Re: [git-users] Using a FUSE with git?

[Charles Manning]

On Thu, Feb 22, 2018 at 7:21 AM,  wrote:

> On Wednesday, February 21, 2018 11:19:52 AM Duane Knesek wrote:
> > Is that only for Windows?
> >
> > If it runs in Linux, then yes that is exactly what I'd need.
>
> I guess I'd divide that into two questions (assuming my understanding is
> correct)--my questions would be:
>
>* can the GVFS "service" run on Linux?
>
>* even if GVFS must run on Windows, can Linux clients use it?
>

> I'm not the OP, but if I was, it would be one thing to install one or a few
> Windows servers, vs. converting (presumably) many Linux users to Windows to
> use it as a client.
>

I can't see that being very successful. All your permissions and symlinks
are going to get hosed.

Won't Google/Android repo do what you want
https://source.android.com/setup/using-repo

-- 
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] Newb question: can some one explain to me why, when pushing Sass files, only the compiled .css file gets flagged as a merge error and not the .scss files?

[Charles Manning]

Git has no magic knowledge of scss files or css files. It will track
whatever you tell it to track.

So for example let's use something more people are familiar with  -
something like C.

If you tell git to track foo.c it will track foo.c.
If you tell it to  track foo.o it will track foo.o.

If you tell it to track both it will track both.

Same with your scss and css files.

Generally you don't want to track derived files (ie. stuff that is an
output from compiling) - though there are exceptions to this.

If you no longer want to track the css file than remove it from git with
git rm

You might also want to add it to your .gitignore file too.




On Wed, Jan 25, 2017 at 2:19 PM, AD S  wrote:

> So, if I make changes to a .scss file and compile it into a .css file and
> then push the entire project to a remote repo, there are 2 files that have
> been altered, right?
>
> I've noticed, however, that only the .css file appears with merge conflict
> notices.
>
> Would anyone know why this is? Could it be the way my company set up git?
> To automatically accept that my version of the .scss files will always
> override the remote repo's version?
>
> --
> 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] Tracking one file's diffs from then to now

[Charles Manning]

git log -p  foo.c

Also have a look at git blame

git blame foo.c

git blame cannot show deleted lines, but it can be really helpful trying to
see file history.


On Tue, Jan 24, 2017 at 1:47 PM, Michael  wrote:

> What command, or gui tool, will let me check one file's diffs over time?
>
> i.e.: Lets say I want to see what happened to one file, only, from commit
> X to the present.
>
> Then, as a GUI tool, being able to look through the list/tree of files,
> and see all the commits (and their locations in the commit tree) and
> patches to that file.
>
> Bonus points if, as recently discussed, I can specify a block of code that
> I want to see the diffs of over time, even if that block of code has moved
> from one file to another.
>
> (Yes, that is the goal here -- seeing how one routine has morphed over two
> developers and a good amount of time.)
>
>
> ---
> Entertaining minecraft videos
> http://YouTube.com/keybounce
>
> --
> 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] Noob question: In a large company setting, is it normal to spend 3+ hours a day sorting out git issues?

[Charles Manning]

On Thu, Jan 12, 2017 at 1:35 PM, AD S  wrote:

> Hi all,
>
> Sorry I really don't mean for this to sound pessimistic or whiny - it
> really is a genuine question.
>
> I come from a background of solo work where I didn't really use git all
> that much. Now I work with a very large company and git is a huge part of
> the workflow.
>
> However, I spend usually around 3hrs a day just trying to sort out strange
> issues I come up against with git. Everyone else here just seems to think
> this is par for the course, but it frustrates me a lot.
>
> Is this normal when working with large 100+ organisations?
>

Nope.

Problems like this generally result from people trying to be too clever and
making more problems than they solve.

A 100+ organisation should likely be working across multiple repos rather
than just one. You are more than likely working as, say 20 teams of 5
people.

Use lots of branching.


-- Charles

-- 
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] Could GIT help with this task?

[Charles Manning]

On Wed, Sep 7, 2016 at 10:19 AM,  wrote:

> Colleague:
>
> Before I decide to learn GIT I would like to know if it can partially
> handle a task which I will describe briefly.  I work with a large numerical
> weather prediction modeling system called WRF.  For sake of simplicity
> let’s suppose I start with version A which has been unmodified by me.  I
> make changes to A which makes version B.  Then comes along a new version C
> from the WRF distributer.  Would GIT help me take the changes I made in
> version B and incorporate them into version C to make a new version D?
> What would that operation be called in the GIT context?
>
> Many thanks,
> Kevin Doty
>

That's merging.

The way you'd do this is start with A, create a branch on which you do your
fiddling and end uopp with B.

Likewise do the same for a branch with the changes ending up with C. Then
merge the two branches and see if that works.

So it sounds like you're hoping git will magically do merges for you...
That might be wishing too much.

Git is pretty good at merging changes in source code, but cannot
effectively merge binary code. Even machine XML is more like binary code
than text and often gets messed up.

>
> --
> 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] Marking a branch "closed"?

[Charles Manning]

On Wed, Jul 20, 2016 at 6:06 PM, Michael <keybou...@gmail.com> wrote:

>
> On 2016-07-19, at 11:02 PM, Charles Manning <cdhmann...@gmail.com> wrote:
>
> Squashing makes sense if you have a really ratty bunch of checkins with
> work-in-progress checkins etc., but unless it's a trivial topic branch I
> would still typically make the final set of commits into a few logical
> steps.
>
> It costs pretty much nothing to leave old topic branches around (but a few
> thousand "nothings" can add up :-)).
>
> Once you've merged a topic branch you can safely delete the branch with no
> harm (apart from losing the branch itself).
>
> As Gergely says it depends on your workflow.
>
> For example let's say you are using some fault tracking database (eg.
> trac). It often makes sense to do the fix on a topic branch  (eg.
> fix-trac-1234). If you leave the branch in place after merging it you can
> then refer to the branch in the trac notes and see what fixes were used to
> fix the bug (and reopen the bug if it needs another kick).
>
>
> This actually is why I'd like to keep the old "messy work-in-progress"
> history, so I can see what had to be done should it be necessary to go
> back. I still would like some way to be able to see, at a glance, "This
> branch is marked as committed, so I don't have to worry that I forgot about
> it".
>
>
Well as you can see from this thread, there is no "one true way".

By "messy work in progress, I mean those little checkins along the way when
you're doing little experiments, deleting them and trying something else.

