similar to what we do for loose objects, avoid deleting pack files that were created "too soon" unless -a is given. This prevents races when gotadmin load, got fetch or gotd are installing a new pack file and a concurrent gotadmin clean attempts to remove it. ok stsp

leftover, the error handling of unlink() a couple of lines below used to carry a similar special-casing but was dropped; drop it here too.

This wraps the cleaning up of loose objects and redundant pack files under a new functions, making the _prepare() and _complete() functions unnecessary. It walks the reachable commits unconditionally since that information is always needed and adapt how we purge loose objects after this change. The progress function is changed too and we log `ncommits' first, followed by the number of loose objects, followed by the rest. Pack files are now considered redundant when all their objects are provided by a bigger pack or are unreachable. ok stsp

ok stsp

This adds the functions got_repo_cleanup_prepare() and _complete() to lock/unlock the repository to prevent multiple `gotadmin cleanup' or `git gc' operations to run in parallel. improvements and ok stsp@

improvements and ok stsp@

and use it to avoid hardcoding the digest type in a few places. ok stsp@

it abstracts over the hash type and ensures that object ids are zero'ed before their sha1 digest is written. Needed by the incoming sha256 support. ok stsp@

It will soon grow functions to deal with sha256 too. stsp@ agrees.

In preparation for wide sha256 support; stsp@ agrees. Change done mechanically with find . -iname \*.[cy] -exec sam {} + X ,x/<sha1\.h>/i/\n#include <sha2.h>

Intended mostly for the regress suite, so we'll be able to test also the ref-delta code paths. ok stsp@

gotd used 0600 (due to mkstemps(3)), gotadmin 0644; change it to 0444 since packfiles shouldn't change once created. Mirrors what git does. ok stsp@

Allows 'got commit' to use a ".diff" suffix for temporary diff files. ok op@

Needed by future gotd(8), where progress reports will be sent to a network socket, rather than a local terminal.

Future gotd(8) needs to run got_pack_create() in a chroot environment, so we can no longer open new temporary files inside got_pack_create(). ok op@

The old code required a seekable output file. This conflicts with requirements of future gotd(8), which will write pack file data to network sockets. ok op@

This can't happen in practice, but llvm's scan-build doesn't know this. OK stsp

Saves us from doing a malloc/free call for every item on the list. ok op@

imsg_clear frees and closes resources allocated as part of enqueueing imsgs so it's a no-op after reads. discussed with and ok stsp@

tested by thomas, naddy, and myself