alkdev/alkgit

0 Fork 0

Files

T

glm-5.3-flash 8f73da5d12 docs(architecture): phase 1 bootstrap — specs, 9 ADRs, OQ tracker

Architecture documentation structure per sdd_process phase 1:

- README index (doc table, ADR table, lifecycle), overview with crate
  map, dependency rules, and security invariants
- Component specs: storage, transport, http, ssh, alkgitd (all draft)
- ADRs 001-009: crate decomposition, front-door-blind core, V2-first
  protocol, pack pipeline (data::output generation / data::input
  ingestion), session substrate types, http adapter composition
  (proposed, OQ-01), ACL-before-advertisement, registry-resolved repo
  identity, bounded-resources budgets
- open-questions.md: OQ-01..08 with two deferred(scope), one
  deferred(unclear), door-type definitions, blocker tracker tasks in
  tasks/architecture/
- v1 ssh-door decision recorded: russh terminates wire SSH in alkgitd;
  alkcall channels stay the internal substrate (OQ-03 partially
  resolved)

Two review rounds (fresh-context subagent): 4 critical + 17 warnings
fixed in round one; zero critical + 4 warnings + 5 suggestions fixed in
round two. All ADR/OQ cross-references verified resolving.

2026-09-21 03:55:33 +00:00

3.8 KiB

Raw Blame History

ADR-003: Protocol V2-first with honest capability advertisement

Status

Accepted

Context

The git smart protocol has three versions in the wild: V0 (the original), V1 (capability line prefixed), and V2 (command-based, default since git 2.18+). POC-1 and POC-2 validated the full V2 fetch path against real git 2.43 end-to-end (advertisement, ls-refs, fetch with sideband packs) over both duplex and http substrates. V0/V1 stateless-http is a different framing (per-round POSTs, stateless ack re-derivation).

The research inventory (git-protocol.md §"Open items") left the V0/V1 question open: support V0/V1 on http? On ssh? Which first?

Decision drivers:

Our deployment anchor (vision §"Primary deployment target") is modern git clients over http and ssh; every git client from the last six years speaks V2.
Honest capability advertisement (vision principle 5) is easiest to keep honest with fewer served shapes.
V0/V1 stateless-http needs multi-round negotiation state re-derivation — meaningful complexity for a client population that is effectively extinct.

Decision

Serve protocol V2 first. alkgit v1 serves V2 on both doors:

Advertisement is once-per-session (duplex) / per-request-set (http), exactly as observed in POC-1 (re-advertising between commands hangs real git).
Advertised capabilities are exactly what we serve: ls-refs=unborn, fetch=wait-for-done (v1 fetch policy is full-closure pack on done; multi-round negotiation, OQ-02, may add capability values when it lands), object-format=sha1 (+ sha256 under the feature flag, if and when tested — OQ-05). Nothing unimplemented is advertised (shallow, filter, packfile-uris, object-info, server-option are declined by omission; POC-1 confirmed real git accepts this). On the push side, git-upload-archive is not served at all — ssh exec requests for it get a fixed refusal (ssh.md; ADR-008's never-execute rule).
HTTP requests protocol V2 only; ssh requests protocol V2 only (see below).

V0/V1 policy: decided now as V2-only for v1. Both doors (http and ssh) speak V2; clients that cannot speak V2 get a clear pkt-line error. Reversal is wire-visible (the advertised version set is a wire format), so treat this as effectively one-way once published — the decision is still made now, and revisiting it needs a new ADR plus a deprecation window for clients. The motivation stands: a concrete consumer needing V0/V1 (e.g. very old CI images) has not been identified. The V0/V1 receive-pack shape (used by git push on http stateless framing in older clients) is unaffected by this choice: receive-pack is mostly version-independent (see transport.md).

Sequencing note (fixes the scope contradiction the review found): the v1 fetch policy that ships first is full-closure-on-done (POC-validated). Multi-round V2 negotiation (haves/acks without done) is in v1's scope but lands after the done-path works end-to-end; its ack logic is OQ-02's open question, and any capability-advertisement change it requires happens then.

Consequences

Smallest honest surface; POC-validated shapes only.
Old-client support is declined explicitly (error, not silence).
The POC's observed done-path shape (no acknowledgments section, no ready, response ends at flush) is our normative wire behavior, not the docs' grammar — git-protocol.md §V2 records this.
If V0/V1 is ever added, the state machines in transport must be re-shaped around a version-dispatch at session start (single point, by design — see ADR-005).

References

docs/research/git-protocol.md §"Protocol surface inventory", §"Negotiation policy"
docs/research/poc-1-findings.md (advertisement-once, capability declination), POC-3 (http V2 framing)
ADR-005 (substrate types), ADR-007 (ACL before advertisement)
transport.md, http.md, ssh.md

3.8 KiB Raw Blame History