Git repository clonning

Git repository clonning

[Martin Přecechtěl]

To Whom It May Concern,
I have a problem and I would like to ask you for help. I need to somehow 
access the git repositry over network and read files from it. In the 
concrete, only thing I need is to read some files from repository, where 
I have files needed for installation of computers. So on  computer where 
installation takes place, I need to read from this repository some files 
needed for installation process. I tried command git clone 
<my_repository> but this command download the whole repository which 
size is now almost 8GB files + 4GB .git folder. This process is very 
slow. So I want to ask you if there is a way how to access my repository 
(read some files from it) but to significantly improve speed. Is there 
for example something what will somehow map the repository to folder and 
then dowload only thouse files which are accessed?

I would appreciate any help.

Thank you for your time,
Yours faithfully,
Martin Precechtel

Re: Git repository clonning

[Jakub Narebski]

Martin Přecechtěl <precechtel@gmail.com> writes:

> I have a problem and I would like to ask you for help. I need to
> somehow access the git repository over network and read files from
> it. In the concrete, only thing I need is to read some files from
> repository, where I have files needed for installation of
> computers. So on  computer where installation takes place, I need to
> read from this repository some files needed for installation
> process.

Where this repository is?  Is it on computer you control?  If yes, why
not use some networking filesystem or just rsync checked out files?

> I tried command git clone <my_repository> but this command
> download the whole repository which size is now almost 8GB files + 4GB
> .git folder. This process is very slow. So I want to ask you if there
> is a way how to access my repository (read some files from it) but to
> significantly improve speed. Is there for example something what will
> somehow map the repository to folder and then dowload only thouse
> files which are accessed?

You can try to use shallow clone, which would download only tip of
history... but all the files:

  $ git clone --depth=0 <URL>

If remote repository is configured correctly, and all the files are
inside one directory, you can try to use remote archive:

  $ git archive --remote=<URL> --prefix=<dir>/ --output=<dir>.tar \
    HEAD:<dir>

It wouldprobably not work.

If there is some web interface, like gitweb or GitHub, you can
download snapshots of some subdirectory from there.

HTH
-- 
Jakub Narębski