audit: stage 2c - the sender check reads From: as OpenDKIM does (owner's decision)
test / test (push) Canceled after 0s
test / test (push) Canceled after 0s
Owner, 2026-10-14, after reading OpenDKIM's source together (dkim_mail_parse, called in mlfi_eoh on the first From:; SenderHeaders is unset in build/opendkim.conf): "Да" to porting the parser so that the check decides by the address OpenDKIM will take. The scan of the display name and comments for another sender of the server - a guard against two parsers parting ways, and the reason L8 is half closed - then goes. New stage 2c (three lines) after 2.5.1 and before stage 3: the port from the version the image runs, checked against OpenDKIM on the stand with generated lines, then docs and wording. Stages 3 and 4 move to 2.5.3 and 2.5.4 (development.md's planned cuts). Plan 65 lines, roadmap 30/65. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
+4
-3
@@ -133,9 +133,10 @@ than the default «meaningful step» rule in [§ Commits and release build](#com
|
||||
| 20 | audit-2026-10-12 stage 0: release hygiene | `2.4.2` (PATCH) |
|
||||
| 21 | audit-2026-10-12 stages 1 and 2 | `2.5.0` |
|
||||
| 22 | audit-2026-10-12 stage 2b: the log tailer | `2.5.1` |
|
||||
| 23 | audit-2026-10-12 stage 3 | `2.5.2` |
|
||||
| 24 | audit-2026-10-12 stage 4 | `2.5.3` |
|
||||
| 25 | audit-2026-10-12 stage 5 | `2.6.0` |
|
||||
| 23 | audit-2026-10-12 stage 2c: the sender check reads From: as OpenDKIM does | `2.5.2` |
|
||||
| 24 | audit-2026-10-12 stage 3 | `2.5.3` |
|
||||
| 25 | audit-2026-10-12 stage 4 | `2.5.4` |
|
||||
| 26 | audit-2026-10-12 stage 5 | `2.6.0` |
|
||||
|
||||
Docs-only prep (checklists in plans, no product code) uses the same per-step
|
||||
commit rule but **no** version cut until the next product stage ships.
|
||||
|
||||
Reference in New Issue
Block a user