I will often do little WIP checkins so I don't lose say 15 minutes of work.
They are not useful commits in themselves,

It is generally worth cleaning those up so that you end up with fewer
"logical" commits without all the little mis-steps.
But why not just start off keep the closed topic branches. If you don't
like them you can always delete them later.

I would suggest giving them a fixed format so that they are easy to filter
out and delete/rename later,\


> ---
> Entertaining minecraft videos
> http://YouTube.com/keybounce
>
> --
> 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] Marking a branch "closed"?

[Charles Manning]

Squashing makes sense if you have a really ratty bunch of checkins with
work-in-progress checkins etc., but unless it's a trivial topic branch I
would still typically make the final set of commits into a few logical
steps.

It costs pretty much nothing to leave old topic branches around (but a few
thousand "nothings" can add up :-)).

Once you've merged a topic branch you can safely delete the branch with no
harm (apart from losing the branch itself).

As Gergely says it depends on your workflow.

For example let's say you are using some fault tracking database (eg.
trac). It often makes sense to do the fix on a topic branch  (eg.
fix-trac-1234). If you leave the branch in place after merging it you can
then refer to the branch in the trac notes and see what fixes were used to
fix the bug (and reopen the bug if it needs another kick).

-- Charles







On Wed, Jul 20, 2016 at 5:29 PM, Gergely Polonkai 
wrote:

> Hello,
>
> that, again, depends on your workflow. For one project, I
> interactive-rebase my feature branches often, and merge them into master
> when they are finished; after that, they can live forever in the repo, and
> if the feature needs a fix or change, I can check them out. For another
> project, I squash-commit the feature branch then delete it. In a third one
> I squash and don't delete. You decide :)
>
> Best,
> Gergely
>
> On Jul 20, 2016 05:55, "Michael"  wrote:
>
>> Lets say I've got a topic branch. I've made a bunch of commits. It's
>> messy. But it's done.
>>
>> As I understand it, best practice is to do a squash commit of the whole
>> thing onto the parent branch (develop or master, depending on workflow).
>> And I can do that.
>>
>> What do I do with the leftover? I thought I could tag it as "closed", but
>> I can't use the same tag more than once. What's the best way to mark it as
>> done, or should I just delete the branch label and let it be garbage
>> collected/deleted?
>>
>> ---
>> Entertaining minecraft videos
>> http://YouTube.com/keybounce
>>
>> --
>> 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.
>

-- 
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] GIT and Hosting Remote Repositories

[Charles Manning]

On Fri, Jul 8, 2016 at 5:18 AM, JaedenRuiner 
wrote:

> I am aware of the convention and the nature of local repositories.
> That's not my problem.
>
> Git repositories can have a "remote repository" by means of the
> origin/[branch] aspect.  This is the "remote" tracked branches that are
> used in Upstreams.
>

You can have multiple remotes if you want.


>
> I want to know how I can define an upstream repository on my local system,
> a network system, or by way of a cloud service.  If it is local it CANNOT
> be Linux.  It must be a windows host.  If I had the capability of doing a
> linux box or VM I would already be using GitLab, because i know it.  But
> since I can't do linux, I need a windows or cloud solution for this.
>

I avoid Windows as much as possible, so I can't help you there.

There is also Cygwin which runs on Windows and has git.


>
> And trust me, I've already searched a hundred times for what i'm looking
> for and I keep getting routed to GitLab or GitHub.  I need a different
> solution.
> Now i don't know what API GitHub uses, and it appears you can do a single
> project, but I want a remote repository for a 100 projects.  They have a
> "personal" that seems to allow unlimited private projects for $7 a month,
> but I don't know if they have an API, a way to automate my stuff via other
> interfaces than a web browser, and I'm not going to pay for anything until
> I can verify it will do what I want.
>

I'm confused by what you mean by an API and what you expect this API to do.

If the server supports basic git protocols (as github or any other git
server/service does) then you can drive it from any git client (eg. command
line, tortoise-git,...)


>
> Thanks
> J"SD"a'RR
>
> On Thursday, July 7, 2016 at 12:00:21 AM UTC-5, charlesmanning wrote:
>>
>> The thing about systems like git is that there is no "central repository"
>> except by convention.
>>
>> That means you can host a local repository on a local server as well as
>> storing it in github or some other cloud service. No one repository is more
>> "central" than the other.
>>
>> "git server for linux" on google will give you many hits.
>>
>>
>>
>> On Thu, Jul 7, 2016 at 4:38 PM, JaedenRuiner 
>> wrote:
>>
>>> So,
>>>
>>> Simple question.  I do my own development, a lot.  Now, I've been using
>>> GIT at work and figured it was about time to set up my own version control
>>> on my personal stuff.
>>> I've installed GIT and it is all ready to go, but I currently have only
>>> "local" repositories.  There is no, "remote" to compare to.  At work we use
>>> GitLab.  This seemed the easy route for me to go, since I already know
>>> both, but I can't seem to use their own API against their website.  It
>>> seems their website may only be a temporary testing ground and not the full
>>> product, which has to be run on a Linux server and Apache.  I do not have
>>> one, nor can I get to one easily, (nor do I want to deal with the hassle).
>>>
>>> So, since GitLab is out, I wanted to know if there is any "software" or
>>> other such Git setup that will allow me to host a central "remote
>>> repository"?  I don't want to deal with only the local repositories,
>>> regardless of the powerful branching system, I want a central repository of
>>> everything, so I can manage my current production versions.  I am perfectly
>>> okay with using a cloud based system of a self hosting concept is not
>>> available, as long as it doesn't cost too much (free would be preferable),
>>> and there is a grace period for validating it can do what I need.  I need
>>> the ability to manage nearly 100 projects, at less than a gig of storage
>>> space necessary.
>>>
>>> Any Ideas?
>>>
>>> Thanks
>>> Jaeden "Sifo Dyas" al'Raec Ruiner
>>>
>>> --
>>> 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+...@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.
>

-- 
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] GIT and Hosting Remote Repositories

[Charles Manning]

The thing about systems like git is that there is no "central repository"
except by convention.

That means you can host a local repository on a local server as well as
storing it in github or some other cloud service. No one repository is more
"central" than the other.

"git server for linux" on google will give you many hits.



