refactor(token-meter): simplify singleton service (round 1)
This commit is contained in:
Hypatia May
@@ -24,7 +24,7 @@ A harness is one [Cordis](cordis-primer.md) context. Packages contribute service
|
||||
| ctx key | Package family | Role |
|
||||
|---|---|---|
|
||||
| `ctx.llm` | [`llm/`](../packages/llm/README.md) | adapter registry and streaming model calls |
|
||||
| `ctx.tokenMeter` | [`llm/token-meter`](../packages/llm/token-meter/README.md) | replay-aware request/surface pressure per model |
|
||||
| `ctx.tokenMeter` | [`llm/token-meter`](../packages/llm/token-meter/README.md) | singleton replay-aware request/surface pressure |
|
||||
| `ctx.bash` | [`bash/`](../packages/bash/README.md) | foreground/background command execution |
|
||||
| `ctx.sandbox` | [`sandbox/`](../packages/sandbox/README.md) | same-world process confinement (argv wrapping, per-call policy) |
|
||||
| `ctx.codeRuntime` | [`code-runtime/`](../packages/code-runtime/README.md) | model-written program execution |
|
||||
|
||||
@@ -194,7 +194,7 @@ flowchart LR
|
||||
| ctx key | Role | Owner | Implementations | Direct consumers | Companion plugins | Note |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| `ctx.llm` | `seam` | [`llm`](../packages/llm/llm) | [`llm-deepseek`](../packages/llm/llm-deepseek), [`llm-pi-ai`](../packages/llm/llm-pi-ai), [`llm-replay`](../packages/support/llm-replay) | [`agent-loop`](../packages/core/agent-loop), [`compact-basic`](../packages/compact/compact-basic) | - | Adapters register provider implementations; the loop and compaction call the provider-neutral stream service. |
|
||||
| `ctx.tokenMeter` | `core` | [`token-meter`](../packages/llm/token-meter) | - | [`compact-basic`](../packages/compact/compact-basic) | - | Owns isolated per-model/session replay folds; pressure consumers share immutable revisioned measurements. |
|
||||
| `ctx.tokenMeter` | `core` | [`token-meter`](../packages/llm/token-meter) | - | [`compact-basic`](../packages/compact/compact-basic) | - | Owns isolated per-session replay folds; pressure consumers share immutable revisioned measurements. |
|
||||
| `ctx.sessions` | `core` | [`session`](../packages/core/session) | - | [`agent-loop`](../packages/core/agent-loop), [`agent`](../packages/core/agent), [`session-persistence`](../packages/session-persistence/session-persistence), [`session-query`](../packages/session-query/session-query), [`subagent-inprocess`](../packages/subagent/subagent-inprocess), [`invariants`](../packages/support/invariants) | - | Owns append-only Session instances and emits the durable session event feed. |
|
||||
| `ctx.sessionPersistence` | `seam` | [`session-persistence`](../packages/session-persistence/session-persistence) | [`session-persistence-jsonl`](../packages/session-persistence/session-persistence-jsonl), [`session-persistence-sqlite`](../packages/session-persistence/session-persistence-sqlite) | [`agent-loop`](../packages/core/agent-loop), [`acp`](../packages/ui/acp), [`session-query`](../packages/session-query/session-query) | - | Backends persist the same SessionEvent vocabulary; apps choose a backend at composition time. |
|
||||
| `ctx.sessionQuery` | `seam` | [`session-query`](../packages/session-query/session-query) | - | - | - | Resolves live and optional persisted logs into one logical corpus for exact reads. |
|
||||
|
||||
@@ -226,8 +226,10 @@ Requires: `llm` · `tokenMeter`
|
||||
```ts config-catalog
|
||||
/** Basic compaction configuration; every common field has a deployment default. */
|
||||
export interface BasicCompactConfig {
|
||||
/** Field-wise pressure/retention overrides keyed by configured token-meter model name. */
|
||||
models?: Record<string, ModelCompactConfig>
|
||||
/** Compact at this fraction of the token meter's context window. Defaults to `0.8`. */
|
||||
thresholdRatio?: number
|
||||
/** Recent surface tokens retained verbatim. Defaults to `floor(contextWindow * 0.16)`. */
|
||||
retainTokens?: number
|
||||
/** Summary model; `''` resolves the latest routed model, then `AgentOptions.model`. Defaults to `''`. */
|
||||
summarizationModel?: string
|
||||
/** Provider generation cap for summarization. Defaults to `8192`. */
|
||||
@@ -237,17 +239,9 @@ export interface BasicCompactConfig {
|
||||
/** Enable the automatic `agent/pre-step` pressure listener. Defaults to `true`. */
|
||||
auto?: boolean
|
||||
