Jeff King wrote:
> On Fri, Nov 22, 2013 at 04:24:05PM -0800, Jonathan Nieder wrote:

>>>  t/                     |  66 ------------------
>> Hmm, not all of these tests are about the "experimental" format.  Do
>> we really want to remove them all?
> I think so. They were not all testing the experimental format, but they
> were about making sure the is-it-experimental heuristic triggered
> properly with various zlib settings.
> Now that we do not apply that heuristic, there is nothing (in git) to
> test. We feed the contents straight to zlib.

Ok, makes sense.

In principle the tests are still useful as futureproofing in case git
starts to sanity-check the objects as a way to notice corruption
earlier or something.  But in practice, that kind of futureproofing is
probably not worth the extra tests to maintain.

For what it's worth,
Reviewed-by: Jonathan Nieder <>
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to
More majordomo info at

Reply via email to