Commit Graph
6 Commits
Author SHA1 Message Date
jcoffey-dev f44382fb09 DLP phase 3: hold for review
ci / fork-checks (pull_request) Successful in 33s
ci / build (pull_request) Successful in 9m58s
The hold action now holds (dlp-and-mail-flow-rules spec, §2.6), where
until now it blocked.

- At DATA a hold decision queues the message with its release a century
  off (the queue's future-release mechanism, so the stored format is
  unchanged and an older node just never sends it), transport rules
  still applied, and replies 250 Held for review. A review record under
  R/h + queue id keeps the sender, recipients, subject, size, rules and
  detector counts. The sender is told when the rule asks.
- smtp/queue/held.rs: release (each recipient due now, its next notice
  as far off as it was, its lifetime counted from the release), reject
  (removed from the queue, the sender told, with the reviewer's note),
  and expiry: the daily clean-up rejects what nobody reviewed in 7 days,
  recorded as the server's doing.
- inbuxa:HeldMessage get/set: the review queue, sysDlpReviewGet to list
  and read (preview, 64 KB of text, only when asked for and recorded as
  blobAccess), sysDlpReviewUpdate to release or reject, a reason
  required and audited by the request layer; no create or destroy;
  server-level only.
- Guards: Emails > Queue refuses to change or delete held mail; the
  sender can't unsend it.
- Privacy catalog entry for inbuxa:HeldMessage; spec §2.6 as built.

Tests: mail_rules_tests gains the whole flow (held and listed with
counts, sender notified and nothing delivered, queue and unsend
refused, preview recorded, reject needs a reason and tells the sender
the note, release delivers, expiry returns it, decisions audited with
reasons). smtp inbound, system_tests (after one BlobNotFound in
antispam, the known flake, then clean), features and common unit tests.
2026-09-28 18:33:12 -07:00
jcoffey-dev d9a6db025b Explain this: the local model reads delivery failures, verdicts, logs and settings
ci / fork-checks (pull_request) Successful in 48s
ci / build (pull_request) Successful in 9m4s
A new method, inbuxa:Explanation/set, asks the node's local model for a
short plain-words reading of one thing an administrator is looking at:
a failed recipient in the queue, a Classify verdict, a log line or trace
event, or one setting with its saved value. The server builds the prompt
itself from stored data and the registry schema, never from text the
console sends, and grounds SMTP replies in RFC 3463 and RFC 5321.

What the model is never shown: secrets (including ones nested inside a
setting, like an AI model's HTTP auth), raw protocol events, and the
contents of any other event. A tag name that doesn't have a tag's shape is
refused before a model is asked.

Calls share the AI gate with spam classification, but mail always keeps
its slot, and Explain has its own hourly count per account and its own
on/off switch in inbuxa:AiLimits. The permission is sysAiExplain,
superuser only; tenant administrators can't use it. The session carries
an aiExplain flag so a console knows when to offer the button.

An install whose roles were stored before the permission existed gets it
added once, at start-up, to the roles that are administrators' alone,
not the User role their defaults share with every account. An operator
who removes it later isn't overruled.

Tests: unit tests in inbuxa-features and jmap, and ai_explain_tests
(run with --ignored) covering the acceptance tests and the upgrade.
2026-09-26 00:52:51 -07:00
jcoffey-dev 6a53d47106 Mark the files this fork changed (AGPL section 5(a))
The AGPL asks a modified version to carry prominent notices saying it was
modified, and giving a date. Publishing the source is the conveyance that
asks for it, so it wants doing before the repository is public rather than
at the release.

Every upstream file the fork changed now says so in its header, beneath the
notice it came with: 164 files, found by diffing against the upstream
snapshot branch rather than by guessing, so the list is what actually
differs. Files the fork wrote itself already carry their own copyright and
need nothing. Upstream's notices are untouched, which its licence requires
and which was already true.

The README says the same thing in prose, since the obligation is on the
work as a whole and not only its Rust files.

Builds unchanged: the server and the test binary both compile.
2026-09-19 23:48:35 -07:00
jcoffey-dev 3896f720a1 Multi-tenancy: queue visibility (MT-5)
A tenant sees queued mail with a recipient on its domains, and mail its own
authenticated senders sent from them, in get, query, update and destroy.
2026-09-18 15:20:04 -07:00
jcoffey-dev c9c761fab5 Quick fixes from the first boot: recovery admin, upsell, warnings
- The recovery administrator (INBUXA_RECOVERY_ADMIN, or STALWART_RECOVERY_ADMIN)
  is honored only in bootstrap and recovery mode. On a configured server it's
  ignored with a startup warning. Before, it was a standing full-admin login
  for as long as the variable stayed set.
- The Enterprise upsell error is replaced by "This feature isn't available in
  INBUXA yet" for the features still to be rebuilt.
- Workspace warnings: 25 to 0. cargo fix removed the unused imports. The
  seven places where Enterprise code used to plug in keep their parameters,
  each with an inbuxa: comment naming the rebuild that uses it again. The
  antispam test's mock-server imports are back behind pending-rebuild.
2026-09-18 11:36:44 -07:00
jcoffey-dev 7dae9b29fd Import upstream v0.16.22, stripped
Upstream commit: 474dd0229cb20cf513036619781ed97bd8073c3f
Enterprise-only files removed or emptied: 63
Enterprise-only snippets removed: 117 in 50 files
Dangling module declarations removed: 5
Cargo edits turning enterprise off: 14
Verification: clean
Enterprise feature gates left for rebuilt features: 19 in 18 files

Produced by tools/fork/strip.py. The full report is in docs/fork/strip-reports/ on main.
2026-09-18 10:21:56 -07:00