| Message ID | b09f21dff786780dfb0ae71efd9a0cabe31c1e0b.1684324059.git.ps@pks.im (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | d1adf85b0a21ea6575b145959c53e7d5e4a0cba4 |
| Headers | show |
| Series | fetch: smallish cleanups | expand |
On Wed, May 17, 2023 at 01:48:51PM +0200, Patrick Steinhardt wrote: > The `fetch_config` structure currently only has a single member, which > is the `display_format`. We're about extend it to contain all parsed > config values and will thus need it available in most of the code. > > Prepare for this change by passing through the `fetch_config` directly > instead of only passing its single member. Makes sense. One small nit: > static int do_fetch(struct transport *transport, > struct refspec *rs, > - enum display_format display_format) > + const struct fetch_config *config) > { > struct ref_transaction *transaction = NULL; > struct ref *ref_map = NULL; > @@ -1639,7 +1639,8 @@ static int do_fetch(struct transport *transport, > if (retcode) > goto cleanup; > > - display_state_init(&display_state, ref_map, transport->url, display_format); > + display_state_init(&display_state, ref_map, transport->url, > + config->display_format); If the point is that fetch_config may start carrying new information, wouldn't we want to pass it as a whole down to display_state_init()? It might eventually want to see some of that other config, too. It's presumably academic for now, and it would not be too hard to change later if needed, so I don't know that it's worth a re-roll. I just found it especially funny here since the purpose of the patch is to treat the config struct as a single unit. -Peff
On Thu, May 18, 2023 at 08:18:03PM -0400, Jeff King wrote: > On Wed, May 17, 2023 at 01:48:51PM +0200, Patrick Steinhardt wrote: > > > The `fetch_config` structure currently only has a single member, which > > is the `display_format`. We're about extend it to contain all parsed > > config values and will thus need it available in most of the code. > > > > Prepare for this change by passing through the `fetch_config` directly > > instead of only passing its single member. > > Makes sense. > > One small nit: > > > static int do_fetch(struct transport *transport, > > struct refspec *rs, > > - enum display_format display_format) > > + const struct fetch_config *config) > > { > > struct ref_transaction *transaction = NULL; > > struct ref *ref_map = NULL; > > @@ -1639,7 +1639,8 @@ static int do_fetch(struct transport *transport, > > if (retcode) > > goto cleanup; > > > > - display_state_init(&display_state, ref_map, transport->url, display_format); > > + display_state_init(&display_state, ref_map, transport->url, > > + config->display_format); > > If the point is that fetch_config may start carrying new information, > wouldn't we want to pass it as a whole down to display_state_init()? It > might eventually want to see some of that other config, too. > > It's presumably academic for now, and it would not be too hard to change > later if needed, so I don't know that it's worth a re-roll. I just found > it especially funny here since the purpose of the patch is to treat the > config struct as a single unit. Well, I decided against passing in the full configuration as it feels a bit like a layering violation: the other code really is about the fetch itself, while this code here is only about display logic. So passing in the `fetch_config` felt weird to me, even more so because we continue to only need that single value at the end of this series. I do see your point though. Given that none of your other comments require a reroll I'll leave this as-is for now. Thanks for your review! Patrick
On Mon, May 22, 2023 at 10:58:51AM +0200, Patrick Steinhardt wrote: > > If the point is that fetch_config may start carrying new information, > > wouldn't we want to pass it as a whole down to display_state_init()? It > > might eventually want to see some of that other config, too. > > > > It's presumably academic for now, and it would not be too hard to change > > later if needed, so I don't know that it's worth a re-roll. I just found > > it especially funny here since the purpose of the patch is to treat the > > config struct as a single unit. > > Well, I decided against passing in the full configuration as it feels a > bit like a layering violation: the other code really is about the fetch > itself, while this code here is only about display logic. So passing in > the `fetch_config` felt weird to me, even more so because we continue to > only need that single value at the end of this series. I do see your > point though. > > Given that none of your other comments require a reroll I'll leave this > as-is for now. Thanks for your review! Yeah, I could see that line of thinking, as well. Leaving sounds good to me. And I think that was my only substantive comment on the whole series, so we can consider the whole reviewed-by: me. -Peff
