[PATCHv2 0/6] duplicate objects and delta cycles, oh my!

From: Jeff King <peff@peff.net>
Date: Aug 22, 2013, 23:12 UTC
Message-ID: <20130822231215.GA16978@sigill.intra.peff.net>
In-Reply-To: <20130822144305.GA21219@sigill.intra.peff.net>

On Thu, Aug 22, 2013 at 10:43:05AM -0400, Jeff King wrote:

Show 5 quoted lines

> > write_idx_file() is called after index-pack processes all delta
> > objects. Could resolve_deltas() go cyclic with certain duplicate
> > object setup?
> 
> Good question. I'm not sure. I'll check it out.

I think the answer is "no", based on both reasoning and testing (both of which are included in patches 3-4 of the series below).

So here's my re-roll of the series.

  [1/6]: test-sha1: add a binary output mode

    New in this iteration; the previous version piped test-sha1 into
    perl to create the pack trailer, but with this simple change we can
    drop the perl dependency.

  [2/6]: sha1-lookup: handle duplicate keys with GIT_USE_LOOKUP

    Same code as before. I've factored the pack-creation bits from the
    tests into lib-pack.sh, so they can be reused elsewhere when we want
    to create bogus packs (and patches 3-4 reuse them here).

  [3/6]: add tests for indexing packs with delta cycles
  [4/6]: test index-pack on packs with recoverable delta cycles

    New tests covering delta cycles.

  [5/6]: index-pack: optionally reject packs with duplicate objects

    Similar to before, but I converted the config flag to a simple
    boolean (since we scrapped the "fix" of the tri-state "allow,
    reject, fix").

  [6/6]: default pack.indexDuplicates to false

    This flips the safety check on by default everywhere (before, it was
    left off for index-pack).

-Peff