From: Patrick Steinhardt <ps@pks.im>
To: Adrian Ratiu <adrian.ratiu@collabora.com>
Cc: git@vger.kernel.org, Emily Shaffer <emilyshaffer@google.com>,
Junio C Hamano <gitster@pobox.com>,
"brian m . carlson" <sandals@crustytoothpaste.net>
Subject: Re: [PATCH 10/10] hook: show disabled hooks in "git hook list"
Date: Wed, 11 Mar 2026 11:24:42 +0100 [thread overview]
Message-ID: <abFC6pEEmicjG6a9@pks.im> (raw)
In-Reply-To: <20260309005416.2760030-11-adrian.ratiu@collabora.com>
On Mon, Mar 09, 2026 at 02:54:16AM +0200, Adrian Ratiu wrote:
> diff --git a/builtin/hook.c b/builtin/hook.c
> index c806640361..ff446948fa 100644
> --- a/builtin/hook.c
> +++ b/builtin/hook.c
> @@ -72,16 +72,20 @@ static int list(int argc, const char **argv, const char *prefix,
> case HOOK_TRADITIONAL:
> printf("%s%c", _("hook from hookdir"), line_terminator);
> break;
> - case HOOK_CONFIGURED:
> - if (show_scope)
> - printf("%s (%s)%c",
> - h->u.configured.friendly_name,
> - config_scope_name(h->u.configured.scope),
> + case HOOK_CONFIGURED: {
> + const char *name = h->u.configured.friendly_name;
> + const char *scope = show_scope ?
> + config_scope_name(h->u.configured.scope) : NULL;
> + if (scope)
> + printf("%s (%s%s)%c", name, scope,
> + h->u.configured.disabled ? ", disabled" : "",
> line_terminator);
> + else if (h->u.configured.disabled)
> + printf("%s (disabled)%c", name, line_terminator);
> else
> - printf("%s%c", h->u.configured.friendly_name,
> - line_terminator);
> + printf("%s%c", name, line_terminator);
> break;
> + }
> default:
> BUG("unknown hook kind");
> }
Hm. This starts to feel less and less like an interface that can easily
be parsed by a machine, even with "-z". I guess this partly comes from
our insistence to reinvent the wheel in Git instead of just using
something like JSON :/
> diff --git a/hook.c b/hook.c
> index 2c03baeaac..4f4f060156 100644
> --- a/hook.c
> +++ b/hook.c
> @@ -119,6 +119,7 @@ static void list_hooks_add_default(struct repository *r, const char *hookname,
> struct hook_config_cache_entry {
> char *command;
> enum config_scope scope;
> + int disabled;
> };
>
> /*
Is there any reason this is an `int` and not a `bool`?
> @@ -217,8 +218,10 @@ static int hook_config_lookup_all(const char *key, const char *value,
> * every item's string is the hook's friendly-name and its util pointer is
> * a hook_config_cache_entry. All strings are owned by the map.
> *
> - * Disabled hooks and hooks missing a command are already filtered out at
> - * parse time, so callers can iterate the list directly.
> + * Disabled hooks are kept in the cache with entry->disabled set, so that
> + * "git hook list" can display them. Hooks missing a command are filtered
> + * out at build time; if a disabled hook has no command it is silently
What exactly does "build time" refer to? To me this reads like invoking
make :)
> + * skipped rather than triggering a fatal error.
