Commit Graph
2835 Commits
Author SHA1 Message Date
jschoubben c81915b039 G findings: distribution-tools and package-registry make the held writer at once and start it after Serve, as it seeds at Start
Builds: design 47 G A-40, A-32
2026-11-11 22:19:22 +01:00
jschoubben 839147b146 G findings: artifact-store-tools and mesh-config read what their held writer holds, so a restart neither drops nor rewrites their findings
Their in-memory open sets go.

Builds: design 47 G A-32, A-29, A-38
2026-11-11 22:17:21 +01:00
jschoubben 652493e2ac G: the package seats' whole test plays the runtime's put and get sessions
Builds: design 47 G A-21, §518, §490

Claude-Session: https://claude.ai/code/session_01NcbgfGKaxw984XyAPfLm4h
2026-11-11 22:15:42 +01:00
jschoubben fc83f204f4 G findings: the held writers of the stdio bundles are made once the runtime has its channel
The held writer reads the keys of its state when it is made, so it is made after Serve answers state calls; the keeper says a finding is not said before it exists.

Builds: design 47 G A-32, A-29, A-37
2026-11-11 22:15:42 +01:00
jschoubben 9f4d4eecba G: the holder's test of a part sent again says what the receiver does since design 51 (chunk 0 again starts the part afresh, a chunk out of sequence is refused)
Builds: design 47 G A-21
2026-11-11 22:12:21 +01:00
jschoubben 18fd2a32bf G findings: distribution and gitea's package registry write through the held writer, so a restart neither drops nor rewrites their findings
Their in-memory open sets go.

Builds: design 47 G A-32, A-29, A-37
2026-11-11 22:11:12 +01:00
jschoubben 8255640d00 G findings: claude-code, mesh-provisioner, distribution, mesh-config and gitea declare the writer's three events
The writer listens to mesh-conditions.findings-asked, so each also consumes it, as the modules that moved before.

Builds: design 47 G A-30, §383, §686

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 22:10:19 +01:00
jschoubben 3215e5eeb9 G findings: distribution and gitea's package registry delete their old finding keys once at start
Builds: design 47 G A-31, A-14, A-22

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 22:08:33 +01:00
jschoubben 1cc67aa9d7 G findings: mesh-config's and mesh-vault's own followers go through a views.Reader holding the writer, adopted
Builds: design 47 G A-28, A-26, A-22, §687

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 22:07:02 +01:00
jschoubben 06ebd7e8cf G findings: the writers name the ask's subject from the contract; the readers take the writer whole
WriterConfig.Asked is the generated mesh-conditions contract's SubjectFindingsAsked (A-17), a reader's Findings is the
module's writer itself (A-22, A-26), and mesh-build-go's single manifest ask writes and clears its first-view-unanswered
finding through the writer with the contract's Finding (A-28). mesh-config and mesh-vault's own followers (A-28) are not
yet moved onto a views.Reader: they do not compile at mesh-sdk 24f876a.

Builds: design 47 G §687, §689, A-17, A-22, A-26, A-28

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 22:03:09 +01:00
jschoubben 95f29d6c54 G findings: the modules with findings of their own write them through mesh-sdk's writer
artifact-store-tools (image-missing, sweep-refused), distribution (program and tools bundle, one writer each over its own
kinds), gitea's package-registry (and the token-refused key the TypeScript bundle watches) and mesh-config keep their
findings open and write and clear them through the generic writer; mesh-vault states its assignments view's. A part of an
id told apart by a ref is in the id (<id>.<ref>), each escaped; the writer starts once Serve has the runtime's channel; the
removed findings helpers and resolved words go.

Builds: design 47 G §686, §687, §689, A-3, A-14, A-16, A-22

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 21:58:08 +01:00
jschoubben dd4449de61 G findings: the view-only modules write their reader's findings through mesh-sdk's stdio writer
nfs-server, minio object-store, claude-code, mesh-watcher, mesh-provisioner and mesh-build-go hand their views reader
views.WriterFindings over views.NewStdioWriter and start it; the removed meshconditionsclient.FindingWriter goes.

Builds: design 47 G §686, §687, §689, A-3, A-16, A-17

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 21:46:49 +01:00
jschoubben b90d3a4556 The delivery gate names the ledger's fill answer as the sdk has it: buscalls.Fill
The sdk (02bea66) answers the ledger's fill call as one buscalls.Fill; the gate's LedgerFilled is a name that is gone.

