Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] The great git migration 2011

On 10 May 2011 12:25, Alex Blewitt <alex.blewitt@xxxxxxxxx> wrote:
I thought the plan was to filter out the old stuff? So the 150M was for the single repo post cruft filtering. 

That's probably about right, yes.
We do have a few exes, not many. Mostly in the tests for the symbol readers.  As a rule there's not much derived content to speak of in the repo.

Not now, but do you know for sure that there has never been am accidental commit? Whilst this wouldn't be obvious from CVS, the Git repo will contain all historic mistakes. 

Well there are a few binary blobs, and these are checked in on purpose.  I agree it's not ideal but I don't think we can do anything about them as removing them will invalidate recent tags and branches.  For example all the native bits are checked and are only rebuilt as necessary due to issues with finding a target platform and the fact that they change infrequently.  

Using xdu, the main contributions to the size are: 
  - The CDT parser is big: 
          cdt.core (16M)
          cdt.core.tests (6.1M)
          cdt.ui (15M)
          cdt.ui.tests (3.7M)
    Total: cdt-core = 43M 
 - documentation pngs 
          images = 15M
- EDC tests 
          resources = 42M. 
These 3 are ~60% of a checkout.

Cheers,
James


Back to the top