Files
promptkit/docs/roadmap/implementation.md

387 lines
15 KiB
Markdown

# Local Backend Convenience Implementation Plan
**Status:** Ready for implementation.
## Purpose
This document is the decision-complete implementation plan for
[local backend convenience](local-backend.md). It is written for a coding
agent that will implement each stage in order.
The feature roadmap owns the motivation, consumer paths, policy choices,
compatibility requirements, non-goals, and target end state. This document
owns the exact proposed API, file-level changes, implementation sequence, test
ownership, documentation work, validation commands, and completion gates.
## Implementation Rules
- Complete the stages in order. Keep the root package compiling and its
focused tests passing at every stage boundary.
- Preserve unrelated working-tree changes. The feature roadmap may already be
uncommitted when implementation begins; retain it.
- Follow every policy under `docs/policy/`, the task-specific reading guide in
`docs/development.md`, and the accepted behavior in
`local-backend.md`.
- Keep the API in the root `promptkit` package. Do not add a public or internal
package for this feature.
- Implement the helper as a transparent constructor for the existing
`Backend` type. Do not add a second backend representation or bypass
`WithBackend`.
- Leave normalization, validation, copying, registry construction, capacity
defaulting, and duplicate detection in their existing owners.
- Do not add environment-variable discovery, package-global registration,
implicit local defaults, model selection, profile construction, or support
for non-OpenAI-compatible transports.
- Keep tests lean and behavior-focused. Do not duplicate the existing
registry and concurrency test matrices merely because the constructor
reaches those mechanisms.
- Update exact GoDoc with the exported declarations. Update current-state
consumer guidance only after the corresponding API is implemented.
- Do not update release notes, create a release, change a module version, or
tag a commit as part of this work.
## Fixed Design
### Exported API
Add these declarations to `backends.go` in package `promptkit`:
```go
// BackendLocal is the conventional ID used by LocalBackend. It is not a
// built-in or reserved backend and must be registered with WithBackend.
const BackendLocal = "local"
// LocalBackend returns a Backend for a conventional local OpenAI-compatible
// endpoint.
func LocalBackend(endpoint string, concurrencyLimit int) Backend {
return Backend{
ID: BackendLocal,
Endpoint: endpoint,
ConcurrencyLimit: concurrencyLimit,
}
}
```
The exact GoDoc may be wrapped or expanded for clarity, but it must own and
communicate all of these contract points:
- `BackendLocal` is the case-sensitive conventional ID `"local"`;
- it is neither built in nor reserved;
- calling `LocalBackend` does not register anything;
- the returned value must be supplied through `WithBackend`;
- `endpoint` and `concurrencyLimit` are copied into the corresponding fields
without normalization or validation;
- `APIKeyEnv`, `ExtraParams`, and `QueueCapacity` retain their zero values; and
- normal `NewEngine` backend validation and concurrency semantics apply after
registration.
Keep `BackendOpenRouter` unchanged. `BackendLocal` must not alias an internal
registry constant because the internal registry has no special local-backend
identity or behavior.
Place `BackendLocal` near `BackendOpenRouter`, and place `LocalBackend` after
the `Backend` declaration and before `WithBackend`. This keeps the conventional
IDs, configured value, convenience constructor, and registration option
discoverable in one file.
### Constructor Semantics
`LocalBackend` is a pure struct constructor. Its complete behavior is
equivalent to the keyed literal shown above.
In particular, the constructor must not:
- trim or parse the endpoint;
- reject blank endpoints or negative limits;
- choose a default limit;
- assign an API-key environment variable;
- allocate an empty `ExtraParams` map;
- assign a queue-capacity pointer;
- read process environment;
- mutate package-global or engine state; or
- call `WithBackend` itself.
Deferred validation is intentional. It keeps one validation path for all
`Backend` values: `WithBackend` copies the public value into engine options,
and `NewEngine` constructs the validated immutable registry. A positive limit
with a nil queue continues to select the existing default queue capacity of
1024; zero continues to mean unlimited; a negative value continues to make
`NewEngine` fail with `ErrInvalidConfig`.
The returned `Backend` remains an ordinary caller-owned value. Consumers can
modify it before passing it to `WithBackend`, although task-oriented
documentation should direct materially customized configurations to an
explicit keyed `Backend` literal.
