Document the package hierarchy and finalize the RFC

Add a README to each group dir (core/llm/bash/session-persistence/ui/
support) stating its role and product-vs-support classification, and
rewrite packages/README.md around the hierarchy (group table, grouped
"what goes where", removed the package-hierarchy FIXME).

Move the package-hierarchy RFC to implemented/architecture/ and rewrite
it to describe what shipped (placement rationale, the paths-wildcard and
publint dedup, the two new guardrail gates). Fold the remaining
tsconfig.build.json references dedup into the discover-package-inventory
proposal and fix its cross-link.

Update AGENTS.md: regrouped repo-layout map, depth-2 globs, the new
verify-package-paths gate in the doc-sync listing, and a note that we
lean toward stricter lint in the agentic-coding era (machine-caught
errors and a consistent foundation outweigh the one-time cost).
This commit is contained in:
Tianyi Cui
2026-06-20 23:25:33 +08:00
parent ca2207e26c
commit 5024cd5757
13 changed files with 219 additions and 117 deletions

View File

@@ -1,59 +0,0 @@
# RFC: Reorganize packages into a modular hierarchy
Status: proposed
## Problem
`packages/` is flat. Core product packages, provider integrations, capability seams, example UI support, and snapshot-only replay support all sit at the same level and look equally foundational. The [package README](../../../../packages/README.md) already has a `FIXME(package-hierarchy)` noting that `ui-stdio` and `llm-replay` were extracted from examples mostly for reuse and coverage. The flat layout makes support packages appear more product-shaped than they are and forces publish/lint/doc scripts to encode intent through comments or static lists.
This is not just cosmetic. A package's location currently says little about whether it is core API, a swappable capability, an adapter integration, an example harness helper, or test infrastructure. That makes future removal harder because every top-level package looks like part of the same public surface.
## Proposal
Move packages into a deliberate hierarchy under `packages/`. The exact layout is deferred to the implementing PR, but it should group packages by modular role rather than keep every package at one flat level.
One plausible shape:
```text
packages/
core/
session/
system-prompt/
tools/
agent/
agent-loop/
invariants/
llm/
llm/
adapters/
llm-deepseek/
llm-pi-ai/
bash/
bash/
bash-local/
tool-bash/
session-persistence/
session-persistence/
session-persistence-jsonl/
session-persistence-sqlite/
acp/
support/
ui-stdio/
llm-replay/
```
The final implementation may choose different names or groupings, but it should keep the same intent: core APIs, package families such as LLM/bash/session persistence, standalone integrations such as ACP, and support/test/example packages are distinguishable from the filesystem alone. Npm package names can stay `@deepseek-ai/dsh-*`; the hierarchy is about repo structure and maintenance policy, not public package renaming.
This proposal does not delete `llm-replay` or `ui-stdio` by itself. It makes their status honest: either they graduate into product packages with documented consumers, or they live under a support/testing/example classification where release and compatibility expectations are lower.
## Acceptance criteria
- Packages move from the flat `packages/<name>/` layout into a documented modular hierarchy.
- The implementing PR chooses the exact hierarchy and updates workspace globs, TypeScript paths, package docs, generated module graphs, `cordis.yml` package paths, build scripts, and publish/lint scripts in one coordinated move.
- Scripts that publish, lint publishability, or generate package inventories use the hierarchy instead of an ad hoc static list where the hierarchy is enough to express the policy.
- Docs explain which package groups are part of the product API and which groups are support/test/example infrastructure.
- New package guidance tells authors where to place a package and discourages new one-off top-level groups.
## What we give up
The restructure churns imports, workspace globs, docs links, and package paths. That churn is acceptable pre-release if it prevents the flat layout from fossilizing support packages as product contracts.

View File

@@ -4,20 +4,22 @@ Status: proposed
## Problem
Package and gate inventories are repeated by hand. [scripts/publint-all.ts](../../../../scripts/publint-all.ts) has a static list of publishable packages. The [package cookbook](../../../cookbook/adding-a-package.md) tells authors to update several files. The [package README](../../../../packages/README.md) carries a hand-written dependency graph. [CI](../../../../.github/workflows/ci.yml) and [development docs](../../../development.md) can drift from the actual `doc-sync` subcommands when new gates are added. These lists are small today, but every new package or gate creates another manual synchronization point.
Package and gate inventories are repeated by hand. The [package cookbook](../../../cookbook/adding-a-package.md) tells authors to update several files. The [package README](../../../../packages/README.md) carries a hand-written dependency graph. [CI](../../../../.github/workflows/ci.yml) and [development docs](../../../development.md) can drift from the actual `doc-sync` subcommands when new gates are added. `tsconfig.build.json` lists all 18 packages as explicit project `references`. These lists are small today, but every new package or gate creates another manual synchronization point.
The [package hierarchy](../../implemented/architecture/2026-06-20-package-hierarchy.md) already removed several of these by hand: `scripts/publint-all.ts` now derives its list from the `packages/<group>/<pkg>` layout, and the two `tsconfig` `paths` maps collapsed to one `@deepseek-ai/dsh-*` wildcard. What remains is the inventory that cannot be globbed away — chiefly `tsconfig.build.json`'s project `references`, which TypeScript requires as an explicit array (no wildcard form).
Static lists are appropriate when they encode policy; they are needless friction when they duplicate manifest data or layout facts that already exist in `package.json`, workspace globs, or the package hierarchy.
## Proposal
Make package/gate inventories discoverable. Publishability should come from the deliberate [package hierarchy](../architecture/2026-06-20-package-hierarchy.md) plus package manifests, not from a static array in a script or the npm `private` flag. Module graph generation should read package manifests. `doc-sync` should be the one command that defines and prints its sub-gates, with docs linking to that command rather than restating a second list.
Make the remaining package/gate inventories discoverable. A single canonical source — the `packages/<group>/<pkg>` hierarchy plus package manifests — should drive `tsconfig.build.json`'s `references`, the module graph, and any other full-package list, with a generate-and-verify step (the existing `gen-module-graph` / `gen-cordis-catalog` pattern: a generator writes the artifact, a `--check` mode in `hygiene`/`doc-sync` fails on a stale committed copy). Module graph generation already reads package manifests. `doc-sync` should be the one command that defines and prints its sub-gates, with docs linking to that command rather than restating a second list.
The hierarchy does not need to encode every fact about a package, but it should encode the broad maintenance policy: core/product packages, integrations, capability seams, and support/test/example packages should not all require a hand-maintained exception list before scripts can tell them apart.
## Acceptance criteria
- `publint-all` discovers publishable packages from the hierarchy plus manifests instead of a hard-coded array.
- Adding a package does not require editing a static package list for every gate.
- `tsconfig.build.json` project `references` are generated from the hierarchy (a generator emits them; a `--check` gate fails when the committed copy is stale), rather than hand-maintained.
- Adding a package does not require editing a static package list for any gate.
- Docs describe the source of truth rather than repeating generated inventories.
- CI invokes the aggregate commands and lets those commands own their sub-gate lists.