On Thu, Jul 7, 2016 at 4:38 PM, JaedenRuiner 
wrote:

> So,
>
> Simple question.  I do my own development, a lot.  Now, I've been using
> GIT at work and figured it was about time to set up my own version control
> on my personal stuff.
> I've installed GIT and it is all ready to go, but I currently have only
> "local" repositories.  There is no, "remote" to compare to.  At work we use
> GitLab.  This seemed the easy route for me to go, since I already know
> both, but I can't seem to use their own API against their website.  It
> seems their website may only be a temporary testing ground and not the full
> product, which has to be run on a Linux server and Apache.  I do not have
> one, nor can I get to one easily, (nor do I want to deal with the hassle).
>
> So, since GitLab is out, I wanted to know if there is any "software" or
> other such Git setup that will allow me to host a central "remote
> repository"?  I don't want to deal with only the local repositories,
> regardless of the powerful branching system, I want a central repository of
> everything, so I can manage my current production versions.  I am perfectly
> okay with using a cloud based system of a self hosting concept is not
> available, as long as it doesn't cost too much (free would be preferable),
> and there is a grace period for validating it can do what I need.  I need
> the ability to manage nearly 100 projects, at less than a gig of storage
> space necessary.
>
> Any Ideas?
>
> Thanks
> Jaeden "Sifo Dyas" al'Raec Ruiner
>
> --
> 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] Multiple backup strategy & S3

[Charles Manning]

On Thu, Jan 7, 2016 at 2:51 PM, Kai Hendry  wrote:

> Hi guys,
>
> Like many of you I'm putty more and more precious things like photos into
> git for safekeeping.
>

I've never heard of anyone using git for storing photos.It's like using a
wrench as a hammer: it does the job, but that's not what it is designed for.



> However the problem I have is how to sync it to a couple of remotes in a
> straightforward manner.
>

Why not just use rsync?


>
> Also I would like to "fsck" the remotes to make sure the remotes are not
> somehow corrupted.
>
> Ideally I would like to backup straight to S3 but this is a little clumsy
> with http://dabase.com/e/14002/
> * I don't want to run commands manually
> * I don't want to trump what's there in case my source is corrupted
>
> So I guess I am looking for more end points to push my git repo to. They
> can be public. Any suggestions? Please let me know.
>
> I'm aware of https://git-annex.branchable.com/ but I don't like how it
> mangles my filenames and such.
>
> Kind regards,
>
> --
> 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] Fairly new to GIT questions

[Charles Manning]

On Thu, Dec 24, 2015 at 7:30 AM, Konstantin Khomoutov <
flatw...@users.sourceforge.net> wrote:

> On Wed, 23 Dec 2015 10:03:49 -0800 (PST)
> Lou Elston  wrote:
>
> > I am a Windows user.  I do not know BASH, so I am using GIT-CMD, and
> > am working my way through the LinkedIn\Lynda, GIT Beginners course.
> > Why (it does not happen every single time), do I have to do a Cntrl\C
> > to get the prompt back (it appears to be hanging\waiting for
> > something)?
>

I don't know what this git command is. Why not just load up Cygwin? That
has full bash and git support.

>
> A guess: you run some command, it outputs something to you and then
> shows the ':' character in the lower left corner of your console window
> (usually with a blinking cursor right beside it as well).
> Is this the case?
>
> If yes, this is the so-called "pager" presenting you the output of the
> Git command you have run, and unless you took special measures to
> override this, the pager Git uses by default is named "less" and is
> bundled with Git for Windows.  Git tries to be smart so, again, unless
> overridden or told not to in a particular command invocation, it sees
> if the prospective output will fit into whatever number of text lines
> your console window shows, and if it sees the output will be larger
> than that it spawns the pages automatically and pipes the output to it.
> That explains why you see the pager when doing `git log --patch`
> but don't see it when doing `git log -2`.
>
> The "less" program [1] allows you to scroll up and down (cursor keys,
> page up/page down keys), search through whatever it displays ('/' or '?'),
> place anchors (and jump to them) etc.  To quit from it just press 'q'
> (for "quit").  To get built-in help, type 'h' ('q' exits from this mode).
>
> Google for less+pager+manual if you want to know more.
> You can start with [1].
>
> > Also, is there any kind of test, exam, or certification for GIT?
> > Something to show that I know the basics, that I can show, and\or use
> > to get ready for an interview?


I don't know about the people you deal with, but I've been involved in
hiring dozens of people. I never
really look at any certificates etc.

IMHO the best approach is to use git for your work as much as you can. Get
yourself a free account on github
and store files there.

git alone is unlikely to get you a job, so do some programming and store
the files there. You can then steer the interviewer to your github
repositories to show examples of your work as well as proving you've used
git for your projects.

-- 
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] religions

[Charles Manning]

I think he's asking about

git blame

On Mon, Dec 21, 2015 at 10:49 PM, Martin Møller Skarbiniks Pedersen <
traxpla...@gmail.com> wrote:

>
>
> On 19 December 2015 at 20:29, wayne capper  wrote:
>
> all individuals worldwide shold legally have a right to there own beliefs
>> without prosecution or war...
>>
>
> Do you have a question about git ?
>
>
> --
> 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] how to git clone/pull/push another PC without entering password ?

[Charles Manning]

It all comes down to what protocol you're using to push.

If you have them attached via nfs, then you can just push with no problems.

Since you talk about ssh-agent I expect you're using the git protocol with
ssh.

What you need to do is set up the remote device to accept passwordless ssh
logins, You do that by using ssh-copy-id to get the remote machine to
recognise your key.


Notes from something a friend sent me:


crash course in ssh use:

You should generate an ssh key on each
computer you use and upload the public half to each computer you want
to talk to. Copying the same key about means that compromising one
machines compromises them all, and neither you nor anyone else can
control access on a per-machine basis.

Gernating a new key is dead easy:
ssh-keygen

and uploading the public half is similarly easy:
ssh-copy-id machine  (often need to give -i .ssh/id_dsa.pub if ssh-add
hasn't been run yet)

ssh keys ought to have a passphrase in them (so that stealing your
computer is not sufficient to give access everywhere). But that can be
annoying unless you use ssh-agent. Your distro almost certainly runs
this for you as part of your X-session so it's actually very eay to
use. You do ssh-add and it stays added until the session logs out. But
doing this is dull, so the clever thing is to add this to your
.bashrc:
alias ssh='ssh-add -l > /dev/null 2>&1 || ssh-add ; \ssh'
that adds it automatically the first time you connect to somewhere so
you only ever have to type your passphrase once.



