Re: [git-users] HELP git cant recognize ANY commands

2015-10-13 Thread Konstantin Khomoutov
On Mon, 12 Oct 2015 23:10:52 +0200
Magnus Therning  wrote:

> > 
> > This is what I got...(Hope I did it right...). The git version is 
> > 2.6.1.windows.1
> 
> All we see there is that you can't run `mount`. I suspect you'll have
> more luck with an image of your attempt to actually run `git`.

My take on this is that your suggestion won't add anything: the
contents of the PATH env. variable suggests the crucial bits of the GfW
installation are not there, and hence the Git entry binary (script in
GfW? -- I can't remember) simply won't be found.

I have no idea why the environment in the OP's case looks the way it
looks, though.

My take is that he did not install GfW "as is" but rather it came
bundled with some other software package, and is just broken when the
user tries to use it outside of that package.

-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[git-users] creating a new commit after rework

2015-10-13 Thread nmh
i did 

git clone.
made some changes.
git status shows changed files.

did 
git commit -a 
git review.

(a,b,c,d,e,f)(lets say these are files changed in this commit)
all okay.
--

Now got some review comments,
worked on files e,f -> e' f'

git status 
shows changed files e' f' only 

For some reason, i wanted to create a separate commit, not a patch for 
these 6 files now.
(a,b,c,d,e',f') 


How to do that ?

if i do 
git commit -a (is it right ?)(Will it take only e' f' or all 6 files 
a,b,c,d,e',f' ?)
==












-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [git-users] Re: Undefined symbols for architecture x86_64: "_reencode_string_iconv"

2015-10-13 Thread Konstantin Khomoutov
On Mon, 12 Oct 2015 12:54:20 -0700 (PDT)
mham...@sandia.gov wrote:

> FWIW, I'm doing a farily straightforward build from source
> 
> tar zxf git-2.6.1.tar.gz
> cd git-2.6.1
> export NO_APPLE_COMMON_CRYPTO=1
> make configure
> ./configure --prefix=/blah/blah/blah
> make

As usually, is [1] what you're looking for?

In either case, we're unlikely to help much with such errors as this
list is for helping users use Git, not build it ;-)
You might have better luck asking on the main Git list instead -- be
sure to read [2] for pointers on how to do this.

1. https://trac.macports.org/ticket/38851
2. https://gist.github.com/tfnico/4441562

-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[git-users] git status after git commit

2015-10-13 Thread nmh
i did 
git clone

made some changes,

git status show files changed.
(a,b,c,d) (lets assume)

git commit -a 

git review.


=

now i am back in my local dir.
git status

git status
On branch bug/1436865
nothing to commit, working directory clean

but i want to rework upon those files.

i want to see the files changed in git status and git diff again.

i want to see their prev version so that i can see 
git diff properly and work upon changes.

how to do that ?


 


-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [git-users] creating a new commit after rework

2015-10-13 Thread Gergely Polonkai
I guess what you need is an amended commit:

git add e f
git commit --amend
git review

This way your old commit will be replaced with a new one, and thus Gerrit
will create a new patch set for you (instead of a new change).

Best,
Gergely
On 13 Oct 2015 11:19, "nmh"  wrote:

> i did
>
> git clone.
> made some changes.
> git status shows changed files.
>
> did
> git commit -a
> git review.
>
> (a,b,c,d,e,f)(lets say these are files changed in this commit)
> all okay.
> --
>
> Now got some review comments,
> worked on files e,f -> e' f'
>
> git status
> shows changed files e' f' only
>
> For some reason, i wanted to create a separate commit, not a patch for
> these 6 files now.
> (a,b,c,d,e',f')
>
>
> How to do that ?
>
> if i do
> git commit -a (is it right ?)(Will it take only e' f' or all 6 files
> a,b,c,d,e',f' ?)
> ==
>
>
>
>
>
>
>
>
>
>
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Git for human beings" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to git-users+unsubscr...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [git-users] HELP git cant recognize ANY commands

2015-10-13 Thread Magnus Therning
On Tue, Oct 13, 2015 at 01:58:02PM +0300, Konstantin Khomoutov wrote:
> On Mon, 12 Oct 2015 23:10:52 +0200
> Magnus Therning  wrote:
> 
> > > 
> > > This is what I got...(Hope I did it right...). The git version is 
> > > 2.6.1.windows.1
> > 
> > All we see there is that you can't run `mount`. I suspect you'll have
> > more luck with an image of your attempt to actually run `git`.
> 
> My take on this is that your suggestion won't add anything: the
> contents of the PATH env. variable suggests the crucial bits of the
> GfW installation are not there, and hence the Git entry binary (script
> in GfW? -- I can't remember) simply won't be found.
>
> I have no idea why the environment in the OP's case looks the way it
> looks, though.
> 
> My take is that he did not install GfW "as is" but rather it came
> bundled with some other software package, and is just broken when the
> user tries to use it outside of that package.