}
|
||||
|
||||
/** Optional pressure and retention policy for one metered model. */
|
||||
export interface ModelCompactConfig {
|
||||
/** Compact at this fraction of the model's configured context window. Defaults to `0.8`. */
|
||||
thresholdRatio?: number
|
||||
/** Recent surface tokens retained verbatim. Defaults to `floor(contextWindow * 0.16)`. */
|
||||
retainTokens?: number
|
||||
}
|
||||
```
|
||||
|
||||
Source: [`packages/compact/compact-basic/src/types.ts:16`](../packages/compact/compact-basic/src/types.ts)
|
||||
Source: [`packages/compact/compact-basic/src/types.ts:8`](../packages/compact/compact-basic/src/types.ts)
|
||||
|
||||
## `@deepseek-ai/dsh-fs-local`
|
||||
|
||||
@@ -875,20 +869,12 @@ Source: [`packages/context/time-context/src/index.ts:22`](../packages/context/ti
|
||||
```ts config-catalog
|
||||
/** Token-meter plugin configuration. */
|
||||
export interface TokenMeterConfig {
|
||||
/** Built-in field overrides and custom model profiles, keyed by routed model name. */
|
||||
models?: Record<string, ModelTokenMeterConfig>
|
||||
}
|
||||
|
||||
/** Optional pricing fields for one configured model. */
|
||||
export interface ModelTokenMeterConfig {
|
||||
/** Provider context-window capacity in tokens. Required for a custom model. */
|
||||
/** Service-wide context-window capacity in tokens. Defaults to `128000`. */
|
||||
contextWindow?: number
|
||||
/** Heuristic text density in characters per token. Defaults to `4`. */
|
||||
charsPerToken?: number
|
||||
}
|
||||
```
|
||||
|
||||
Source: [`packages/llm/token-meter/src/types.ts:19`](../packages/llm/token-meter/src/types.ts)
|
||||
Source: [`packages/llm/token-meter/src/types.ts:10`](../packages/llm/token-meter/src/types.ts)
|
||||
|
||||
## `@deepseek-ai/dsh-tool-bash`
|
||||
|
||||
|
||||
@@ -263,13 +263,17 @@ Source: [`packages/tasks/tasks/src/index.ts:76`](../../packages/tasks/tasks/src/
|
||||
|
||||
## `ctx.tokenMeter` — `TokenMeterService`
|
||||
|
||||
Concrete registry and replay owner for all configured model meters.
|
||||
Replay owner for one service-wide estimator and isolated per-session folds.
|
||||
|
||||
```ts cordis-catalog
|
||||
resolve(model: string): ModelTokenMeter
|
||||
measure(session: Session, requestHeader?: EpochHeader): TokenMeasurement
|
||||
measureSurface(session: Session): TokenSurfaceMeasurement
|
||||
estimateMessage(message: Message): number
|
||||
```
|
||||
|
||||
Source: [`packages/llm/token-meter/src/index.ts:145`](../../packages/llm/token-meter/src/index.ts)
|
||||
Types: [Message](../core-data-structures/core.md)
|
||||
|
||||
Source: [`packages/llm/token-meter/src/index.ts:92`](../../packages/llm/token-meter/src/index.ts)
|
||||
|
||||
## `ctx.tools` — `ToolRegistry`
|
||||
|
||||
|
||||
@@ -50,7 +50,7 @@ interface CompactionResult {
|
||||
|
||||
## The service
|
||||
|
||||
`CompactService` exposes `compactIfNeeded(...)` for pressure-triggered compaction, returning `null` when no compaction is needed, and `compactRegion(...)` for an explicit inclusive surface range. The pre-step caller supplies the agent, full prompt, session prefix, and abort signal; implementations must forward that signal to summarization. The seam owns no pricing API: `dsh-compact-basic` resolves the routed model through [`ctx.tokenMeter`](token-meter.md), whose model-bound handle owns estimation and replay, while the backend owns retention, event sequencing, and summarization.
|
||||
`CompactService` exposes `compactIfNeeded(...)` for pressure-triggered compaction, returning `null` when no compaction is needed, and `compactRegion(...)` for an explicit inclusive surface range. The pre-step caller supplies the agent, full prompt, session prefix, and abort signal; implementations must forward that signal to summarization. The seam owns no pricing API: [`ctx.tokenMeter`](token-meter.md) directly owns estimation and replay, while `dsh-compact-basic` owns retention, event sequencing, routed summarization calls, and their configuration.
|
||||
|
||||
Auto-compaction runs at serial `agent/pre-step`, before the step and request derivation, so it can replace surface nodes while keeping trace events outside the step. Region boundaries preserve tool-call/result pairing but do not preserve whole turns, allowing early closed steps of one oversized turn to compact. `dsh-compact-basic` owns the retention and failure details.