On Mon, Oct 19, 2015 at 8:51 AM, liu wen  wrote:

> I'm using a fedora and ubuntu. Ubuntu as the server.
>
> On Sun, Oct 18, 2015 at 6:38 PM, Konstantin Khomoutov <
> flatw...@users.sourceforge.net> wrote:
>
>> On Sat, 17 Oct 2015 02:45:28 -0700 (PDT)
>> liu wen  wrote:
>>
>> > I have two laptop A and B, I want to create a shared repository on A,
>> > and I can clone/pull/push the shared repo on B.
>> > is it possible? if so, what are the clone command like?
>> >
>> > and do I need to enter password again and again? can I avoid entering
>> > password( like ssh-agent)?
>>
>> Too little information to provide you with a constructive answer.
>>
>> To begin with, what OSes those laptops are running?
>>
>> If it's Windows, you will be able to interoperate with A from B using
>> nothing more than a conventional Windows share created to host the
>> directory containing the repository.
>> Password management in this case depends on what class this Windows
>> network is (domain vs single-rank).
>>
>> If it's not Windows (or Windows and something else), things should
>> probably be set up differently, and password management would depend on
>> the setup selected.
>>
>
> --
> 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] Looking for a change to a file

[Charles Manning]

When you're trying to find why a specific change was made, then git blame
is also your friend.

On Mon, Sep 14, 2015 at 7:16 AM, Paul Smith  wrote:

> On Sun, 2015-09-13 at 11:32 -0700, Michael wrote:
> > I've got a file, with a diff/change, that I thought had already been
> > checked into a previous commit. I want to see if this change exists in
> > any checkin of that file.
> >
> > Is there any way to ask for "all the checked-in versions of file X"?
> > As far as I know, git only knows files by their sha hashes, and not by
> > names -- which already tells me that there's probably something that I
> > really don't understand about git.
>
> Git doesn't know files by SHA hashes.  Git knows "a set of changes to
> one or more files" by its SHA hash.  The SHA hash names the _commit_ not
> a specific file.
>
> However, if you want to see the commits that impacted a given file it's
> trivial:
>
>   git log 
>
> If you want to see the commits that impacted any file in a subdirectory
> that's trivial as well:
>
>   git log 
>
>
> --
> 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] Why git doesn't track file's read/write permission?

[Charles Manning]

On Thu, Aug 27, 2015 at 12:40 PM, Enzo Chi enzo.chi...@gmail.com wrote:

 I think git track executable permission, right?

 If so, files in git is read only and not deployment too, why it track
 x permission?


x permissions alter the functionality of the file.

For example, if you have a build system that needs scripts to build and
those scripts need to be x, then the x better be stored or the scripts are
useless.



 Keep read only permission is useful in some scenario. And most important
 thing is there's on harm to keep it (I am not a software developer, correct
 me if I am wrong)?


That does not achieve anything from a safety perspective. If you lose/alter
a file then you just checkout the version in HEAD again.

Getting r/w permissions right would be hell anyway.

Consider this:

Start with a readonly file.

Now I need to change it, so I make is locally writable and change it.
Then I do a commit and push it again. Oops! I forgot to make it read only
again before I pushed it, so now it is writable in the repo too.

Better to just dodge this hangover by not having r/w.






 On Thursday, August 27, 2015 at 6:55:55 AM UTC+10, Philip Oakley wrote:


 I have post an question at
 http://superuser.com/questions/962861/how-to-use-git-to-commit-read-only-file


 I just want to know why GIT doesn't track read/write permission?

 What I want is just GIT keep what every I checked in? ( I am OK with the
 executable permission control)


 It's sort of a philosophical issue. If you are placing a file into a
 repository, it is by definition read only. You can never 'write' the same
 revision, but with a different content - it would be a contradiction. Hence
 the r/w flags are ignored.

 It's important to remember that as concieved, Git is not a deployment
 tool, so it didn't need r/w permissions, and as open source DVCS,
 everything checked out would be local so the user would have full control,
 so read-only couldn't be relied on anyway, and we hope the user will
 contribute a change/improvement so 'write' it is!

 Likewise it doesn't store timestamps (of the files) either..

 There is a Linus 'rant' somewhere on the issue..

 Philip

 --
 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] Workflow With QA Team

[Charles Manning]

You might consider something that wraps git up in extra QA/management
layers. For example, have a look at gerrit which is used by Google, amongst
others.

https://code.google.com/p/gerrit/



On Sat, Sep 6, 2014 at 5:28 AM, Eric Kolotyluk eric.koloty...@gmail.com
wrote:

  What sort of workflow do people use with their QA teams?

 For example, when a feature branch is ready for testing, do people do a
 pull request to master, development, or to a special test/*branch*? When
 QA have confirmed the feature is good, do they then do a pull request to
 master, release, or somewhere else.

 Does anyone know of any good documentation that talks about git workflows
 incorporating a test team?

 Cheers, Eric

 --
 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] Looking for Suggestions

[Charles Manning]

 There is an argument whether Git is actually easier to learn than SVN for
 people who have never used version control systems before. After all, Git
 has more in common with the kind of version control system you would invent
 by yourself if there were none available [1].

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


By that argument, git should have emerged as the first version control
system. Instead we got rcs!

SVN is surely simpler to learn - there are less concepts: just a client
checkout and a repository. No staging and no real branching.

Git though is not that much harder to learn. It has far more flexibility
than svn. For that reason it is worth the extra effort.

-- 
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] Looking for Suggestions

[Charles Manning]

Git is certainly a bit harder to learn than SVN, but it is so much more
flexible.

git's strength is that it works seamlessly as a single-user system and and
a multi-user + server system.

I use git to track changes during development even if I never move it from
the original PC.

Git has the added benefit that all clones are equal. If you lose an SVN
server you're screwed. With git you can just reclone it from a developer
machine.

Git is a far better solution in the long term.



On Thu, Jun 12, 2014 at 3:03 AM, Dale R. Worley wor...@alum.mit.edu wrote:

  From: HOWARD ROSENBLOOM hrsn...@gmail.com
 
  We are a small company that has not been using any form of
  configuration/tracking in the past. We have 2-3 programmers and have
 gotten
  into the situation where we need to get control of any work being done.
  Since none of us have any experience in this area I am looking for
  suggestions as to whether GIT is right for what we are looking for, or is
  there a better/more appropriate solution.