Builds: design 47 G §462

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 21:26:20 +01:00
jschoubben 9c1cb63c3b mesh-provisioner uses the catalog key's codec meshcatalogkey, as meshcatalogclient's wrappers go
Builds: design 47 G §670

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 21:18:40 +01:00
jschoubben fa0f851f81 mesh-build-go, mesh-build-image and mesh-build-node read a step key's node as SD4 has it: ParseStepKey answers asker, node and id
Builds: design 47 G §484, §649

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 21:16:18 +01:00
jschoubben 65570636b2 Merge branch 'feat/47-G-conditions' into feat/47-G-dev2-join 2026-11-11 21:06:13 +01:00
jschoubben 1af839fb72 COND9 (moved-stream half): the keeper keeps a moved part's conditions, drops the old source and asks once on the new stream
Builds: design 47 G §663, §690

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 20:44:44 +01:00
jschoubben ecf597adce COND11 (in part): the keeper's ask bound and first-view bound are settings read at its start; the two hours wait on COND7
Builds: design 47 G §557

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 20:32:33 +01:00
jschoubben 811b59b454 COND9: the keeper takes a part whose finding stream the bus module made as a source (the moved-stream half stops on a question)
Builds: design 47 G §515, §663

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 20:27:47 +01:00
jschoubben 8f4f70f088 COND6: the keeper follows one catalog view per watched branch and adopts a branch registered while it runs
Builds: design 47 G §365, §479, §521, §532

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 20:26:40 +01:00
jschoubben ef60e5d09e COND2 (whole): the keeper's whole test starts with a kept source and no view of the catalog half
Builds: design 47 G §388

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 15:49:13 +01:00
jschoubben 26c82a507a OBS4: artifact-store-tools, mesh-watcher, minio and nfs-server ask findings and finding-events in requires
Builds: design 47 G §529, §387

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 15:48:35 +01:00
jschoubben 76be864bff HEAL2 (manifest only): mesh-healers asks findings and finding-events in requires; the writer's use waits on the contracts' finding subjects
Builds: design 47 G §529, §384, §461

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 15:48:35 +01:00
jschoubben 9e2311029a COND2: the keeper keeps its sources in the state findings-views and hears them before its first view of the catalog half
Builds: design 47 G §388, §365

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 15:46:35 +01:00
jschoubben bea1c446ea NATS10: the bus module serves no restore point tool and keeps no state kept
Reverts bbd9f638, c2e017a4 and b6884796; cmd/mesh-nats-snapshot/restore.go is as before them. The manifest test
asserts the claim serves only the reading verbs and the keeper's state is asked, requests, spent.

Builds: design 47 G §554, §555

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 15:45:03 +01:00
jschoubben 1e55ea8e4e Merge branch 'feat/47-G-build-conditions' into feat/47-G-conditions 2026-11-11 15:44:22 +01:00
jschoubben 9f62e0a2ba BLD6: the real unit level's broken fixture fails in its tests alone, go vet being the lint step's
Builds: design 47 G §328, §417

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:59:16 +01:00
jschoubben 579cd33cd5 GATE2: the whole-module test counts the merges handed in the state merges, the merge as an owed act being left out of the join
Builds: design 47 G §160, tests 34, 39 (eb131beb left out, G §167, §542)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:55:49 +01:00
jschoubben d3286ad0eb BLD3, BLD4: the joined step test carries no names list, the lint's fixed-name rule being taken out
Builds: design 47 G §7, §328, §329

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:53:43 +01:00
jschoubben a7b83b27f0 The issue tracker links a research effort by its number (milestone 47/G, W15.2)
The trace from research to the running build is links all the way (design 47).

