git/refs, branch v2.37.2

Merge branch 'ep/maint-equals-null-cocci'

2022-05-20T22:26:59Z

Introduce and apply coccinelle rule to discourage an explicit comparison between a pointer and NULL, and applies the clean-up to the maintenance track. * ep/maint-equals-null-cocci: tree-wide: apply equals-null.cocci tree-wide: apply equals-null.cocci contrib/coccinnelle: add equals-null.cocci

Merge branch 'ep/maint-equals-null-cocci' for maint-2.35

2022-05-02T17:06:04Z

* ep/maint-equals-null-cocci: tree-wide: apply equals-null.cocci contrib/coccinnelle: add equals-null.cocci

tree-wide: apply equals-null.cocci

2022-05-02T16:50:37Z

Signed-off-by: Junio C Hamano

Revert "Merge branch 'ps/avoid-unnecessary-hook-invocation-with-packed-refs'"

2022-04-13T22:51:33Z

This reverts commit 991b4d47f0accd3955d05927d5ce434e03ffbdb6, reversing changes made to bcd020f88e1e22f38422ac3f73ab06b34ec4bef1.

Merge branch 'ab/refs-various-fixes'

2022-03-29T19:22:02Z

Code clean-up. * ab/refs-various-fixes: refs debug: add a wrapper for "read_symbolic_ref" packed-backend: remove stub BUG(...) functions misc *.c: use designated initializers for struct assignments refs: use designated initializers for "struct ref_iterator_vtable" refs: use designated initializers for "struct ref_storage_be"

Merge branch 'ps/fsync-refs'

2022-03-25T23:38:25Z

Updates to refs traditionally weren't fsync'ed, but we can configure using core.fsync variable to do so. * ps/fsync-refs: core.fsync: new option to harden references

refs debug: add a wrapper for "read_symbolic_ref"

2022-03-17T17:40:14Z

In cd475b3b038 (refs: add ability for backends to special-case reading of symbolic refs, 2022-03-01) when the "read_symbolic_ref" callback was added we'd fall back on "refs_read_raw_ref" if there wasn't any backend implementation of "read_symbolic_ref". As discussed in the preceding commit this would only happen if we were running the "debug" backend, e.g. in the "setup for ref completion" test in t9902-completion.sh with: GIT_TRACE_REFS=1 git fetch --no-tags other Let's improve the trace output, but and also eliminate the now-redundant refs_read_raw_ref() fallback case. As noted in the preceding commit the "packed" backend will never call refs_read_symbolic_ref() (nor is it ever going to). For any future backend such as reftable it's OK to ask that they either implement this (or a wrapper) themselves. Signed-off-by: Ævar Arnfjörð Bjarmason Signed-off-by: Junio C Hamano

packed-backend: remove stub BUG(...) functions

2022-03-17T17:38:05Z

Remove the stub BUG(...) functions previously used by the "struct ref_storage_be refs_be_packed" backend. We never call any functions in the packed backend by using it as a "normal" primary ref store, instead we'll always initialize a "files" backend ref-store. It will then via the "packed_ref_store" member of "struct files_ref_store" call selected functions in the "packed" backend, and we'll in addition call others via wrappers in refs.c. So while these would arguably give us *slightly* more meaningful error messages we'll NULL the missing members in the initializer anyway, so we'll reliably get a segfault if we're ever changing the backend and having it call something it doesn't have. So there's no need for this verbose boilerplate, and as shown in a subsequent commit it might even lead to some confusion about the packed backend being a "real" backend. Let's make it clear that it's not. As an aside, this also fixes a warning emitted by SunCC in at least versions 12.5 and 12.6 of Oracle Developer Studio: "refs/packed-backend.c", line 1599: warning: Function has no return statement : packed_create_symref "refs/packed-backend.c", line 1606: warning: Function has no return statement : packed_rename_ref) "refs/packed-backend.c", line 1613: warning: Function has no return statement : packed_copy_ref "refs/packed-backend.c", line 1648: warning: Function has no return statement : packed_create_reflog Signed-off-by: Ævar Arnfjörð Bjarmason Signed-off-by: Junio C Hamano

refs: use designated initializers for "struct ref_iterator_vtable"

2022-03-17T17:36:11Z

Signed-off-by: Ævar Arnfjörð Bjarmason Signed-off-by: Junio C Hamano

refs: use designated initializers for "struct ref_storage_be"

2022-03-17T17:36:04Z

Change the definition of the three refs backends we currently carry to use designated initializers. The "= NULL" assignments being retained here are redundant, and could be removed, but let's keep them for clarity. All of these backends define almost all fields, so we're not saving much in terms of line count by omitting these, but e.g. for "refs_be_debug" it's immediately apparent that we're omitting "init" when comparing its assignment to the others. This is a follow-up to similar work merged in bd4232fac33 (Merge branch 'ab/struct-init', 2021-07-16), a4b9fb6a5cf (Merge branch 'ab/designated-initializers-more', 2021-10-18) and a30321b9eae (Merge branch 'ab/designated-initializers' into ab/designated-initializers-more, 2021-09-27). Signed-off-by: Ævar Arnfjörð Bjarmason Signed-off-by: Junio C Hamano