 I've used both Git and Subversion.  If you're a small operation, and
 you've not used version control before, and if you do not need the
 distributed development facilities of Git, I believe that you'll find
 Subversion easier to learn and use.  The drawback of Subversion is
 that you need to have a server that is accessible by each worker for
 all check-ins and check-outs.

 Dale

 --
 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] diff most recent commit

[Charles Manning]

git log -p filename



On Fri, May 30, 2014 at 4:54 PM, dexter ietf dexter.i...@gmail.com wrote:

 hi,

 how to see the diff of a file with it's most recent commit.
 also show to take the diff of file:lineno with its most
 recent commit on the line. note that the recent commit for
 a file and file:lineno might be different.

 -thanks

  --
 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] What is the best way to visualize branch history line?

[Charles Manning]

gitk has some useful options, so make sure you read the man pages.

Of most interest is

gitk --all

That is really handy for seeing the branches relative to eachother.



On Wed, May 14, 2014 at 9:23 AM, Leonardo Petry leonardo.petry...@gmail.com
 wrote:

 I tried gitk today and got blown away that I completely missed this. It
 helps me understand the current state of my local branch in a way I
 couldn't before. Many Thanks.


 On Monday, May 12, 2014 10:57:06 PM UTC-4, charlesmanning wrote:

 I like gitk.


 On Tue, May 13, 2014 at 2:20 PM, Leonardo Petry leonardo...@gmail.comwrote:

 What would be the best command to view a branch history? Right now I use 
 *git
 show-branch* but I get a huge list of commits and branches. Ideally I
 would like to see a couple of previous commits and be able to see the which
 files where added to each commit.


 On another topic I would appreciate any tips on when I should prefer
 merge over rebase.

 --
 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+...@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.


-- 
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] What is the best way to visualize branch history line?

[Charles Manning]

I like gitk.


On Tue, May 13, 2014 at 2:20 PM, Leonardo Petry leonardo.petry...@gmail.com
 wrote:

 What would be the best command to view a branch history? Right now I use *git
 show-branch* but I get a huge list of commits and branches. Ideally I
 would like to see a couple of previous commits and be able to see the which
 files where added to each commit.


 On another topic I would appreciate any tips on when I should prefer merge
 over rebase.

 --
 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] How to refresh my local repo copy.

[Charles Manning]

You want git pull

git fetch will fetch the files
git merge will then merge in the changes

git pull does it in one step.




On Fri, Apr 18, 2014 at 8:21 AM, rhugga chuck.car...@gmail.com wrote:


 I'm on a team of about 15 people and we are using a central server to
 store our master repo.

 Each user then pulls down (git clone) a copy of the master repo to where
 ever they happen to be working at the time.

 Is there an easy way for someone to refresh their local copy, so they
 pull down any files someone else has committed? The only way I can see to
 do this is commit and push your local changes, delete your local copy and
 pull down again using clone.

 Thx.


  --
 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] GSoC'14: confuced about submitting the .diff file

[Charles Manning]

You should not generate .diff files because it is hard to know the context
and that throws away some useful info.

It is better to use git format-patch which sets up the diff as a proper
patch, with the commit message and everything. That is far more useful to
people trying to understand your patch.

You can then email it to the list you are interested in. The best way to do
this is with:

git send-email




On Mon, Mar 10, 2014 at 11:24 AM, Nemina Amarasinghe nemi...@gmail.comwrote:

 Hi,

 My name is Nemina Amarasinghe, I am a third year computer engineering
 student form university of Peradeniya Sri Lanka. I would like to
 participate to GSoC'14 under the GIT organization.
 As the Gsoc page mentioned currently I have cloned and build the git
 source code. And now I'm working with one of micro-projects you have
 mentioned.


 *I want to submit the .diff file I have created. Should I attach it in a
 email or should I have to submit it somewhere else. *
 if anyone could give me a answer deeply appreciated

 Best regards,
 Nemina Amarasinghe

 --
 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] How to list branches

[Charles Manning]

Try typing man git branch.

You will see that existing branches are shown. That means branches that
exist in your repository.

Once you check out versionx it will become a local branch.



On Sat, Oct 19, 2013 at 1:51 PM, Blake McBride blake1...@gmail.com wrote:

 Greetings,

 I am having trouble getting up to speed with git.  Your help is greatly
 appreciated.  For example:

 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git pull
 Already up-to-date.
 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git branch
 * master
 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git checkout Version_2_6_10pre
 Branch Version_2_6_10pre set up to track remote branch Version_2_6_10pre
 from origin.
 Switched to a new branch 'Version_2_6_10pre'
 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git branch
 * Version_2_6_10pre
   master
 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git checkout master
 Switched to branch 'master'
 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git branch
   Version_2_6_10pre
 * master
 blake@vm-mint-14 ~/Backup/gcl/gcl.git $ git --version
 git version 1.7.10.4


 The first time I did a 'git branch' it only showed master.  Then, when I
 switch to a branch that I magically know about, git then knows about the
 branch.  Is there some sort of logic to this?  I don't get git because it
 defies any kind of understanding I try to apply to it.

 Your help is greatly appreciated.

 Blake McBride

  --
 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/groups/opt_out.


-- 
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/groups/opt_out.

Re: [git-users] How to list branches

[Charles Manning]

I suggest you work through the example again using something like gitk.
That will show you what branches there are and show you the difference
between the local and remote branches of the same name. Using gitk between
operations is highly instructive as you learn to use git.


Forget all you know about SVN branches. With git you want a completely
different mindset. Under git branches are free and yout should (almost)
always do your development in a branch then merge it.




On Sat, Oct 19, 2013 at 4:05 PM, Felipe Contreras 
felipe.contre...@gmail.com wrote:

 On Fri, Oct 18, 2013 at 9:57 PM, Blake McBride blake1...@gmail.com
 wrote:
  Thank you very much for the help!  I have that book.  I think I'll some
  reading.  My mind is so SVN oriented that when I read the books I keep
  thinking 'but how would I do x?  y?  Z?'  Perhaps x, y, and z don't make
  sense with this new model.

 In my opinion Subversion and Git are so different it's not helpful to
 think on those terms. Yes, some commands are equivalent, but many
 others seem similar but do drastically different things.

  From what you are saying, I gather that branches created remotely are
 kept
  noted as such (even though I have a local copy).  When I check the branch
  out, it becomes a totally new local branch.

