Skip to content

refactor: enforce ≤600-line limit — split remaining oversized modules (no behavior change) #176

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ltmoerdani merged 7 commits into from
Aug 22, 2026
Merged

refactor: enforce ≤600-line limit — split remaining oversized modules (no behavior change) #176

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
4c7c9f6
refactor(models): extract static data tables to modelTables.ts
Fahad090NP Aug 21, 2026
5fb4583
refactor(usage): split dashboard.ts into responsibility modules
Fahad090NP Aug 21, 2026
8aecdfc
refactor(usage): split tracker.ts into types/windows/summary modules
Fahad090NP Aug 21, 2026
03029a6
refactor(provider): split OpenCodeProvider.ts into focused modules
Fahad090NP Aug 21, 2026
b0a9bc7
refactor(test): split goUsageTracker tests into focused suites with s…
Fahad090NP Aug 21, 2026
b03b3f7
docs: log file-size refactor wave in devlog and changelog
Fahad090NP Aug 21, 2026
db89613
refactor(provider): rebase on main (byte-cap scaling in chatPrep) + r…
Fahad090NP Aug 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
79 changes: 50 additions & 29 deletions ARCHITECTURE-MAP.md
View file
Open in desktop

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@ All notable changes to the **OpenCode Go BYOK Provider** extension are documente
- `src/provider/` — `OpenCodeProvider` class, `definitions` (PROVIDERS table + model types), `messages`/`tokens` (message conversion + token estimation), `settings` (schema/getSettings/limits/capabilities), `visionProxy`.
- `src/models/` — `metadata`, `modelLimits`, `modelCapabilities`, `modelNames`, `pricing`, `metadataFetcher` (models.dev cache).
- `src/commands/` — provider, agent-window, diagnostics and thinking-picker command handlers; `src/request/headers.ts` for the OpenCode request headers.
- **`[Internal]` Remaining oversized files split to enforce the ≤600-line limit (no behavior change).** `src/provider/OpenCodeProvider.ts` (1259 lines) now delegates to `chatPrep.ts` (request preparation: message conversion, vision proxy, history trimming, budgets), `modelInfo.ts` (`provideLanguageModelChatInformation` assembly), `modelList.ts` (`ModelListFetcher` with retry/cache), `transportLog.ts` (`TransportSummaryLog`), `providerDialogs.ts` (Manage/Test Connection flows) and `providerUtils.ts`; the class itself is 584 lines. `src/usage/tracker.ts` is split into `trackerTypes.ts`, `trackerWindows.ts` and `trackerSummary.ts` (pure summary builders taking an injected context). The goUsageTracker tests are split into focused suites sharing `src/test/helpers/goUsageTestUtils.ts`.
- `extension.ts` is now a thin entry (~400 lines) that only wires activation + command registration. The two compat barrels (`streaming.ts`, `goUsageTracker.ts`) are removed and every importer references canonical paths. All behavior-preserving — verified by `npm run compile` + 291 unit tests + mock-server retry E2E (`npm run test-retry`) + `npm run lint`.

- **`[Internal]` Data-driven model registry (`src/core/registry.ts`).** The transport router and the thinking-family detector previously each owned a hardcoded model-prefix table. Both now read ONE data-driven table: `MODEL_REGISTRY` rows map model-family patterns → `{ endpointKind, sdkPackage, thinkingFamily, vendors? }`. `resolveModelRouting()` honors per-vendor restrictions (e.g. MiniMax `m2.x` → Messages on Go, Gemini → Google on Zen); `thinkingFamily()` reads the same table vendor-agnostically. Adding a new model family = adding one row (+ optionally a thinking strategy class). Context limits / capabilities stay metadata-driven (live models.dev) rather than duplicated in a static table. Behavior-preserving — verified by 14 new registry tests (305 total).
Expand Down
1 change: 1 addition & 0 deletions docs/devlog.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1974,6 +1974,7 @@ rg -n "sk-[A-Za-z0-9]|apiKey.*[A-Za-z0-9]{20,}|Authorization: Bearer [A-Za-z0-9]
| Date | Version | Summary |
| ---------- | --------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| 2026-08-21 | docs | Docs sync for PR wave #157–#172: created issues 72 (PR #164 think-tag force-strip) + 73 (#162 GLM 5.3) + 74 (PR #157 bug-hunt) + features 18 (Muse Spark 1.2); features/02 + issues/36 updated; 3 missing CHANGELOG [Unreleased] entries added (PR #157, #160, #164); devlog refreshed. Open PRs #161, #170 tracked. |
| 2026-08-21 | refactor/file-size-limit | Completed the ≤600-line file-size policy — split the remaining oversized modules with zero behavior change: `src/provider/OpenCodeProvider.ts` (1259→584) into `chatPrep`, `modelInfo`, `modelList`, `transportLog`, `providerDialogs`, `providerUtils`; `src/usage/tracker.ts` (989) into `trackerTypes`/`trackerWindows`/`trackerSummary`; `src/test/goUsageTracker.test.ts` (904) into focused suites + `test/helpers/goUsageTestUtils`. Full lint gate green after each split. 3 commits on `refactor/split-oversized-files`. |
| 2026-08-13 | refactor/thinking-request-modules | Thinking refactor (per-provider strategy classes + single VS Code per-model config authority + removed globalState shadow + `effectiveModelId`) + request module split (`src/request/`) + Windows lint fixes (`.cmd` shims + `.gitattributes` LF). 6 commits. CHANGELOG [Unreleased] updated. |
| 2026-06-13 | docs | Deep audit — all 4 🟢 Active docs verified against codebase + git history + CHANGELOG. All marked ✅ Solved: issue #19 (PR #15 merged), references #01 (research complete), architecture #01 (living ref complete), issue #01 (all code fixed v0.1.9/v0.1.10, remaining tool-call loop is model behavior not code bug). 0 Active docs remain. |
| 2026-06-13 | docs | Rewrote devlog into work-context format and flagged unrelated `WORK-CONTEXT.md` content |
Expand Down
134 changes: 4 additions & 130 deletions src/models/metadata.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
import { GO_VENDOR, ZEN_VENDOR, type ProviderVendor, type AllProviderVendor } from "../providerTypes";
import { MODEL_LIMITS_BY_PROVIDER, MODELS_WITHOUT_TEMPERATURE, VISION_CAPABLE_MODELS } from "./modelTables";

export interface BaseModelLimits {
contextWindow: number;
Expand Down Expand Up @@ -160,140 +161,13 @@ const MODELS_DEV_PROVIDER_BY_VENDOR: Record<ProviderVendor, keyof ModelsDevRespo
[ZEN_VENDOR]: "opencode",
};

const MODEL_LIMITS_BY_PROVIDER: Record<ProviderVendor, Record<string, BaseModelLimits | undefined>> = {
[GO_VENDOR]: {
// OpenCode Go caps completion at 131072 even though models.dev still lists
// output: 384000 — the gateway rejects larger values with HTTP 400 (#171).
"deepseek-v4-flash": { contextWindow: 1000000, maxOutputTokens: 131072 },
"deepseek-v4-pro": { contextWindow: 1000000, maxOutputTokens: 384000 },
"mimo-v2.5": { contextWindow: 1000000, maxOutputTokens: 128000 },
"mimo-v2.5-pro": { contextWindow: 1048576, maxOutputTokens: 128000 },
"mimo-v2-omni": { contextWindow: 262144, maxOutputTokens: 128000 },
"mimo-v2-pro": { contextWindow: 1048576, maxOutputTokens: 128000 },
"kimi-k2.7-code": { contextWindow: 256000, maxOutputTokens: 262144 },
"kimi-k2.6": { contextWindow: 262144, maxOutputTokens: 65536 },
"kimi-k2.5": { contextWindow: 262144, maxOutputTokens: 65536 },
"glm-5.1": { contextWindow: 202752, maxOutputTokens: 32768 },
"glm-5": { contextWindow: 202752, maxOutputTokens: 32768 },
"minimax-m3": { contextWindow: 512000, maxOutputTokens: 131072 },
"minimax-m2.7": { contextWindow: 204800, maxOutputTokens: 131072 },
"minimax-m2.5": { contextWindow: 204800, maxOutputTokens: 65536 },
"minimax-m2.1": { contextWindow: 204800, maxOutputTokens: 131072 },
"minimax-m2": { contextWindow: 204800, maxOutputTokens: 131072 },
"qwen3.7-max": { contextWindow: 1000000, maxOutputTokens: 65536 },
"qwen3.7-plus": { contextWindow: 262144, maxOutputTokens: 65536 },
"qwen3.6-plus": { contextWindow: 262144, maxOutputTokens: 65536 },
"qwen3.5-plus": { contextWindow: 262144, maxOutputTokens: 65536 },
"gpt-5.6-luna": { contextWindow: 1050000, maxOutputTokens: 128000 },
"hy3-preview": { contextWindow: 256000, maxOutputTokens: 64000 },
"muse-spark-1.2-contributor": { contextWindow: 1048576, maxOutputTokens: 131072 },
},
[ZEN_VENDOR]: {
"claude-opus-4-7": { contextWindow: 1000000, maxOutputTokens: 128000 },
"claude-opus-4-6": { contextWindow: 1000000, maxOutputTokens: 128000 },
"claude-opus-4-5": { contextWindow: 200000, maxOutputTokens: 64000 },
"claude-opus-4-1": { contextWindow: 200000, maxOutputTokens: 32000 },
"claude-sonnet-4-6": { contextWindow: 1000000, maxOutputTokens: 64000 },
"claude-sonnet-4-5": { contextWindow: 1000000, maxOutputTokens: 64000 },
"claude-sonnet-4": { contextWindow: 1000000, maxOutputTokens: 64000 },
"claude-haiku-4-5": { contextWindow: 200000, maxOutputTokens: 64000 },
"deepseek-v4-flash-free": { contextWindow: 200000, maxOutputTokens: 128000 },
"gemini-3.5-flash": { contextWindow: 1048576, maxOutputTokens: 65536 },
"gemini-3.1-pro": { contextWindow: 1048576, maxOutputTokens: 65536 },
"gemini-3-flash": { contextWindow: 1048576, maxOutputTokens: 65536 },
"glm-5.1": { contextWindow: 204800, maxOutputTokens: 131072 },
"glm-5": { contextWindow: 204800, maxOutputTokens: 131072 },
"gpt-5.5": { contextWindow: 1050000, maxOutputTokens: 128000 },
"gpt-5.5-pro": { contextWindow: 1050000, maxOutputTokens: 128000 },
"gpt-5.4": { contextWindow: 1050000, maxOutputTokens: 128000 },
"gpt-5.4-pro": { contextWindow: 1050000, maxOutputTokens: 128000 },
"gpt-5.4-mini": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.4-nano": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.3-codex": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.3-codex-spark": { contextWindow: 128000, maxOutputTokens: 128000 },
"gpt-5.2": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.2-codex": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.1": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.1-codex": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.1-codex-max": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5.1-codex-mini": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5-codex": { contextWindow: 400000, maxOutputTokens: 128000 },
"gpt-5-nano": { contextWindow: 400000, maxOutputTokens: 128000 },
"grok-build-0.1": { contextWindow: 256000, maxOutputTokens: 256000 },
"kimi-k2.6": { contextWindow: 262144, maxOutputTokens: 65536 },
"kimi-k2.5": { contextWindow: 262144, maxOutputTokens: 65536 },
"minimax-m2.7": { contextWindow: 204800, maxOutputTokens: 131072 },
"minimax-m2.5": { contextWindow: 204800, maxOutputTokens: 131072 },
"minimax-m2.5-free": { contextWindow: 204800, maxOutputTokens: 131072 },
"qwen3.6-plus": { contextWindow: 262144, maxOutputTokens: 65536 },
"qwen3.6-plus-free": { contextWindow: 262144, maxOutputTokens: 65536 },
"qwen3.5-plus": { contextWindow: 262144, maxOutputTokens: 65536 },
"trinity-large-preview-free": { contextWindow: 131072, maxOutputTokens: 131072 },
"nemotron-3-super-free": { contextWindow: 204800, maxOutputTokens: 128000 },
"big-pickle": { contextWindow: 200000, maxOutputTokens: 128000 },
"muse-spark-1.2-contributor-free": { contextWindow: 1048576, maxOutputTokens: 131072 },
},
};

/**
* Models where the `temperature` request parameter is unsupported / deprecated.
* The upstream API rejects any non-default temperature with HTTP 400 for these.
* Mirrors the models.dev `temperature: false` flag so the extension still
* omits temperature even when the live metadata fetch fails.
*/
const MODELS_WITHOUT_TEMPERATURE = new Set([
// Kimi K2.7-code: Moonshot API returns "invalid temperature: only 1 is allowed"
"kimi-k2.7-code",
]);

export function isFreeModel(modelId: string): boolean {
return FREE_ZEN_MODEL_IDS.has(modelId) || modelId.endsWith("-free");
}

export const VISION_CAPABLE_MODELS = new Set([
"minimax-m2.7",
"minimax-m2.5",
"minimax-m2.5-free",
"kimi-k2.7-code",
"kimi-k2.6",
"kimi-k2.5",
"glm-5.1",
"glm-5",
"mimo-v2.5",
"mimo-v2.5-pro",
"mimo-v2-omni",
"mimo-v2-pro",
"claude-opus-4-7",
"claude-opus-4-6",
"claude-opus-4-5",
"claude-opus-4-1",
"claude-sonnet-4-6",
"claude-sonnet-4-5",
"claude-sonnet-4",
"claude-haiku-4-5",
"gemini-3.5-flash",
"gemini-3.1-pro",
"gemini-3-flash",
"gpt-5.5",
"gpt-5.5-pro",
"gpt-5.4",
"gpt-5.4-pro",
"gpt-5.4-mini",
"gpt-5.4-nano",
"gpt-5.3-codex",
"gpt-5.2",
"gpt-5.2-codex",
"gpt-5.1",
"gpt-5.1-codex",
"gpt-5.1-codex-max",
"gpt-5.1-codex-mini",
"gpt-5",
"gpt-5-codex",
"gpt-5-nano",
"grok-build-0.1",
"gpt-5.6-luna",
]);
// Static per-model data tables live in `modelTables.ts`; re-exported here so
// existing import paths keep working.
export { VISION_CAPABLE_MODELS };

export function isFreshModelMetadata(snapshot: CachedModelMetadataSnapshot): boolean {
return Date.now() - snapshot.fetchedAt < MODEL_METADATA_CACHE_TTL_MS;
Expand Down
Loading
Loading