diff options
| author | Francois Kritzinger <francois@codesynthesis.com> | 2024-12-02 13:36:45 +0200 |
|---|---|---|
| committer | Francois Kritzinger <francois@codesynthesis.com> | 2024-12-03 08:25:21 +0200 |
| commit | e938c4a00c46cdff5a5718a0cec3ae7af714c583 (patch) | |
| tree | bce0a8c043ea4f06ddd84f83024c3641b67f4b0e /mod/mod-ci-github.cxx | |
| parent | d67a516238a4a004d60b6ea6b0b6862abeb9d27c (diff) | |
Clean up
Diffstat (limited to 'mod/mod-ci-github.cxx')
| -rw-r--r-- | mod/mod-ci-github.cxx | 25 |
1 files changed, 2 insertions, 23 deletions
diff --git a/mod/mod-ci-github.cxx b/mod/mod-ci-github.cxx index 113da2e..e04570e 100644 --- a/mod/mod-ci-github.cxx +++ b/mod/mod-ci-github.cxx @@ -21,19 +21,6 @@ // @@ Remaining TODOs // -// - Rerequested checks -// -// - check_suite (action: rerequested): received when user re-runs all -// checks. -// -// - check_run (action: rerequested): received when user re-runs a -// specific check or all failed checks. -// -// @@ TMP I have confirmed that the above is accurate. -// -// Will need to extract a few more fields from check_runs, but the layout -// is very similar to that of check_suite. -// // - Choose strong webhook secret (when deploying). // // - Check that delivery UUID has not been received before (replay attack). @@ -280,9 +267,6 @@ namespace brep // is that we want be "notified" of new actions at which point we can // decide whether to ignore them or to handle. // - // @@ There is also check_run even (re-requested by user, either - // individual check run or all the failed check runs). - // if (event == "check_suite") { gh_check_suite_event cs; @@ -558,10 +542,6 @@ namespace brep l3 ([&]{trace << "installation_access_token { " << *iat << " }";}); - // @@ What happens if we call this functions with an already existing - // node_id (e.g., replay attack). See the UUID header above. - // - // While it would have been nice to cancel CIs of PRs with this branch as // base not to waste resources, there are complications: Firstly, we can // only do this for remote PRs (since local PRs will most likely share the @@ -2253,9 +2233,8 @@ namespace brep NOTIFICATION_DIAG (log_writer); - // @@ TODO Include service_data::event_node_id and perhaps ts.id in - // diagnostics? E.g. when failing to update check runs we print the - // build ID only. + // @@ TODO Include ts.id in diagnostics? Check run build ids alone seem + // kind of meaningless. Log lines get pretty long this way however. service_data sd; try |