 Not exactly. 'git checkout origin/foo' is how you checkout a remote
 branch, but that sends to a kind of limbo called a detached HEAD,
 which means you are in no branch, you are in a commit, and more often
 than not that's not what you want. 'git checkout foo' will checkout
 the local branch 'foo', if there's no local branch with that name it
 finds out if there's a remote branch with that name that would make
 sense to checkout, and will replace that command with 'git checkout -b
 foo origin/foo' (create a new branch 'foo' from the starting point
 'origin/foo')

 --
 Felipe Contreras

 --
 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/groups/opt_out.


-- 
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/groups/opt_out.

Re: [git-users] Re: git.for solo developer

[Charles Manning]

On Thu, Dec 20, 2012 at 9:38 AM, Thomas Ferris Nicolaisen
tfn...@gmail.com wrote:
 On Wednesday, December 19, 2012 9:30:24 PM UTC+1, Dimitris Papageorgiou
 wrote:

 Do you think git would be useful for a solo developer as I am right now?

 I am developing a web app alone-no other developers.

 VCS is always useful kind of aid for a developer.

 But I think Git is intended more for teams of developers that wotk on the
 same project
 so that they can take advantage of Git's distributed functionality.

 I am asking these things because I am complete beginner in Git and in
 general to version control systems


 Git works awesomely for solo projects. Just git init and you're up and
 running. Here's a starter's guide: http://rogerdudler.github.com/git-guide/


I absolutely agree with this.

git is great for solo developers, particularly if you hook up to
github or equivalent. That way losing a disk or having equipment
stolen is not so dramatic.

Even just having the ability to compare between revisions and manage
different branches of development makes git very powerful.


When my teenage kids started programming, the first thing I did with
them was get them to sign up to github and get into the habit of using
git.

Nothing in programming is more tragic than having something work, then
make a few benign changes which screws everything up. git helps with
that

-- CHarles

-- 

Re: [git-users] Re: Why do I have to be a source control engineer just to be a software developer?

[Charles Manning]

Ok, I like git now that I can more or less function, but I agree with
the original poster git is a huge learning curve even with reading
a heap of books and reading all the tutorials.

git could really use some training wheels to make it easier to get going.



On Tue, Oct 30, 2012 at 1:43 PM, Alexandru Pătrănescu
dreal...@gmail.com wrote:
 Git is very different than other VCS (like SVN or whatever you used before).
 I suggest reading the Pro Git book with a clear mind, open to new ideas, not
 linking what you read with what you know that command did in other VCS.
 Git is really not that complicated, as his creator described it, it is a
 stupid content tracker. Please prove yourself smart, and show us that you
 are worth of the software developer attribute.

 I'll answer you only the last question:
 If you will learn how to use git, you will be a better software developer.
 Even if Git is mainly an VCS, it can do more than that, changing the way you
 develop software, letting you do stuff you wouldn't think is possible.
 The rest of the questions you will answer yourself in time...






 On Tuesday, October 30, 2012 12:31:13 AM UTC+2, kramer.newsreader wrote:

 I am a fairly experienced developer and I have never had issues working
 with source control tools before git.

 I take a new job.  I am working with git.  I am thinking about quitting
 over having to use it.

 Every source control tool I have used before has an easy command that
 says: Use these changes right here.  Yes there are conflicts, but these are
 correct.

 How can I get to Not currently on any branch when I was on a branch and
 didn't ask to switch branches?

 Where is that with git?

 Why is the git information model SO COMPLICATED?

 Why can't there be a way to

 Why is the documentation so inadequate?

 Why do I have to be a source control engineer just to be a software
 developer?

 --
 You received this message because you are subscribed to the Google Groups
 Git for human beings group.
 To view this discussion on the web visit
 https://groups.google.com/d/msg/git-users/-/u6mtA5lJd_kJ.

 To post to this group, send email to git-users@googlegroups.com.
 To unsubscribe from this group, send email to
 git-users+unsubscr...@googlegroups.com.
 For more options, visit this group at
 http://groups.google.com/group/git-users?hl=en.

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-users@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Is Git for me?

[Charles Manning]

I think git is great for backup. Here's why:

With svn etc there is a central server. If that fails, then you are screwed.

With git, all the clones have exactly the same data, including history
etc. There is no master except by convention. If the master server
dies, then just repopulate it from a client and you have everything
restored.

I use git all the time even if the code does not leave one machine. It
allows me to track changes and experiment and easily give my clients
patches.



On Sun, Oct 7, 2012 at 5:38 PM, Max Hodges m...@whiterabbitpress.com wrote:
 Git is for version tracking--most often for code, but it could also be used
 to track any files. Its not a backup tool and its not a deployment tool.
 Tracking changes to your source code is a very useful function. Sometimes
 during development you realize you're taken a wrong turn, and a tool like
 Git allows you to backtrack very easily. Also sometimes you may comment out
 a lot of lines while you're refactoring some code. With Git you no longer
 need to comment them out. Just delete them, because you can always compare
 your new code with an older version. So this makes your code a bit more
 clean.

 I'd recommend Git any software development project. Source code version
 tracking is as fundamental as testing. Its the professional way to develop.

 If you don't care about the benefits of source code tracking, and are only
 looking for a backup solution, then there are more simple ways to
 implementing a backup, including a simply copy command to an external hard
 drive or USB thumb drive.

 As far as setting up your code for tracking, its extremely easy. You can use
 a GUI to manage things--I use the SmartGit GUI, but the command-line
 tutorials like this one are still very useful to get familiar with the
 concepts and capabilities:
 http://gitimmersion.com/

 airborne IR cameras huh? for spotting the tell-tale footprints of grow light
 marijuana production by chance?

 Cheers!

 --
 You received this message because you are subscribed to the Google Groups
 Git for human beings group.
 To post to this group, send email to git-users@googlegroups.com.
 To unsubscribe from this group, send email to
 git-users+unsubscr...@googlegroups.com.
 For more options, visit this group at
 http://groups.google.com/group/git-users?hl=en.

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-users@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Any writers out there?

[Charles Manning]

Why not commit every time you save?

Anyone who worked in the 1980s using VAX would tell you how great it was that 
thefile system kept versions of the file. Much, much better than the single 
file version we have now.

