[MERGE] integrated EOL conversion support

[Alexander Belchenko]

(Fw to list. Sorry if it arrives twice.)

--- Исходное сообщение ---
От кого: John Arbash Meinel <john at arbash-meinel.com>
Кому: Alexander Belchenko <bialix at ukr.net>
Дата: 29 марта, 16:34:50
Тема: Re: [MERGE] integrated EOL conversion support

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1

    Alexander Belchenko wrote:
    > Ian Clatworthy пишет:
    >> Brian de Alwis wrote:
    >>> On 28-Mar-2009, at 12:41 PM, Alexander Belchenko wrote:
    >>>> I need the way to force lf on windows. How can I achieve this?
    >>> And similarly I need to force crlf on MacOS X (I do VCS ops from MacOS
    >>> X, but edit the files via VisualStudio running through VMWare Fusion).
    >> Done in the attached.
    >>
    >> "Are we there yet?" :-)
    >>
    >> Ian C.
    >
    > bb:approve
    >
    >
    >

    I'm still a bit concerned that the labels are the "wrong way" around. In
    that you are setting the line-endings in the repository, rather than
    settting the line-endings in the working tree. I'm pretty sure other
    systems like svn have the value indicate the working tree line endings.


I'm strongly disagree here. I think Ian approach is right thing.

If you have committed files with CRLF and you want eventually switch your windows-only development
to cross-platform development (or Linux-only development) you either have to convert all your files
to LF and therefore lost annotations, OR use crlf settings. Annotations are very important thing to
lost them, so Ian approach is right here. I know this because I've converted one my project in 2006
year, and this is reallly bad when you try to find "real" annotation. Believe me, it's a real pain.

May be labels is misleading? Yes, maybe. But approach is right.

I should admit that year ago I've used another approach: store all files only with LF in the repo
and convert to required format for working tree only. My approach was backward incompatible, because
it requires to break annotation. Ian approach is much better.


    (At least last I checked, 'crlf' meant native locally and crlf in the
    repo, rather than whatever in the repo and crlf when I'm working with it.)


How can you have "whatever" in the repo? You either lost annotations or will have big diffs from
revision to revision.


    I honestly think people care more about the value they get when they are
    working on the file, because it matters what other tools are going to do
    with it.

    I'm certainly not trying to block, just address a possible point of
    confusion.


I agree about *possible* confusion.


    John
    =:->
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.4.9 (Cygwin)
    Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

    iEYEARECAAYFAknPePoACgkQJdeBCYSNAAOKvACgsE9vzLy99jyOgLEjhv1g6dDz
    RKoAoLbcHCo2Kzh5WYY5X4zKeeO0oxRi
    =A1wG
    -----END PGP SIGNATURE-----