### Identity And Compatibility
Do not add `"local"` to the internal built-in or reserved ID set. A consumer
must be able to register either:
```go
promptkit.LocalBackend(endpoint, limit)
```
or:
```go
promptkit.Backend{
ID: promptkit.BackendLocal,
Endpoint: endpoint,
}
```
through the existing `WithBackend` option. Both forms participate in ordinary
duplicate-ID detection. Existing consumers that already use the literal ID
`"local"` remain source- and behavior-compatible.
No existing `Backend`, `WithBackend`, profile, registry, execution-target, or
capacity semantics change. Do not modify `internal/backend`,
`internal/capacity`, `internal/domain`, `engine.go`, `profiles.go`, or
`types.go` for this feature.
### Test Ownership
The root external-package contract suite in `public_contract_test.go` owns the
new public behavior. Add one focused test named:
```go
func TestLocalBackendConstructsAndRegistersConventionalBackend(t *testing.T)
```
The test must:
1. call `promptkit.LocalBackend` with a test endpoint and a positive,
test-owned concurrency limit;
2. compare the returned value with this complete expected value:
```go
promptkit.Backend{
ID: promptkit.BackendLocal,
Endpoint: endpoint,
ConcurrencyLimit: limit,
}
```
A whole-value comparison is appropriate here because the exact zero-value
fields are part of this small public constructor's contract.
3. register that returned value with `WithBackend`;
4. add an in-memory profile whose `BackendID` is
`promptkit.BackendLocal`;
5. construct an engine through the existing contract-test prompt fixture;
6. call `Prepare`; and
7. assert that the prepared result exposes `BackendLocal` as the selected
backend and the supplied endpoint as the effective endpoint.
This single test protects the realistic compatibility risks: accidental field
defaults, a changed conventional ID, failure to compose with `WithBackend`,
and accidental treatment of `"local"` as reserved. It also demonstrates that
the helper uses the existing backend/profile path.
Do not add separate tests for blank endpoints, malformed endpoints, negative
limits, queue defaulting, duplicate IDs, engine isolation, runtime capacity,
or caller mutation. Those mechanisms are unchanged and already have tests at
their owning boundaries. Do not add internal-package tests for this root
facade constructor.
### Consumer Documentation
Update `docs/consumers/pkg-promptkit.md` after the API exists. Keep Go
declarations and GoDoc as the exact API owner; the guide should help consumers
choose a workflow and link to `backends.go` for precise semantics.
Restructure the backend guidance to present these paths in increasing order of
configuration:
1. **Endpoint-only profile.** Show a small in-memory `Profile` with
`Endpoint` and `Model`. Explain that this is the simplest choice when only
one profile needs the endpoint and shared backend identity or capacity
policy is unnecessary.
2. **Local convenience constructor.** Show
`WithBackend(promptkit.LocalBackend("http://localhost:8000/v1", 2))`
together with a profile using
`BackendID: promptkit.BackendLocal`. Explain briefly that the helper is
explicit, is not pre-registered, does not read environment variables, and
leaves the queue capacity at the existing default for a positive limit.
3. **Complete backend value.** Preserve an advanced example using a keyed
`Backend` literal for needs such as `APIKeyEnv`, an explicit
`QueueCapacity`, extra parameters, a custom ID, or multiple local
endpoints. Use a custom ID other than `"local"` in that example so the
distinction from the conventional helper is clear.
Keep the existing backend-routing, selected-backend identity, concurrency,
credential, and error guidance unless a small wording adjustment is required
to make the new decision path coherent. Avoid repeating the complete field
contract or registry validation rules from GoDoc.
Do not add a new maintained example, README section, format-reference entry,
integration-contract change, internal-document change, or release note. The
consumer-guide snippets are sufficient for this small convenience API, and
none of those other documents owns the affected task or contract.
## Stage 1: Add The Public Constructor And Contract Test
**Status:** Not started.
### Objective
Add the smallest public API that expresses the accepted local-backend
convention and protect its compatibility through the root public boundary.
### Implementation Prompt
1. Re-read `docs/development.md`, all files under `docs/policy/`,
`local-backend.md`, `backends.go`, the backend-related portion of
`public_contract_test.go`, and the existing backend/concurrency GoDoc before
editing.
2. Confirm the working tree and preserve the uncommitted roadmaps and any
unrelated consumer changes.
3. Add the untyped exported string constant `BackendLocal = "local"` to
`backends.go` without changing `BackendOpenRouter`.
4. Add `LocalBackend(endpoint string, concurrencyLimit int) Backend` to
`backends.go` using the exact keyed-literal implementation in the fixed
design.