There is absolutely nothing wrong with doing work in progress commits after 
every save,

That will make a hell of a history though so you probably want to clean up 
whenever you get to a milestone or before you push. That can be achieved with 

git reset -- soft 25252 - last commit that you actually want to keep
git commit



On Wednesday 09 February 2011 09:06:33 Anas Mughal wrote:
 You definitely do not want a commit every time you save your file. Just
 commit when you reach a milestone, etc.

 Also, have you looked into using a online document hosting service like
 GoogleDocs. I presume you need to use advanced editing features that are
 available part of Word or some other thick client tool.

 Best Regards.

 On Mon, Feb 7, 2011 at 6:16 PM, Mark (my words) elib...@gmail.com wrote:
  Thanks for all the advice. I’ll apologize upfront for not crediting each
  of you for your individual contributions in my response—I’m feeling a bit
  overwhelmed.
 
  I just made a repository of 23 directories of poems, and yeah, it’s
  unwieldy. Thanks for the info on multi-repository tools that will come in
  handy.
 
  Too many commits do get out of hand quickly, I should save commits for
  those breakthrough moments in a project, not when I just change a
  comma—unless it’s an extremely important comma, in which case it better
  be commented.
 
  It just hit me, in my current workflow file names serve as comments I’m
  working on a file called /ladybug/ladybug new 3 past tense.markdown: that
  shows me the branch, version, and the major change: The third version of
  a new branch in which I move to the past tense.
 
  *Git compares lines:* I just diff’ed versions of a poem and wondered why
  it appeared I had deleted a block and replaced it with an identical
  block.
 
  I appreciate the git-show code, but I’m looking for a batch operation,
  but that’s best addressed in another post which I’ll make here in a few
  minutes.
 
  I’m moving toward a stripped down all text workflow using minimal
  text-editors, open/libre/neo-office is a bit…full-featured…for my sanity.
 
  I need to rethink my organization structure. My short-stories are
  arranged by series, but my poems are scattered about under whichever
  parent directory grabs my attention at the time—they’re in this order of
  priority, level of surrealism, emotional content right now and I find
  them by openmeta tags (which, yeah, can be convenient until you start
  using git).
 
  *Don’t have chaos:* I like that. I’m happy I decided to address my issues
  upfront instead of waiting until I went into production with it, these
  are some amazing insights you guys have, it saves me from walking blindly
  into a mine-field! I like the idea of a local master with local
  individual repos. It’s gonna take me a while to find a good balance to
  fit my needs.
 
  Wow, this has been the most fun I have ever had in a tech forum. You guys
  should form a comic troupe.
 
  Thanks for the wonderfully detailed and insightful comments,
 
  Mark


-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-users@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Re: get rid of old commits

[Charles Manning]

You don't need 3 years of commits until you do!

If you live on the wrong end of a really thin pipe you can use 

git clone --depth 10

or something to limit things to recent history.


On Tuesday 28 September 2010 22:19:06 canna wrote:
 didn't anybody deal with this problem?
 I really don't understand how can it be, Git exist for a long time
 now,
 nobody needed to archive old commits (and eliminate them from the
 repository) to reduce the size of the repository and get rid of
 ancient history?
 do all Git users just push and push commits forever and ever?
 I mean who needs 3 years of changes on your hard disk? why wont you
 want to backup part of it and leave only the recent changes on your
 hard drive?

 On Sep 20, 5:57 pm, canna c.ne...@gmail.com wrote:
  Hello Everybody!
 
  I hope someone can help me with this because I didn't find any
  information on that subject in the internet, and I'm struggling with
  this for a whole two days now
 
  we're using git for a year now and the repository grow very big,
  the main remote repository is located on a local network computer and
  all the developers push and pull from it
 
  the problem is, it's taking a lot of time for simple everyday
  operations like pull, push, show log (TortoiseGit), check for
  modifications (TortoiseGit) - to complete. also all the old commits
  are irrelevant (there is no way in the world we could ever revert to
  those commits)
 
  is there a way to somehow cut off half of the repository, meaning get
  rid of the old commits, for instance throw away all the first half of
  year of commits in order to make the repository lighter and easier to
  manage?
 
  Scott Chacon (http://git-scm.com/) recommended his post: Replace
  Kicker (http://progit.org/2010/03/17/replace.html).  I like the idea
  of pushing all history (old commits) to a separate (history)
  repository while removing the same history from the main repository.
  unfortunately his tutorial doesn't explain how to synchronize this
  change with the main remote repository and with all other local
  branches.
 
  we are around 5 developers, so it's still reasonable to go over all
  the repositories of everybody and rebase every brunch in the system
  manually if that what it's take to make it work...
 
  Thank you for any help on that subject!
  Netta


-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Git clone resume using git fetch ??

[Charles Manning]

On Tuesday 03 August 2010 22:25:25 Flak37 wrote:
 A large clone was interrupted at around 10%, and as a result I have
 122 mb of unfinished data left on my system.
 Is it possible to resume it using git-fetch or git pull ???

IIRC when this has happened to me just doing git clone again resumed from 
where things were.

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Multiple Projects Within a Project

[Charles Manning]

On Friday 06 August 2010 01:03:09 DAZ wrote:
 Hi,

 I was wondering if somebody could point me in the right direction
 here:

 I am working on a Sinatra base web framework and am using git for
 source control and also putting it up on github.

 The framework also contains a CSS framework. I would actually like the
 CSS framework to be in a separate github repository so they can be
 developed separately, but I'd also like it if I could work on the CSS
 within the Sinatra framework and for those changes to take effect in
 it's own repository.

 I am also using a one page CSS project that I have hosted on github

 I hope what I am saying makes sense. I have been reading the git docs
 and can't figure out if I should clone the CSS repository into the
 Sinatra repository or if I need to use submodules.

 Any advice on the best way to do this would be much appreciated,
 thanks


It would seem that git submodules would do what you want.

http://book.git-scm.com/5_submodules.html


-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Re: Pissed at Git!

[Charles Manning]

On Wednesday 02 June 2010 10:40:53 Konstantin Khomoutov wrote:

 Also, my idea was this: I found myself in situations like this one
 being discussed, and my first temptation after finding the relevant
 top commit in the reflog always was tag that immediately, then cool
 down, analyse the situation and think, that is, just give that
 regained precious commit a name so it's now referenced.

This is always a good policy while you still have training wheels on.

Before you do anything too exciting put some sort of tag in place. Then if you 
get lost it is easier to find where you were before your adventure.

You can do the same for rebasing, merging etc.