|
||||
|
||||
|
||||
@@ -8,8 +8,6 @@ Source: [`packages/llm/token-meter/src/types.ts`](../../packages/llm/token-meter
|
||||
|
||||
```ts type-equiv
|
||||
interface TokenMeasurement {
|
||||
/** Model profile used for every heuristic component. */
|
||||
readonly model: string
|
||||
/** Number of durable events consumed; equal to the next unread event seq. */
|
||||
readonly logRevision: number
|
||||
/** Provider or heuristic anchor used for this measurement. */
|
||||
@@ -21,7 +19,7 @@ interface TokenMeasurement {
|
||||
}
|
||||
```
|
||||
|
||||
`baseline.kind === 'usage'` means a successful provider call has the same model and canonical envelope. `estimated` means the meter repriced the complete envelope and surface. Signed `surfaceDeltaTokens` preserves growth and shrinkage relative to a matching provider or estimated anchor.
|
||||
`baseline.kind === 'usage'` means the latest successful provider call has the same canonical request envelope. `estimated` means the service priced the complete envelope and surface with its fixed heuristic. A later successful request replaces the earlier anchor; signed `surfaceDeltaTokens` preserves growth and shrinkage relative to a matching anchor.
|
||||
|
||||
## `TokenSurfaceNode`
|
||||
|
||||
@@ -38,8 +36,6 @@ interface TokenSurfaceNode {
|
||||
|
||||
```ts type-equiv
|
||||
interface TokenSurfaceMeasurement {
|
||||
/** Model profile used to price every node. */
|
||||
readonly model: string
|
||||
/** Number of durable events consumed; equal to the next unread event seq. */
|
||||
readonly logRevision: number
|
||||
/** Total heuristic tokens across the current surface. */
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write
|
||||
2026-07-15-replay-token-meter-service.md: 4452c151e122c4a4ad72e3f0bc2616cd2fa28b9d
|
||||
2026-07-15-replay-token-meter-service.zh.md: 23edc11ffd19b9cfaeb794f3608e9ced4e7dbb7b
|
||||
2026-07-15-replay-token-meter-service.md: 981e789a51bbe7e2b09d47a76d71ac79fe14c999
|
||||
2026-07-15-replay-token-meter-service.zh.md: 85565b6f3db77ab81712f9c128e570c7a16bae8d
|
||||
|
||||
@@ -6,52 +6,52 @@ English | [中文](2026-07-15-replay-token-meter-service.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
Context pressure is useful outside compaction. A compaction backend, an overflow guard, or a future request-policy plugin can all need the same answer: how much of one model's window does the durable request consume? Keeping that fold inside `dsh-compact-basic` duplicates replay logic, makes measurement unavailable without compaction, and encourages callers to reuse accounting from the wrong model.
|
||||
Context pressure is useful outside compaction. A compaction backend, an overflow guard, or a future request-policy plugin can all need the same answer: how much of the configured context window does the durable request consume? Keeping that fold inside `dsh-compact-basic` duplicates replay logic, makes measurement unavailable without compaction, and encourages callers to reuse stale accounting.
|
||||
|
||||
Provider usage is not a complete answer. It describes one successful call under one exact request envelope, while the current surface can grow, shrink, or be replaced afterward. Sessions also switch models, old logs can lack chunk provenance, and provider fields separate input, cache-read, cache-write, output, and reasoning counts. A useful service therefore combines exact anchors with conservative model-specific repricing and exposes the log revision consumed by each result.
|
||||
Provider usage is not a complete answer. It describes one successful call under one exact request envelope, while the current surface can grow, shrink, or be replaced afterward. Sessions also switch models, old logs can lack chunk provenance, and provider fields separate input, cache-read, cache-write, output, and reasoning counts. A useful service therefore combines the latest exact anchor with conservative heuristic repricing and exposes the log revision consumed by each result.
|
||||
|
||||
## Decision
|
||||
|
||||
### One concrete LLM-family service
|
||||
|
||||
`@deepseek-ai/dsh-token-meter` is one concrete package under `packages/llm/` and registers `ctx.tokenMeter`. It is not split into an interface and backend before a second implementation exists. Its public entry point resolves an exact model name to a stable `ModelTokenMeter`; unknown names throw `TokenMeterError` with `TOKEN_METER_MODEL_UNCONFIGURED` instead of inheriting a universal window.
|
||||
`@deepseek-ai/dsh-token-meter` is one concrete package under `packages/llm/` and registers `ctx.tokenMeter`. It is not split into an interface and backend before a second implementation exists. `TokenMeterService` itself exposes `contextWindow`, `measure(session, requestHeader?)`, `measureSurface(session)`, and `estimateMessage(message)`; consumers call the singleton service directly.
|
||||
|
||||
The built-in `deepseek-v4-flash` and `deepseek-v4-pro` profiles use a 128,000-token context window and four characters per estimated token. `models` overrides merge field-by-field. A custom name requires `contextWindow`, while `charsPerToken` defaults to four. Direct construction reports typed profile errors; Loader mounts first apply the package's Schemastery shape validation.
|
||||
The service has one `contextWindow`, defaulting to 128,000 tokens and configurable as a positive integer. Estimation uses a fixed four-characters-per-token heuristic plus structural overhead. There are no model profiles, density settings, tokenizer backends, or language-specific strategies.
|
||||
|
||||
### Model-bound replay folds
|
||||
### Per-session replay folds
|
||||
|
||||
Each model/session pair owns an isolated incremental fold. Active folds advance from `session/event`; every read catches up through the durable tail, so listener ordering, seeded sessions, and service reload do not change the answer. The fold tracks canonical request headers and deltas, step boundaries, surface appends and replacements, assistant usage, and assistant-chunk provenance. A malformed next event fails transactionally and remains unread rather than partially mutating state.
|
||||
Each session owns one isolated incremental fold. Active folds advance from `session/event`; every read catches up through the durable tail, so listener ordering, seeded sessions, and service reload do not change the answer. The fold tracks canonical request headers and deltas, step boundaries, surface appends and replacements, assistant usage, and assistant-chunk provenance. A malformed next event fails transactionally and remains unread rather than partially mutating state.
|
||||
|
||||
`measure(session, requestHeader?)` returns scalar pressure. `measureSurface(session)` returns positional per-node prices for retention and replacement decisions. `estimateMessage(message)` applies the handle's profile without session state. Results are detached, deeply immutable snapshots carrying `logRevision`; a consumer compares scalar and surface revisions before making one decision.
|
||||
`measure(session, requestHeader?)` returns scalar pressure. `measureSurface(session)` returns positional per-node prices for retention and replacement decisions. `estimateMessage(message)` applies the fixed heuristic without session state. Results are detached, deeply immutable snapshots carrying `logRevision`; a consumer compares scalar and surface revisions before making one decision.
|
||||
|
||||
Provider usage is reused only when the handle's model and canonical request envelope equal the successful-call anchor. Any system, prefix, tool, or call-config change causes complete repricing under the requested model. Surface changes remain a signed delta from a matching anchor, including negative values after a shrinking replacement. A success by another model changes the shared surface but never overwrites this model's anchor.
|
||||
Provider usage is reused only when the measured canonical request envelope equals the latest successful-call anchor. Any model, system, prefix, tool, or call-config change causes complete heuristic repricing. Surface changes remain a signed delta from a matching anchor, including negative values after a shrinking replacement. A later successful request replaces the earlier anchor, including across model switches.
|
||||
|
||||
Usage sums the disjoint input, cache-read, cache-write, and output buckets. Reasoning is not added a second time. Every successful model call records an `assistant/message`, including content-less and max-token calls, with its exact earlier chunk seqs. An explicit empty provenance list means a known empty provider stream; absent legacy provenance conservatively treats the durable assistant output as provider output.
|
||||
|
||||
### Compact-basic consumes, but does not own, measurement
|
||||
|
||||
`dsh-compact-basic` requires `ctx.tokenMeter`; `CompactService` gains no token methods or types. The backend is factored into configuration, automatic triggering, region transaction, and summarizer modules, while `summarize()` remains its sole subclass hook. The conversation model's meter consistently prices pressure, retention, shadowed content, provenance, and non-shrinking-summary rejection.
|
||||
`dsh-compact-basic` requires `ctx.tokenMeter`; `CompactService` gains no token methods or types. The backend is factored into configuration, automatic triggering, region transaction, and summarizer modules, while `summarize()` remains its sole subclass hook. The singleton service consistently prices pressure, retention, shadowed content, provenance, and non-shrinking-summary rejection.
|
||||
|
||||
Every metered model receives a compact policy with defaults: threshold ratio `0.8`, retained tail `floor(contextWindow × 0.16)`, summarization model `''`, maximum summary output `8192`, one extra compaction attempt, and automatic triggering enabled. Per-model compact overrides merge `thresholdRatio` and `retainTokens`; retention must remain below the resulting threshold. Empty summarization model resolves the latest logged routed model, then `AgentOptions.model`.
|
||||
Compact policy has service-wide defaults: threshold ratio `0.8`, retained tail `floor(contextWindow × 0.16)`, summarization model `''`, maximum summary output `8192`, one extra compaction attempt, and automatic triggering enabled. Top-level `thresholdRatio` and `retainTokens` override the pressure policy; retention must remain below the resulting threshold. Empty summarization model resolves the latest logged routed model, then `AgentOptions.model`.
|
||||
|
||||
The pre-step trigger measures a provisional envelope: the current prompt and prefix override logged values, while the latest logged header supplies model, tools, and other call config. A model-less router-only agent skips that provisional check because `agent/request` can route later; naming an unknown model remains an error.
|
||||
The pre-step trigger measures a provisional envelope: the current prompt and prefix override logged values, while the latest logged header supplies model, tools, and other call config. A model-less router-only agent skips that provisional check because `agent/request` can route later; any routed model name can use the singleton estimator.
|
||||
|
||||
## Testing
|
||||
|
||||
Unit coverage pins profiles, field-wise overrides, custom and unknown models, envelope invalidation, model switching, usage and missing-usage paths, seeded append/replace replay, signed deltas, provenance modes, malformed boundaries, immutable snapshots, listener ordering, reload, compact defaults, routing fallback, retention, convergence, and transaction rollback. A real Loader/Include YAML fixture loads the exact zero-config token-meter and compact-basic package names in dependency order.
|
||||
Unit coverage pins service configuration, fixed estimation, envelope invalidation, latest-anchor replacement across model switches, usage and missing-usage paths, seeded append/replace replay, signed deltas, provenance modes, malformed boundaries, immutable snapshots, listener ordering, reload, compact defaults, routing fallback, retention, convergence, and transaction rollback. A real Loader/Include YAML fixture loads the exact zero-config token-meter and compact-basic package names in dependency order.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
- **Keep estimation inside `CompactService`** — rejected because measurement has consumers and replay semantics independent of compaction; it would also force every compactor to expose the same unrelated API.
|
||||
- **Split a token-meter interface from a heuristic backend immediately** — rejected because only one implementation exists. One concrete service preserves the future seam without speculative packages or configuration.
|
||||
- **Give unknown models a 128,000-token fallback** — rejected because a plausible but wrong capacity can trigger destructive policy at the wrong point. Unknown routed names fail with their exact name.
|
||||
- **Keep model-keyed windows and density profiles** — rejected because the deployment currently has one context policy and one estimator. Model registries, unknown-model failures, and configurable density add branches without a second behavior to select.
|
||||
- **Copy complete history into each scalar result** — rejected because below-threshold reads are common. Immutable revisioned scalars and a separate surface snapshot preserve consistency without an O(history) copy.
|
||||
- **Treat provider usage as portable between models or envelopes** — rejected because tokenization, context capacity, tools, prefixes, and call config are model/request facts. Mismatch reprices the whole current request.
|
||||
- **Treat provider usage as portable between envelopes** — rejected because model, tools, prefixes, and call config are request facts. Mismatch reprices the whole current request.
|
||||
|
||||
## Consequences
|
||||
|
||||
- Token pressure has one replay-aware owner that compaction and future plugins can share.
|
||||
- Defaults make the bundled DeepSeek composition usable with two zero-config plugin entries, while custom models must state the one fact that cannot be guessed safely: context capacity.
|
||||
- Heuristic density and provider usage remain estimates of provider behavior. Maintainers must update built-in profiles and overflow wording as models evolve.
|
||||
- The default makes the bundled composition usable with two zero-config plugin entries; deployments override one context capacity when needed.
|
||||
- Fixed heuristic pricing remains an estimate of provider behavior and is not an exact tokenizer or request serializer.
|
||||
- Measurements fail loudly on malformed durable boundaries. This turns corrupted replay into a named integration failure instead of silently drifting pressure.
|
||||
- The pre-step compact integration can skip a router-only first check and can miss tool or routing changes applied later in request middleware.
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# RFC: 重放式 token 计量服务
|
||||
# RFC: 回放式 token 计量服务
|
||||
|
||||
Status: implemented
|
||||
|
||||
@@ -6,52 +6,52 @@ Status: implemented
|
||||
|
||||
## 问题
|
||||
|
||||
上下文压力并不只对压缩有用。压缩后端、溢出保护或未来的请求策略插件都可能需要回答同一个问题：持久请求占用了某个模型多少上下文窗口？如果把该折叠逻辑留在 `dsh-compact-basic` 内部，就会重复实现重放逻辑，使未加载压缩的调用方无法使用计量，并诱使调用方错误复用其他模型的核算结果。
|
||||
上下文压力并不只对压缩有用。压缩后端、溢出保护或未来的请求策略插件都可能需要回答同一个问题：持久请求占用了已配置上下文窗口的多少容量？如果把该折叠逻辑留在 `dsh-compact-basic` 内部，就会重复实现回放逻辑，使未加载压缩的调用方无法使用计量，并诱使调用方复用陈旧的核算结果。
|
||||
|
||||
提供方 usage 也不是完整答案。它只描述某个精确请求信封下的一次成功调用，而当前表层之后还可能增长、缩小或被替换。会话也可能切换模型，旧日志可能缺少 chunk 来源，提供方字段还会分别报告输入、缓存读取、缓存写入、输出与推理计数。因此，可用的服务必须把精确锚点与保守的逐模型重新定价结合起来，并公开每个结果已经消费的日志修订号。
|
||||
提供方 usage 也不是完整答案。它只描述某个精确请求信封下的一次成功调用，而当前表层之后还可能增长、缩小或被替换。会话也可能切换模型，旧日志可能缺少分片来源，提供方字段还会分别报告输入、缓存读取、缓存写入、输出与推理计数。因此，可用的服务必须把最新精确锚点与保守的启发式重新定价结合起来，并公开每个结果已经消费的日志修订号。
|
||||
|
||||
## 决策
|
||||
|
||||
### 一个具体的 LLM 家族服务
|
||||
|
||||
`@deepseek-ai/dsh-token-meter` 是 `packages/llm/` 下的单个具体 package，并注册 `ctx.tokenMeter`。在第二种实现出现之前，它不会被拆成接口与后端。公开入口把精确模型名解析为稳定的 `ModelTokenMeter`；未知名称抛出带 `TOKEN_METER_MODEL_UNCONFIGURED` 的 `TokenMeterError`，而不是继承通用窗口。
|
||||
`@deepseek-ai/dsh-token-meter` 是 `packages/llm/` 下的单个具体包，并注册 `ctx.tokenMeter`。在第二种实现出现之前，它不会被拆成接口与后端。`TokenMeterService` 本身公开 `contextWindow`、`measure(session, requestHeader?)`、`measureSurface(session)` 与 `estimateMessage(message)`；消费方直接调用这个单例服务。
|
||||
|
||||
内置的 `deepseek-v4-flash` 与 `deepseek-v4-pro` profile 都采用 128,000 token 上下文窗口，以及每 token 四个字符的估算密度。`models` 覆盖按字段合并。自定义名称必须提供 `contextWindow`，而 `charsPerToken` 默认为四。直接构造会报告类型化 profile 错误；Loader 挂载则先应用 package 的 Schemastery 形状校验。
|
||||
服务只有一个 `contextWindow`，默认值为 128,000 token，并允许配置为正整数。估算采用固定的每 token 四个字符启发式规则，并加上结构开销。服务不提供模型 profile、密度设置、分词器后端或语言专用策略。
|
||||
|
||||
### 绑定模型的重放折叠
|
||||
### 逐会话回放折叠
|
||||
|
||||
每个模型/会话对都有隔离的增量折叠。活跃折叠通过 `session/event` 前进；每次读取都会追到持久日志尾部，因此监听器顺序、种子会话与服务重载不会改变答案。折叠跟踪规范请求头及其增量、步骤边界、表层追加与替换、assistant usage，以及 assistant chunk 来源。下一个畸形事件会以事务方式失败并保持未读，不会让状态只修改一半。
|
||||
每个会话都有一个隔离的增量折叠。活跃折叠通过 `session/event` 前进；每次读取都会追到持久日志尾部，因此监听器顺序、种子会话与服务重载不会改变答案。折叠跟踪规范请求头及其增量、步骤边界、表层追加与替换、assistant usage，以及 assistant 分片来源。下一个畸形事件会以事务方式失败并保持未读，不会让状态只修改一半。
|
||||
|
||||
`measure(session, requestHeader?)` 返回标量压力。`measureSurface(session)` 返回用于保留与替换决策的逐位置节点价格。`estimateMessage(message)` 不依赖会话状态，直接应用该 handle 的 profile。结果是分离且深度不可变的快照，并携带 `logRevision`；消费者在一次联合决策前比较标量与表层修订号。
|
||||
`measure(session, requestHeader?)` 返回标量压力。`measureSurface(session)` 返回用于保留与替换决策的逐位置节点价格。`estimateMessage(message)` 不依赖会话状态，直接应用固定启发式规则。结果是分离且深度不可变的快照，并携带 `logRevision`；消费方在一次联合决策前比较标量与表层修订号。
|
||||
|
||||
只有当 handle 的模型与规范请求信封都等于成功调用锚点时，服务才复用提供方 usage。系统提示词、前缀、工具或调用配置任一变化都会在请求模型下重新定价完整当前请求。表层变化相对匹配锚点保留有符号增量，包括缩小替换后的负值。其他模型的成功调用会改变共享表层，但绝不会覆盖当前模型的锚点。
|
||||
只有当待计量的规范请求信封等于最近一次成功调用的锚点时，服务才复用提供方 usage。模型、系统提示词、前缀、工具或调用配置任一变化都会触发完整的启发式重新定价。表层变化相对匹配锚点保留有符号增量，包括缩小替换后的负值。后续成功请求会替换先前锚点，模型切换时也一样。
|
||||
|
||||
