[Monotone-devel] Re: Bug in CRLF conversions

[Bruce Stephens]

Ethan Blanton <address@hidden> writes:

[...]

> This is precisely why Yury is correct.  Because, on many systems, \r
> and \n have _different meanings_, they should be properly and
> reversibly preserved even in text files.

I can believe that's true of some text files.  But for (to take a
random example) C++ source files, it's surely not true, is it?  

I'd expect to get C++ source files with native line endings, and to be
able to commit them with native line endings, and I'd be surprised if
I had to do something special to avoid ending up with CRLF endings on
Unix, or LF endings on Windows.

So how can we have both, while keeping the solution sufficiently
comprehensible that people can work out how to get it to work for
them?

[...]