_BUFFER_SIZE and every module would be okay automatically?

I don't think there's a strong reason these values need to be same. Note, even today, they likely won't be the same since io.DEFAULT_BUFFER_SIZE is only a fallback buffer size used if st_blksize is not set.

I concur with Benjamin.

Although two variables are similar, they are used in the different layer.