CHANGELOG: v1.34.0 !458
1 file changed, +17 −1
Layout: unified · split
CHANGELOG.org +17 −1
| @@ -4,7 +4,7 @@ Versioning follows semver from v0.1.0. Database migrations run | |||
| 4 | automatically on daemon start; upgrade notes appear per release when | 4 | automatically on daemon start; upgrade notes appear per release when |
| 5 | anything beyond "replace the binary and restart" is needed. | 5 | anything beyond "replace the binary and restart" is needed. |
| 6 | 6 | ||
| 7 | * Unreleased | 7 | * v1.34.0 — 2026-09-21 |
| 8 | 8 | ||
| 9 | The profile's tabs, reworked. | 9 | The profile's tabs, reworked. |
| 10 | 10 | ||
| @@ -61,6 +61,22 @@ Eleven web findings, mostly alignment and affordance (#247). | |||
| 61 | - =nav.tabs= has a bottom margin. | 61 | - =nav.tabs= has a bottom margin. |
| 62 | - The footer links the iOS client. | 62 | - The footer links the iOS client. |
| 63 | 63 | ||
| 64 | The e2e suite runs in parallel (#246). | ||
| 65 | |||
| 66 | - =startInstance= linked a byte-identical 35MB gitbayd for each of the | ||
| 67 | 207 tests, about 0.8s apiece: go's build cache covers the compile, | ||
| 68 | not the link. Each binary is now built once per process, lazily, so | ||
| 69 | a =-run= of one test links only what it needs. | ||
| 70 | - 202 of the 207 tests take =t.Parallel=; the five calling =t.Setenv= | ||
| 71 | stay serial. =freePorts= allocates from an atomic counter instead of | ||
| 72 | binding =:0= and closing it, which raced once tests overlapped — | ||
| 73 | =waitForPort= only asks whether something is listening, so the loser | ||
| 74 | talked to another test's server. | ||
| 75 | - Locally on 12 cores the suite goes from 637s to two or three minutes. | ||
| 76 | The =-timeout= floor the harness enforced (#143) is gone with the | ||
| 77 | reason for it; =make test= and =.gitbay/ci.yml= still pass their own | ||
| 78 | as a ceiling for a real hang. | ||
| 79 | |||
| 64 | * v1.33.0 — 2026-09-21 | 80 | * v1.33.0 — 2026-09-21 |
| 65 | 81 | ||
| 66 | Five findings from the outside review of the web UI (#241–#245). | 82 | Five findings from the outside review of the web UI (#241–#245). |