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

[git-users] squashing

[Charles Manning]

If you're old enough you might remember VAX VMS. One really neat feature that 
this had was file versioning that made it really easy to revert a few edits 
back.

I was trying to something like this in git storing each file after an edit 
with

$ git commit -a -m wip $(date)

After a few edits you end up with a log that looks like:

$ git log --oneline
d7f2462 wip Mon May  3 11:40:09 NZST 2010
a744bf9 wip Mon May  3 11:40:09 NZST 2010
cd27722 wip Mon May  3 11:40:08 NZST 2010
4130b3b wip Mon May  3 11:40:08 NZST 2010
3ab5dc6 wip Mon May  3 11:40:06 NZST 2010
0f0a662 wip Mon May  3 11:39:42 NZST 2010
25669f0 Add starting point file x

I can then do a git rebase -i and squash all the wips

and end up with

$ git log --oneline
fd041b4 wip Mon May  3 11:39:42 NZST 2010
25669f0 Add starting point file x

Then do a git commit --amend and change the commit message.

$ git log --oneline
3a770a7 finally fixed
25669f0 Add starting point file x


That's all very manual.

Is there some easier way?

Thanks

Charles


-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

Re: [git-users] Fixing up a branch on a stale branch point

[Charles Manning]

On Friday 16 April 2010 11:31:18 Rick DeNatale wrote:
 On Thu, Apr 15, 2010 at 6:45 PM, Charles Manning

 mannin...@actrix.gen.nz wrote:
  On Friday 16 April 2010 01:15:42 Rick DeNatale wrote:
  On Thu, Apr 15, 2010 at 12:59 AM, Charles Manning
 
  mannin...@actrix.gen.nz wrote:
   Hi All
  
   I have an issue that I think I can resolve in a messy way, but I
   suspect there has to be a cleaner way to do this too.
  
   On one computer I have master, plus a branch with some work on it.
   Let's call that branch B.
  
   On a different computer I made a change and pushed that to the shared
   repository.
  
   Now I want to go back to the first computer, pull the changes so that
   master reflects the current shared repository. That would put master
   ahead of the branch point, but I want to have those changes viewed in
   branch B too.
  
   Ideally there would be some way to rip a branch off and re-graft it on
   to a different branch point.
  
   It seems I could do something like
  
   -- Starting on branch B --
   git commit -m B work in progress
   git checkout master
   git pull
   git branch B-continued
   git checkout B-continued
   git merge B
   git branch -d B
   -- Continue working on B-continued--
 
  If I understand what you are asking I think it's just
 
  - Starting on Branch B
  git commit -m'B work in progress'
  git checkout master
  git pull
  git checkout B
  git merge master
  -- continue working on B
 
  I did some reading last night. It looks like the following would do what
  I want:
  -starting on B -
  git commit -m B work in progress
  git checkout master
  git pull
  git checkout B
  git rebase master B

 Be careful, you don't want to rebase a branch which you have already
 pushed, or it will wreak havoc for other users.

 If you HAVE pushed branch be then you should merge master into it to
 pick up changes from master instead of rebasing it.

Thanks for that warning,

The branch has not been pushed. It is just a private branch.

My intention is to always merge stuff onto master before pushing master. That 
should keep things a lot neater.

Thanks

-- CHarles

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

[git-users] Fixing up a branch on a stale branch point

[Charles Manning]

Hi All

I have an issue that I think I can resolve in a messy way, but I suspect there 
has to be a cleaner way to do this too.

On one computer I have master, plus a branch with some work on it. Let's call 
that branch B.

On a different computer I made a change and pushed that to the shared 
repository.

Now I want to go back to the first computer, pull the changes so that master 
reflects the current shared repository. That would put master ahead of the 
branch point, but I want to have those changes viewed in branch B too.

Ideally there would be some way to rip a branch off and re-graft it on to a 
different branch point.

It seems I could do something like

-- Starting on branch B --
git commit -m B work in progress
git checkout master
git pull
git branch B-continued
git checkout B-continued
git merge B
git branch -d B
-- Continue working on B-continued--


But there must be a neater way.


Thanks

Charles

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.

[git-users] Should gitk show all branches?

[Charles Manning]

Hi All

I find that if I do the following:
stash branch br, 
switch back to master,
modify master,
then switch back to branch br 
stash apply

gitk only shows branch x and does not show master.

I was expect to be seeing a fork-shaped graph showing both 


Here's what I did
char...@qcore:~/junk$ git init xx
Initialized empty Git repository in /home/charles/junk/xx/.git/
char...@qcore:~/junk$ cd xx
char...@qcore:~/junk/xx$ git branch
char...@qcore:~/junk/xx$ git branch -a
char...@qcore:~/junk/xx$ ls a
char...@qcore:~/junk/xx$ git add a
char...@qcore:~/junk/xx$ git commit -m add a
[master (root-commit) 6e4d94d] add a
 1 files changed, 1 insertions(+), 0 deletions(-)
 create mode 100644 a
char...@qcore:~/junk/xx$ git branch
* master
char...@qcore:~/junk/xx$ git branch br
char...@qcore:~/junk/xx$ git checkout br
Switched to branch 'br'
char...@qcore:~/junk/xx$ ls  b
char...@qcore:~/junk/xx$ git add b
char...@qcore:~/junk/xx$ ls  b
char...@qcore:~/junk/xx$ git stash
Saved working directory and index state WIP on br: 6e4d94d add a
HEAD is now at 6e4d94d add a
char...@qcore:~/junk/xx$ git checkout master
Switched to branch 'master'
char...@qcore:~/junk/xx$ ls -ial  a
char...@qcore:~/junk/xx$ git add a
char...@qcore:~/junk/xx$ git commit -m change a
[master 879e2c3] change a
 1 files changed, 5 insertions(+), 1 deletions(-)
char...@qcore:~/junk/xx$ git checkout br
Switched to branch 'br'
char...@qcore:~/junk/xx$ git stash apply
# On branch br
# Changes to be committed:
#   (use git reset HEAD file... to unstage)
#
#   new file:   b
#
char...@qcore:~/junk/xx$ gitk

This shows branch br, but not master.

Is that correct?

Thanks

Charles

-- 
You received this message because you are subscribed to the Google Groups Git 
for human beings group.
To post to this group, send email to git-us...@googlegroups.com.
To unsubscribe from this group, send email to 
git-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/git-users?hl=en.