[Numpy-discussion] whitespace in git repo

Friedrich Romstedt friedrichromstedt@gmail....
Wed Oct 20 14:12:51 CDT 2010


.nsi.in has '$\n\r'  Is this now LFCR or CRLF?

doall.py and py3tool.py have '$\n' precisely.  I'm on Mac.  \n is the
at least vim-default on Mac.

So yes, confirming that it's already in the repo.

What about the svg eols?

Do we need my branch now, or do we want to leave it alone?  When
others change the file, it might get converted, when we do not apply
my commit, assumed that they have some personal preference about eols.
 For the standard user he will not notice the difference.  Someone
else than me has to think through it for approval.  (I thought thru
and say it's okay.)

"When deciding what attributes are assigned to a path, git consults
$GIT_DIR/info/attributes file (which has the highest precedence),
.gitattributes file in the same directory as the path in question, and
its parent directories up to the toplevel of the work tree (the
further the directory that contains .gitattributes is from the path in
question, the lower its precedence). Finally global and system-wide
files are considered (they have the lowest precedence)."

I tried to put it in the same dir but it didn't work for me.  Maybe my
git is ageing.

Used $grep -PU '$\r' and similar; '$\r^' does not work, just for completeness.

The section "End-of-line conversion" in
http://www.kernel.org/pub/software/scm/git/docs/gitattributes.html is
quite clear.  Considering this section I come to the conclusion to
leave it to the user how he/she checks in/out the files, but we define
for files with defined gitattributes the attributes in .gitattributes.
 (Alternative to .gitattributes: $GIT_DIR/info/attributes, I don't
know if this goes into the repo too).  People are advised then to not
attribute files to themselves by just an eol conversion.

Opinions, Comments?  I know, this topic is unattractive.
http://www.astro.gla.ac.uk/users/labrosse/PORTFOLIO/images/cosmicsexinessladder.png
(from http://www.astro.gla.ac.uk/users/labrosse/PORTFOLIO/philo.html)

This is somewhere at the other end of the world :-)

Friedrich

2010/10/20 Charles R Harris <charlesr.harris@gmail.com>:
> On Wed, Oct 20, 2010 at 9:56 AM, Friedrich Romstedt
> <friedrichromstedt@gmail.com> wrote:
>> 2010/10/20 Darren Dale <dsdale24@gmail.com>:
>> > On Wed, Oct 20, 2010 at 6:12 AM, Friedrich Romstedt
>> > <friedrichromstedt@gmail.com> wrote:
>> >> Due to Darren's config file the .nsi.in file made it with CRLF into the
>> >> repo.
>> >
>> > Uh, no.
>>
>> You mean I'm wrong?  Wait, I'll check, but ...Hmm, how do I check the
>> line endings?  `file` told it for the .nsi.in file, but not for the
>> .py ones.  I was starting from the assumtion that Mac OS X native form
>> is LF, or even CR, but not CRLF, which is Windows?
>>
>> Sorry if I wasn't diligent enough or too sure.
>>
>
> You can grep for CR: grep -P '\r' foo . On windows and mac you might need to
> add -U. For vim ":set list" will show the line endings.


More information about the NumPy-Discussion mailing list