> */
> void hook_cache_clear(struct hook_config_cache *cache)
> {
> @@ -268,21 +271,26 @@ static void build_hook_config_map(struct repository *r,
> struct hook_config_cache_entry *entry;
> char *command;
>
> - /* filter out disabled hooks */
> - if (unsorted_string_list_lookup(&cb_data.disabled_hooks,
> - hname))
> - continue;
> + int is_disabled =
> + !!unsorted_string_list_lookup(
> + &cb_data.disabled_hooks, hname);
>
> command = strmap_get(&cb_data.commands, hname);
> - if (!command)
> - die(_("'hook.%s.command' must be configured or "
> - "'hook.%s.event' must be removed;"
> - " aborting."), hname, hname);
> + if (!command) {
> + if (is_disabled)
> + warning(_("disabled hook '%s' has no "
> + "command configured"), hname);
> + else
> + die(_("'hook.%s.command' must be configured or "
> + "'hook.%s.event' must be removed;"
> + " aborting."), hname, hname);
> + }
>
> /* util stores a cache entry; owned by the cache. */
> CALLOC_ARRAY(entry, 1);
> - entry->command = xstrdup(command);
> + entry->command = command ? xstrdup(command) : NULL;
You can use `xstrdup_or_null()` here.
> entry->scope = scope;
> + entry->disabled = is_disabled;
> string_list_append(hooks, hname)->util = entry;
> }
>
> @@ -401,7 +411,16 @@ struct string_list *list_hooks(struct repository *r, const char *hookname,
> int hook_exists(struct repository *r, const char *name)
> {
> struct string_list *hooks = list_hooks(r, name, NULL);
> - int exists = hooks->nr > 0;
> + int exists = 0;
> +
> + for (size_t i = 0; i < hooks->nr; i++) {
> + struct hook *h = hooks->items[i].util;
> + if (h->kind == HOOK_TRADITIONAL ||
> + !h->u.configured.disabled) {
Is the first condition required? I would expect that `disabled` would
always be false for traditional hooks.
> + exists = 1;
> + break;
> + }
> + }
> string_list_clear_func(hooks, hook_free);
> free(hooks);
> return exists;
> diff --git a/hook.h b/hook.h
> index 0d711ed21a..0432df963f 100644
> --- a/hook.h
> +++ b/hook.h
> @@ -31,6 +31,7 @@ struct hook {
> const char *friendly_name;
> const char *command;
> enum config_scope scope;
> + int disabled;
> } configured;
> } u;
>
Same question here regarding the type of the struct member.
Patrick
next prev parent reply other threads:[~2026-03-11 10:24 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-09 0:54 [PATCH 00/10] config-hook cleanups and two small 'git hook list' features Adrian Ratiu
2026-03-09 0:54 ` [PATCH 01/10] hook: move unsorted_string_list_remove() to string-list.[ch] Adrian Ratiu
2026-03-10 19:56 ` SZEDER Gábor
2026-03-11 11:08 ` Adrian Ratiu
2026-03-09 0:54 ` [PATCH 02/10] hook: fix minor style issues Adrian Ratiu
2026-03-09 2:12 ` Eric Sunshine
2026-03-09 0:54 ` [PATCH 03/10] hook: rename cb_data_free/alloc -> hook_data_free/alloc Adrian Ratiu
2026-03-11 10:24 ` Patrick Steinhardt
2026-03-11 11:09 ` Adrian Ratiu
2026-03-09 0:54 ` [PATCH 04/10] hook: detect & emit two more bugs Adrian Ratiu
2026-03-09 0:54 ` [PATCH 05/10] hook: replace hook_list_clear() -> string_list_clear_func() Adrian Ratiu
2026-03-09 2:18 ` Eric Sunshine
2026-03-10 14:20 ` Adrian Ratiu
2026-03-09 0:54 ` [PATCH 06/10] hook: make consistent use of friendly-name in docs Adrian Ratiu
2026-03-09 0:54 ` [PATCH 07/10] t1800: add test to verify hook execution ordering Adrian Ratiu
2026-03-09 0:54 ` [PATCH 08/10] hook: refactor hook_config_cache from strmap to named struct Adrian Ratiu
2026-03-09 21:59 ` Junio C Hamano
2026-03-10 14:19 ` Adrian Ratiu
2026-03-09 0:54 ` [PATCH 09/10] hook: show config scope in git hook list Adrian Ratiu
2026-03-09 21:59 ` Junio C Hamano
2026-03-10 14:45 ` Adrian Ratiu
2026-03-11 10:24 ` Patrick Steinhardt
2026-03-11 11:47 ` Adrian Ratiu
2026-03-09 0:54 ` [PATCH 10/10] hook: show disabled hooks in "git hook list" Adrian Ratiu