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from dd68c96d.)
Builds: design 47 G §318 (ISS1)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:48:07 +01:00
jschoubben 4839884d93 A decision links the milestone that builds it: link kind built-in (milestone 47/G, W15.2)
ADR 0493 section 4 names the links; the knowledge base took no milestone as a target.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 5f4e4d29.)
Builds: design 47 G §318 (KB1)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:48:07 +01:00
jschoubben 6d4ad2c249 Test that the operator's Approve says accepted, by whom and on which warrant (issue 595, W15.1)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 65c6ca40.)
Builds: design 47 G §319, test 59 (KB1)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:48:07 +01:00
jschoubben 6f7d070769 M1-M5: the router holds back the asker's oldest ask behind a fourth and offers it again when one ends; the seat operator-channel's ask-again offers it now (milestone 47/G, issue 1966)
Claude-Session: https://claude.ai/code/session_01JG73KK8HD1gufnqyFxMFdW

ask-again is the seat operator-channel's verb (key in; state or error out), not the module tool messenger_ask_again, and records no hand act (G §394, §450).
(Joined from 681975ec.)
Builds: design 47 G §40, §83, §403 (RTR1, RTR2)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:48:07 +01:00
jschoubben 85023b3b52 gate_record_repair asks the owners' read verbs for the outcomes a record's node entries lack and applies them as the consumer would (milestone 47/G, gate_record_repair)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 46d3053c.)
Builds: design 47 G §159 (GATE13)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:45:57 +01:00
jschoubben 8484d349aa The gate says record-opened when it opens a delivery record (milestone 47/G, W9.3)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 649fa9b4.)
Builds: design 47 G §304, §305 (GATE)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:45:50 +01:00
jschoubben 1f1aa70580 The gate's whole-module test traces a delivery in one record and starts the gate again over the same bus and ledger: the record whole, nothing handed or asked again (milestone 47/G, W11.16: acceptance 34, 39)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from d162f899.)
Builds: design 47 G §160, tests 34, 39 (GATE2)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:45:50 +01:00
jschoubben 46c125c07c The gate keeps no mark of an own ask of a check: the dead field is gone and a row holds it (milestone 47/G, W11.8, issue 1794: its cause went with the gate's checks, issue 1318)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 70ded58f.)
Builds: design 47 G (main file G row: issues 1785, 1794) (GATE)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:45:50 +01:00
jschoubben 435d1206e4 A verdict, composed check or status of the same heads that changes what a refusal waited on starts the count of repeated refusals again (milestone 47/G, W11.7, issue 1785)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 80d401c4.)
Builds: design 47 G (main file G row: issues 1785, 1794) (GATE)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:45:50 +01:00
jschoubben 6fe1bba067 The gate hears the teller's answered and the view's reached, reads their answers and reach through their tools, and reads neither state (milestone 47/G, ruling 3: ADR 0596 §2-§4)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from ace22b9f.)
Builds: design 47 G §431 (the reads of the teller's answers and the view's reach are by their events); the commit's calls of teller_answers and view_reach are kept whole, the map names them no line