diff --git a/builtin/fetch.c b/builtin/fetch.c index f268322e6f..401543e05d 100644 --- a/builtin/fetch.c +++ b/builtin/fetch.c @@ -1553,7 +1553,7 @@ static int backfill_tags(struct display_state *display_state, static int do_fetch(struct transport *transport, struct refspec *rs, - enum display_format display_format) + const struct fetch_config *config) { struct ref_transaction *transaction = NULL; struct ref *ref_map = NULL; @@ -1639,7 +1639,8 @@ static int do_fetch(struct transport *transport, if (retcode) goto cleanup; - display_state_init(&display_state, ref_map, transport->url, display_format); + display_state_init(&display_state, ref_map, transport->url, + config->display_format); if (atomic_fetch) { transaction = ref_transaction_begin(&err); @@ -1828,7 +1829,7 @@ static int add_remote_or_group(const char *name, struct string_list *list) } static void add_options_to_argv(struct strvec *argv, - enum display_format format) + const struct fetch_config *config) { if (dry_run) strvec_push(argv, "--dry-run"); @@ -1864,7 +1865,7 @@ static void add_options_to_argv(struct strvec *argv, strvec_push(argv, "--ipv6"); if (!write_fetch_head) strvec_push(argv, "--no-write-fetch-head"); - if (format == DISPLAY_FORMAT_PORCELAIN) + if (config->display_format == DISPLAY_FORMAT_PORCELAIN) strvec_pushf(argv, "--porcelain"); } @@ -1874,7 +1875,7 @@ struct parallel_fetch_state { const char **argv; struct string_list *remotes; int next, result; - enum display_format format; + const struct fetch_config *config; }; static int fetch_next_remote(struct child_process *cp, @@ -1894,7 +1895,7 @@ static int fetch_next_remote(struct child_process *cp, strvec_push(&cp->args, remote); cp->git_cmd = 1; - if (verbosity >= 0 && state->format != DISPLAY_FORMAT_PORCELAIN) + if (verbosity >= 0 && state->config->display_format != DISPLAY_FORMAT_PORCELAIN) printf(_("Fetching %s\n"), remote); return 1; @@ -1927,7 +1928,7 @@ static int fetch_finished(int result, struct strbuf *out, } static int fetch_multiple(struct string_list *list, int max_children, - enum display_format format) + const struct fetch_config *config) { int i, result = 0; struct strvec argv = STRVEC_INIT; @@ -1945,10 +1946,10 @@ static int fetch_multiple(struct string_list *list, int max_children, strvec_pushl(&argv, "-c", "fetch.bundleURI=", "fetch", "--append", "--no-auto-gc", "--no-write-commit-graph", NULL); - add_options_to_argv(&argv, format); + add_options_to_argv(&argv, config); if (max_children != 1 && list->nr != 1) { - struct parallel_fetch_state state = { argv.v, list, 0, 0, format }; + struct parallel_fetch_state state = { argv.v, list, 0, 0, config }; const struct run_process_parallel_opts opts = { .tr2_category = "fetch", .tr2_label = "parallel/fetch", @@ -1972,7 +1973,7 @@ static int fetch_multiple(struct string_list *list, int max_children, strvec_pushv(&cmd.args, argv.v); strvec_push(&cmd.args, name); - if (verbosity >= 0 && format != DISPLAY_FORMAT_PORCELAIN) + if (verbosity >= 0 && config->display_format != DISPLAY_FORMAT_PORCELAIN) printf(_("Fetching %s\n"), name); cmd.git_cmd = 1; if (run_command(&cmd)) { @@ -2028,7 +2029,7 @@ static inline void fetch_one_setup_partial(struct remote *remote) static int fetch_one(struct remote *remote, int argc, const char **argv, int prune_tags_ok, int use_stdin_refspecs, - enum display_format display_format) + const struct fetch_config *config) { struct refspec rs = REFSPEC_INIT_FETCH; int i; @@ -2095,7 +2096,7 @@ static int fetch_one(struct remote *remote, int argc, const char **argv, sigchain_push_common(unlock_pack_on_signal); atexit(unlock_pack_atexit); sigchain_push(SIGPIPE, SIG_IGN); - exit_code = do_fetch(gtransport, &rs, display_format); + exit_code = do_fetch(gtransport, &rs, config); sigchain_pop(SIGPIPE); refspec_clear(&rs); transport_disconnect(gtransport); @@ -2383,7 +2384,7 @@ int cmd_fetch(int argc, const char **argv, const char *prefix) if (filter_options.choice || repo_has_promisor_remote(the_repository)) fetch_one_setup_partial(remote); result = fetch_one(remote, argc, argv, prune_tags_ok, stdin_refspecs, - config.display_format); + &config); } else { int max_children = max_jobs; @@ -2403,7 +2404,7 @@ int cmd_fetch(int argc, const char **argv, const char *prefix) max_children = fetch_parallel_config; /* TODO should this also die if we have a previous partial-clone? */ - result = fetch_multiple(&list, max_children, config.display_format); + result = fetch_multiple(&list, max_children, &config); } /* @@ -2424,7 +2425,7 @@ int cmd_fetch(int argc, const char **argv, const char *prefix) if (max_children < 0) max_children = fetch_parallel_config; - add_options_to_argv(&options, config.display_format); + add_options_to_argv(&options, &config); result = fetch_submodules(the_repository, &options, submodule_prefix,
The `fetch_config` structure currently only has a single member, which is the `display_format`. We're about extend it to contain all parsed config values and will thus need it available in most of the code. Prepare for this change by passing through the `fetch_config` directly instead of only passing its single member. Signed-off-by: Patrick Steinhardt <ps@pks.im> --- builtin/fetch.c | 31 ++++++++++++++++--------------- 1 file changed, 16 insertions(+), 15 deletions(-)