2026-03-11 10:24 ` Patrick Steinhardt [this message]
2026-03-11 12:24 ` Adrian Ratiu
2026-03-11 13:53 ` Patrick Steinhardt
2026-03-09 20:14 ` [PATCH 00/10] config-hook cleanups and two small 'git hook list' features Junio C Hamano
2026-03-10 14:37 ` Adrian Ratiu
2026-03-09 20:27 ` Junio C Hamano
2026-03-20 11:52 ` [PATCH v2 " Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 01/10] hook: move unsorted_string_list_remove() to string-list.[ch] Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 02/10] hook: fix minor style issues Adrian Ratiu
2026-03-24 8:37 ` Patrick Steinhardt
2026-03-24 19:19 ` Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 03/10] hook: rename cb_data_free/alloc -> hook_data_free/alloc Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 04/10] hook: detect & emit two more bugs Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 05/10] hook: replace hook_list_clear() -> string_list_clear_func() Adrian Ratiu
2026-03-24 8:37 ` Patrick Steinhardt
2026-03-24 22:33 ` Adrian Ratiu
2026-03-25 5:26 ` Patrick Steinhardt
2026-03-20 11:52 ` [PATCH v2 06/10] hook: make consistent use of friendly-name in docs Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 07/10] t1800: add test to verify hook execution ordering Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 08/10] hook: introduce hook_config_cache_entry for per-hook data Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 09/10] hook: show config scope in git hook list Adrian Ratiu
2026-03-24 8:37 ` Patrick Steinhardt
2026-03-25 11:28 ` Adrian Ratiu
2026-03-20 11:52 ` [PATCH v2 10/10] hook: show disabled hooks in "git hook list" Adrian Ratiu
2026-03-24 8:38 ` Patrick Steinhardt
2026-03-24 16:14 ` Junio C Hamano
2026-03-24 19:23 ` Adrian Ratiu
2026-03-23 16:11 ` [PATCH v2 00/10] config-hook cleanups and two small 'git hook list' features Junio C Hamano
2026-03-24 8:38 ` Patrick Steinhardt
2026-03-24 18:56 ` Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 00/12] config-hook cleanups and three small git-hook features Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 01/12] hook: move unsorted_string_list_remove() to string-list.[ch] Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 02/12] builtin/receive-pack: properly init receive_hook strbuf Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 03/12] hook: fix minor style issues Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 04/12] hook: rename cb_data_free/alloc -> hook_data_free/alloc Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 05/12] hook: detect & emit two more bugs Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 06/12] hook: replace hook_list_clear() -> string_list_clear_func() Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 07/12] hook: make consistent use of friendly-name in docs Adrian Ratiu
2026-03-25 19:54 ` [PATCH v3 08/12] t1800: add test to verify hook execution ordering Adrian Ratiu
2026-03-25 19:55 ` [PATCH v3 09/12] hook: introduce hook_config_cache_entry for per-hook data Adrian Ratiu
2026-03-25 19:55 ` [PATCH v3 10/12] hook: show config scope in git hook list Adrian Ratiu
2026-03-25 19:55 ` [PATCH v3 11/12] hook: show disabled hooks in "git hook list" Adrian Ratiu
2026-03-25 19:55 ` [PATCH v3 12/12] hook: reject unknown hook names in git-hook(1) Adrian Ratiu
2026-03-25 21:17 ` [PATCH v3 00/12] config-hook cleanups and three small git-hook features Junio C Hamano
2026-03-26 10:21 ` Adrian Ratiu
2026-03-27 8:04 ` Patrick Steinhardt
2026-03-27 16:11 ` Junio C Hamano
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=abFC6pEEmicjG6a9@pks.im \
--to=ps@pks.im \
--cc=adrian.ratiu@collabora.com \
--cc=emilyshaffer@google.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=sandals@crustytoothpaste.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox