Zimbra Port

[Torsten Zuehlsdorff]

On 02.06.2016 07:41, Bradley T. Hughes wrote:
>
>> On 01 Jun 2016, at 16:08, Torsten Zuehlsdorff <mailinglists at toco-domains.de> wrote:
>>
>>> * The Zimbra source is huge, a git clone is about 13 GigaBytes. I am not
>>> sure on how source that big is handled correctly in ports. (e.g. is it
>>> OK that every make does a git clone and you have to wait until you get
>>> the 13 GB of data? Would this be a problem for the FreeBSD build cluster
>>> infrastructure to create the packages?, ...)
>>
>> A git clone contains the complete development *history*. If a try to fetch a packet from their homepage it is "just" around 850 MB which should not be a problem.
>>
>>> * On the porters handbook it says to fetch a tarball from http/ftp, is
>>> it also possible to directly work with git and clone a repository?
>>
>> I'm not aware of such a way. But since there is a Git-Repo: isn't there a management tool like GitLab, which could provides tarballs? At least it is possible to set one up, clone the repo and provide it this way.
>
> There's also the --depth option to git clone that can help limit the size of the resulting checkout.
>
>        --depth <depth>
>            Create a shallow clone with a history truncated to the specified
>            number of commits. Implies --single-branch unless
>            --no-single-branch is given to fetch the histories near the tips of
>            all branches.

Yes, but i doubt this could be a solution. The ports ship a distinfo 
file, which contains size and checksum of the distfiles for the port. 
This is not possible with a pure git.

But since this is open source: in the worst case we can create the 
distfiles by ourself and host them. I don't see this as stopper.

Greetings,
Torsten