Usage 会对互不重叠的输入、缓存读取、缓存写入与输出 bucket 求和，不会再次加入推理计数。每次成功模型调用都会记录 `assistant/message`，包括无内容调用与达到 token 上限的调用，并带上精确的更早 chunk seq。显式空来源列表表示已知为空的提供方流；旧日志中缺失的来源则保守地把持久 assistant 输出视为提供方输出。
|
||||
Usage 会对互不重叠的输入、缓存读取、缓存写入与输出 bucket 求和，不会再次加入推理计数。每次成功模型调用都会记录 `assistant/message`，包括无内容调用与达到 token 上限的调用，并带上精确的更早分片 seq。显式空来源列表表示已知为空的提供方流；旧日志中缺失的来源则保守地把持久 assistant 输出视为提供方输出。
|
||||
|
||||
### compact-basic 消费计量，但不拥有计量
|
||||
|
||||
`dsh-compact-basic` 要求 `ctx.tokenMeter`；`CompactService` 不增加 token 方法或类型。后端拆分为配置、自动触发、区域事务与摘要器模块，而 `summarize()` 仍是唯一的子类 hook。会话模型的 meter 一致用于压力、保留、被遮蔽内容、来源以及非缩小摘要拒绝。
|
||||
`dsh-compact-basic` 要求 `ctx.tokenMeter`；`CompactService` 不增加 token 方法或类型。后端拆分为配置、自动触发、区域事务与摘要器模块，而 `summarize()` 仍是唯一的子类 hook。单例服务一致用于压力、保留、被遮蔽内容、来源以及非缩小摘要拒绝的定价。
|
||||
|
||||
每个已计量模型都会获得默认压缩策略：阈值比例 `0.8`、保留尾部 `floor(contextWindow × 0.16)`、摘要模型 `''`、摘要最大输出 `8192`、一次额外压缩尝试，以及启用自动触发。逐模型压缩覆盖按字段合并 `thresholdRatio` 与 `retainTokens`；保留值必须小于最终阈值。空摘要模型先解析最近记录的实际路由模型，再使用 `AgentOptions.model`。
|
||||
压缩策略采用服务级默认值：阈值比例 `0.8`、保留尾部 `floor(contextWindow × 0.16)`、摘要模型 `''`、摘要最大输出 `8192`、一次额外压缩尝试，以及启用自动触发。顶层 `thresholdRatio` 与 `retainTokens` 覆盖压力策略；保留值必须小于最终阈值。空摘要模型先解析最近记录的实际路由模型，再使用 `AgentOptions.model`。
|
||||
|
||||
pre-step 触发器计量临时请求信封：当前提示词与前缀覆盖日志值，最近记录的请求头提供模型、工具及其他调用配置。没有模型的纯路由 agent 会跳过该临时检查，因为 `agent/request` 仍可稍后路由；显式命名未知模型仍然报错。
|
||||
pre-step 触发器计量临时请求信封：当前提示词与前缀覆盖日志值，最近记录的请求头提供模型、工具及其他调用配置。没有模型的纯路由 agent（智能体）会跳过该临时检查，因为 `agent/request` 仍可稍后路由；任意路由模型名都可使用这个单例估算器。
|
||||
|
||||
## 测试
|
||||
|
||||
单元覆盖固定 profile、按字段覆盖、自定义与未知模型、信封失效、模型切换、有无 usage 的路径、种子追加/替换重放、有符号增量、来源模式、畸形边界、不可变快照、监听器顺序、重载、压缩默认值、路由回退、保留、收敛与事务回滚。真实 Loader/Include YAML fixture 按依赖顺序加载精确的零配置 token-meter 与 compact-basic package 名称。
|
||||
单元覆盖固定服务配置、固定估算、信封失效、模型切换时替换最新锚点、有无 usage 的路径、种子追加/替换回放、有符号增量、来源模式、畸形边界、不可变快照、监听器顺序、重载、压缩默认值、路由回退、保留、收敛与事务回滚。真实 Loader/Include YAML fixture 按依赖顺序加载精确的零配置 token-meter 与 compact-basic 包名称。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
- **把估算保留在 `CompactService` 内**——不予采纳，因为计量拥有独立于压缩的消费者与重放语义；它还会强迫每个压缩器暴露同一套无关 API。
|
||||
- **立即把 token meter 拆成接口与启发式后端**——不予采纳，因为目前只有一种实现。单个具体服务保留未来接缝，同时避免推测性的 package 与配置。
|
||||
- **给未知模型提供 128,000 token 回退**——不予采纳，因为看似合理但错误的容量会在错误时点触发破坏性策略。未知路由名称会携带精确名称失败。
|
||||
- **把估算保留在 `CompactService` 内**——不予采纳，因为计量拥有独立于压缩的消费方与回放语义；它还会强迫每个压缩器暴露同一套无关 API。
|
||||
- **立即把 token meter 拆成接口与启发式后端**——不予采纳，因为目前只有一种实现。单个具体服务保留未来接缝，同时避免推测性的包与配置。
|
||||
- **保留模型键控的窗口与密度 profile**——不予采纳，因为当前部署只有一种上下文策略与一个估算器。模型注册表、未知模型错误和可配置密度只增加分支，却没有第二种行为可供选择。
|
||||
- **在每个标量结果中复制完整历史**——不予采纳，因为低于阈值的读取很常见。不可变且带修订号的标量与独立表层快照，在不进行 O(history) 复制的情况下保持一致性。
|
||||
- **在模型或信封之间移用提供方 usage**——不予采纳，因为分词、上下文容量、工具、前缀与调用配置都是模型/请求事实。不匹配时会重新定价完整当前请求。
|
||||
- **在不同信封之间移用提供方 usage**——不予采纳，因为模型、工具、前缀与调用配置都是请求事实。不匹配时会重新定价完整当前请求。
|
||||
|
||||
## 后果
|
||||
|
||||
- Token 压力拥有一个可供压缩与未来插件共享的重放感知所有者。
|
||||
- 默认值让内置 DeepSeek 组合只需两个零配置插件条目即可使用，而自定义模型必须声明唯一不能安全猜测的事实：上下文容量。
|
||||
- 启发式密度与提供方 usage 仍然只是提供方行为的估计。随着模型演进，维护者必须更新内置 profile 与溢出措辞。
|
||||
- 遇到畸形持久边界时，计量会明确失败。这会把损坏的重放转化为具名集成错误，而不是让压力静默漂移。
|
||||
- Token 压力拥有一个可供压缩与未来插件共享的回放感知所有者。
|
||||
- 默认值让内置组合只需两个零配置插件条目即可使用；部署需要时只覆盖一个上下文容量。
|
||||
- 固定启发式定价仍然只是提供方行为的估计，并不是精确分词器或请求序列化器。
|
||||
- 遇到畸形持久边界时，计量会明确失败。这会把损坏的回放转化为具名集成错误，而不是让压力静默漂移。
|
||||
- pre-step 压缩集成可能跳过纯路由的首次检查，也可能错过请求中间件稍后应用的工具或路由变化。
|
||||
|
||||
@@ -64,7 +64,7 @@ Auto-compaction always starts at the surface head, merging the prior checkpoint
|
||||
|
||||
### Approximate convergence invariant
|