5. Write complete GoDoc for both declarations. Make their conventional,
explicit, non-built-in, non-reserved, and deferred-validation semantics
unambiguous.
6. Add
`TestLocalBackendConstructsAndRegistersConventionalBackend` to
`public_contract_test.go` exactly as specified under Test Ownership. Reuse
the existing contract prompt fixture rather than adding a fixture or test
helper.
7. Do not edit internal packages. If the constructor appears to require an
internal change, stop and reconcile the implementation with the fixed
transparent-constructor design instead.
### Focused Validation
Run:
```sh
gofmt -w backends.go public_contract_test.go
go test . -run '^TestLocalBackendConstructsAndRegistersConventionalBackend$'
go test .
go vet .
go build .
git diff --check
```
Inspect the diff and confirm that this stage changes only `backends.go`,
`public_contract_test.go`, and the already-present roadmap files.
### Completion Gate
Stage 1 is complete when:
- the exported constant and constructor match the fixed API;
- the constructor returns only the three specified non-zero fields;
- the helper registers through the ordinary `WithBackend` path;
- `"local"` remains a valid consumer registration rather than a reserved
built-in;
- the focused public-contract test passes; and
- the root package test, vet, build, formatting, and whitespace checks pass.
## Stage 2: Publish Consumer Guidance And Validate The Repository
**Status:** Not started.
### Objective
Make the simplest suitable local-endpoint configuration easy to discover,
confirm the complete change across the repository, and close the temporary
roadmaps.
### Implementation Prompt
1. Re-read the implemented declarations and GoDoc before describing them.
2. Update `docs/consumers/pkg-promptkit.md` according to the three-path
structure under Consumer Documentation.
3. Keep examples illustrative, minimal, secret-free, and consistent with the
implemented declarations. Link precise semantics to `backends.go` rather
than duplicating its field-by-field contract.
4. Follow every added or changed Markdown link and confirm its target exists.
Confirm that all local repository-relative links in
`local-backend.md`, this plan, and the changed consumer guide resolve.
5. Run the complete maintainer validation sequence below.
6. Inspect the final diff for accidental internal behavior changes,
generated artifacts, credentials, local workspace files, or module
replacements.
7. After every completion gate passes, set both stage statuses, this plan's
status, and the status in `local-backend.md` to `Complete`. Do not retire or
remove the roadmaps in the implementation change; roadmap retirement
follows implementation review.
### Full Validation
Run the complete sequence from `docs/development.md`:
```sh
go test ./...
go test -race ./...
go vet ./...
go build ./...
go run ./examples/go-library/prepare
gofmt -l $(git ls-files '*.go')
git diff --check
```
The formatting command must produce no paths. Follow every added or changed
Markdown link and confirm its target and heading exist.
Also inspect:
```sh
git status --short
git diff --stat
git diff
```
Confirm that:
- production changes are limited to the root convenience API;
- no internal backend, registry, capacity, profile, or engine behavior
changed;
- `BackendLocal` is conventional and consumer-registerable, not built in or
reserved;
- `LocalBackend` performs no validation, normalization, environment lookup,
registration, allocation, or hidden default selection;
- the returned `Backend` leaves `QueueCapacity` nil so existing positive-limit
defaulting remains owned by the registry;
- existing endpoint-only profiles and complete custom backends remain
documented and supported;
- current-state documentation describes only the now-implemented API and
links to the canonical GoDoc for exact semantics;
- no `go.work`, `go.work.sum`, local module replacement, credential,
generated binary, or unrelated change was introduced; and
- the feature and implementation roadmaps contain no unresolved work marked
complete.
### Completion Gate
Stage 2 is complete when every target-end-state item in
`local-backend.md` is implemented, the consumer guide clearly presents all
three configuration paths, all complete validation commands pass, all changed
links resolve, and the roadmap statuses accurately report completion.
## Implementation Handoff
The implementation handoff should report:
- the new `BackendLocal` and `LocalBackend` public API;
- that the helper remains explicit and composes with the ordinary backend
registry;
- the focused public-contract coverage added;
- the consumer-guide decision path added;
- the complete validation commands and results; and
- any unrelated working-tree changes that were preserved.
Do not claim that a local backend is built in, pre-registered, configured from
environment variables, or assigned a default model or concurrency limit.
## Open Questions
None. The feature roadmap and this plan fix the exported API, constructor
semantics, identity treatment, compatibility behavior, test boundary,
documentation ownership, implementation sequence, validation gates, and
non-goals required for implementation.