So it's actually a MinGW issue, not an issue with the git installation.

I would never have seen that from the image the OP posted...

/M

-- 
Magnus Therning  OpenPGP: 0xAB4DFBA4 
email: mag...@therning.org   jabber: mag...@therning.org
twitter: magthe   http://therning.org/magnus

Finagle's Fifth Law:
Always draw your curves, then plot your readings.

-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


signature.asc
Description: PGP signature


Re: [git-users] git commit , git review, want to submit a patch not a new commit

2015-10-13 Thread Magnus Therning
On Tue, Oct 13, 2015 at 02:11:30AM -0700, nmh wrote:
> I cloned a git repository.
> I made some changes. I could see changes in git diff all okay.
> I did git commit -a and git review.
> 
> I got one commit id and could see my patch submitted in gerrit.
> 
> 
> 
> Now git status in my local directory says
> git status
> On branch bug/1777879
> nothing to commit, working directory clean
> 
> Now i worked on review comments made some changes to few files.
> 
> git status shows newly changed files.
> 
> I want that the patch i am going to submit should show the files from prev 
> commit and the
> changed files from this commit.
> 
> e.g a,b,c,d (Files in commit 1)
> d -> d' (in review rework)
> 
> i want my patch to consist of - a, b,c,d'
> 
> 
> so I did 
> 
> git commit --amend this time (Purpose is to submit a patch to earlier 
> commit- is it right to do it this way ?)
> git review 
> 
> 
> is this right ?
> 

This is a mailing list for git users... I'm sure there are quite a few
gerrit users here too, but if I were you I'd post this on some gerrit
mailing list instead. I suspect you'll have more luck with getting good
answers there.

/M

-- 
Magnus Therning  OpenPGP: 0xAB4DFBA4 
email: mag...@therning.org   jabber: mag...@therning.org
twitter: magthe   http://therning.org/magnus

As far as the laws of mathematics refer to reality, they are not
certain, and as far as they are certain, they do not refer to reality.
 -- Albert Einstein

-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


signature.asc
Description: PGP signature


Re: [git-users] An unbelievably newby question

2015-10-13 Thread Konstantin Khomoutov
On Mon, 12 Oct 2015 11:19:21 -0700
Kenoli Oleari  wrote:

> Thanks for these replies.  I will jump in and try them out.
> 
> For all the cheers about GIT, If find it incredibly difficult to use
> (or to even understand)
> 
> I appreciate the support.

I wonder if you have any prior experience with other version control
systems.  Git was my third VCS (after CVS and Subversion; oh, well, I
also used Visual Source Safe and RCS), and the first distributed one,
and I recall that at first I did have great trouble understanding how on
Earth I would just freaking create a repository to begin with!

Now I understand that the centralized model was so entrenched in my
gray matter that I simply fumbled around trying to find any HOWTO or
a blog post which would map my usual Subversion routine -- go to the
server, initialize a repo there, set it up, return back to the client,
check it out -- to that of Git, and there were literally none.  None!
And all that -- given the huge volume of material available online on
Git even back then when I started using it (something like Git 1.5.x).
I was deeply frustrated ;-)

My problem clearly was in applying certain mental patterns entrenched
in me after mastering two centralized VC systems to a system which used
a set of completely different paradigms.

That's why I'm asking.  If you have a centralized VC model controlling
your thinking about version control, -- that's one thing.  We can help
you unlearn that and learn the new ways.

If, conversely, Git is your first VCS that's a wholly different thing.
Then the problem is supposedly the complexity of the set of paradigms
you have to master to fully understand how a DVCS system, and Git,
works.  Once you know that, using Git won't be hard.  In other words,
the oft-cited claim Git UI is hard is mostly a red herring -- for
instance, try making yourself around, say, Mercurial or Fossil, without
reading 2-3 intros or HOWTOs, and you'll see that there's nothing
ingerently "self-evident" about their UIs as well.  As they say,
«the only truly self-evident user interface is the mother's nipple».
Anything else requires understanding the underlying paradigms.

If that's your case, I'd recommend to start with [1] to get the basics
of a DVCS conveyed in the simplest form possible.  Then some more deep
intro material might help.  Depends on your overall skill level in IT.

1. http://tom.preston-werner.com/2009/05/19/the-git-parable.html

-- 
You received this message because you are subscribed to the Google Groups "Git 
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to git-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.