||||
|
||||
`resolveConfig` supplies usable common defaults: threshold ratio `0.8`, retained tail `floor(contextWindow × 0.16)`, empty summarization-model override, `maxTokens: 8192`, `compactionRetries: 1`, and `auto: true`. Optional per-model threshold/retention fields merge over those defaults and must name a configured meter profile; retained tokens must be below the resulting threshold. Convergence remains dynamic because provider output caps can be spent on hidden or surfaced reasoning tokens and summary size is unpredictable. If the compacted surface remains over threshold, `compactIfNeeded()` re-compacts the head checkpoint up to the configured retry count, but each committed summary must be smaller than what it shadows.
|
||||
`resolveConfig` supplies usable defaults: threshold ratio `0.8`, retained tail `floor(contextWindow × 0.16)`, empty summarization-model override, `maxTokens: 8192`, `compactionRetries: 1`, and `auto: true`. Optional top-level `thresholdRatio` and `retainTokens` override the policy for the token meter's single context window; retention must remain below the resulting threshold. Convergence remains dynamic because provider output caps can be spent on hidden or surfaced reasoning tokens and summary size is unpredictable. If the compacted surface remains over threshold, `compactIfNeeded()` re-compacts the head checkpoint up to the configured retry count, but each committed summary must be smaller than what it shadows.
|
||||
|
||||
### Surface replacement: `compact/*` events are log-only; one `user/message` carries the summary
|
||||
|
||||
@@ -115,7 +115,7 @@ Two failure paths, both documented:
|
||||
- **`SessionEventMap`** gains `compact/start` / `compact/summary` / `compact/end` by declaration merging (merge-extensible); `SurfaceEventType` is **not** touched. These are session events, not cordis `Events`, so the event-taxonomy gate needs no entry.
|
||||
- **`dsh-compact`** owns `toolPairingBalancedBefore(session, node)` and `toolPairingBalancedAfter(session, node)`, the cached surface-edge checks that `compactRegion` and `compactIfNeeded` use to avoid splitting a tool-call/result pair. The cache validates current membership by seq and resolves after-edges from its positional successor map instead of trusting a caller-retained `node.next`; stale or missing seqs and orphan results reject. `dsh-session` continues to own the surface `replace` operation, positional nodes, and rewrite generation.
|
||||
- **`dsh-invariants`** drops its `surface replace: start must be <= end` assertion: a head-anchored compaction lands a high-seq replacement node at an older range's *position*, so `start > end` numerically is normal and valid (the range is positional, validated by the surface's `indexOf` checks that remain). The turn-enclosure invariant is reused unchanged.
|
||||
- **Wiring**: `examples/coding-agent/cordis.yml` loads zero-config `dsh-token-meter` before `dsh-compact-basic`; bundled DeepSeek profiles and compact defaults make the pair usable without repeated numeric policy.
|
||||
- **Wiring**: `examples/coding-agent/cordis.yml` loads zero-config `dsh-token-meter` before `dsh-compact-basic`; the service-wide window and compact defaults make the pair usable without repeated numeric policy.
|
||||
|
||||
## Testing
|
||||
|
||||
|
||||
Reference in New Issue
Block a user