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:45:50 +01:00
jschoubben ed188268a0 The gate asks the ledger's calls and its fill of its machine's tool runner (milestone 47/G, W11.20 ruling)
The gate reads its records' ledger fill before it opens a record; its manifest asked no
capability, so it could run beside a runner answering neither. It now asks ledger-calls and
ledger-fill, the fill a capability of its own.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from c20044a2.)
Builds: design 47 G §181, §462, §488 (GATE12)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:44:20 +01:00
jschoubben bf6bb3d094 The gate names the issues linked in the issue tracker and the forge's author, and reads no description (milestone 47/G, W11.5)
ADR 0659 §4 kept three description reads that shape words only until they move to the bus: the
issue numbers the ask names, the author a stalled line names and the notice that after: lines are
not read. The ask now names the issues linked to each member's pull request, read once per head
through the seat issue-tracker's list (a tracker that does not answer is said, never waited on);
the author is the forge's pull request record's, from the seat git's events; the after: notice is
gone. The gate's forms of a pull request and a delivery hold no description field.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from ddd93a78.)
Builds: design 47 G §50, §143 (GATE15)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:44:17 +01:00
jschoubben 8ee28d08c3 The gate says its records' ledger filling, full and room, and holds a merge whose record finds no room (milestone 47/G, W11.20)
Design 47: a store filling is said before it is full. The gate reads its ledger's fill through the
tool runner's mesh/ledger.fill, says filling of the seat mesh-delivery when a record takes the ledger
past 80 % of its cap, refuses the next record when none fits and says full naming it, and keeps the
merge waiting for room on its delivery (held_full), so a restarted gate judges alike. Once the
ledger is below 80 % again, after the retention's removals, it says room and hands the merge.
gate_records says the fill against its cap, or why it is unread.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 061be00d.)
Builds: design 47 G §181 (GATE12)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:44:17 +01:00
jschoubben 6ac5c934d2 The gate holds no merge for a check: the hold, the turns, their stalled line and tools, and check-gates-merge go; the merge judges the operator's Approve alone (milestone 47/G, W11.14)
ADR 0661: the check gates the ask, in one place (askWaits); once the Approve covers
the heads (approvedWaits) no verdict, carry, composed check or required status is
judged, an Approve given before a trunk move merges onto the new trunk, and a
failure after the Approve is said once on the pull request and as a stalled line
naming the merge it follows. A check failing before the answer withdraws the ask.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 43696644.)
Builds: design 47 G §278, test 52 (GATE10)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:43:01 +01:00
jschoubben ee9c389888 The gate takes owed.ended and take-in.taken from its manifest's consumes and names neither the forge bridge nor the indexer in code (milestone 47/G, the names move)
The module's own manifest is embedded (package at the module's root); an event the gate hears from another module is
the consumes entry with that event's name after its owner. No new seat, no second publisher on the seat git (the
lead's ruling). Rows in the manifest test, which also reads the gate's sources for those names.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 2de87fd8.)
Builds: design 47 G §7, §146 (GATE16); the bridge's name stays in code for the old merge follower (eb131beb left out, G §167, §542)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:41:20 +01:00
jschoubben 59f9a00c64 A merge the forge made with no record's merge-pull opens a forge-made delivery record, said once as its transition with why forge-made (milestone 47/G, W11.17e)
The record holds the pull request at its head and has no Approve; the delivery keeps that it was recorded
(forge_recorded). Rows in issue 888's test. The test bridge never performs a switched act: the forge bridge decides.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 674f49ef.)
Builds: design 47 G §170 (GATE6)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:40:50 +01:00
jschoubben badfe0d166 The switch says every act the old states owed and merges hold under its switched id, so the forge bridge performs only what the old bridge did not (milestone 47/G, W11.17d carry)
Every key of owed, and the member merging in every key of merges, is said as owed under
meshdeliveryclient.SwitchedActID(state, key, revision), at every switch, the same act each time. The switch reads
the old states through one reader (OldState). Rows in the switch's test.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from dd0c4294.)
Builds: design 47 G §169 (GATE5, GATE6); only the owed carry: the member merging in the old state merges is not carried as a merge act (G §167, §542)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:40:34 +01:00
jschoubben 64ed69db9b The gate's groups answers the seat's groups: members in merge order, the acts said with their ids and the acts recorded with their outcomes, one group alone, and none for a group it does not know (milestone 47/G, W11.10; F-forge-7, F-forge-8)
The forge bridge reads a group's members and what it owes from here, never from the gate's state. The text form of
the members and composed_gate are gone with the answer of before.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from de368344.)
Builds: design 47 G §157, §304 (GATE5)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:37:37 +01:00
jschoubben c6cf8b64f1 The gate serves the seat mesh-delivery's verb source: each repository's trunk branch and head and the revision, read from source.main (milestone 47/G, source verb)
A node-engine reads the repositories here to call a reverted build's catch-up tool once per repository, never the
gate's state. Rows in issue 888's test and the bus test, which call the verb over the runtime.

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from b45fd0ab.)
Builds: design 47 G §15 in part (GATE7)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:37:33 +01:00
jschoubben 32528aa58f An act owed to the forge with no outcome 15 minutes after it was said is listed by stalled (milestone 47/G, W11.17b)
Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from 0fa400d5.)
Builds: design 47 G §166, §169 (GATE6)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:37:33 +01:00
jschoubben 00b95f1f68 Every act owed to the forge is said as the seat's event owed under one id and stays on its delivery's or group's state until its owed.ended is recorded (milestone 47/G, W11.17a)
At the gate's start every unanswered act is said again under the same id; an outcome is written into the record that
holds the pull request, then answered; an outcome other than performed is said; the checker's are not the gate's.
Needs the forge bridge's forgemsg owed.ended (feat/47-G-forge).

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ

(Joined from cf3715ea.)
Builds: design 47 G §162, §163, §164, §165, §166 (GATE5)

Claude-Session: https://claude.ai/code/session_01JG73KEq4zoDfM9WtYpcQgZ
2026-11-11 14:37:33 +01:00