Module compaction 

Conversation-history compaction.

Rust equivalent of (a self-contained subset of) upstream _compaction.py (see UPSTREAM_DRIFT.md §9). Upstream’s _compaction.py is a large (1500+ line), annotation-driven system: it groups messages into logical spans (system / user / assistant-text / tool-call), stamps grouping and token-count metadata onto Message.additional_properties, and never deletes history — it flags messages _excluded and lets the client filter them out when building the payload sent to the model. It ships seven strategies (Truncation, SlidingWindow, SelectiveToolCall, ToolResult, LLM-backed Summarization, TokenBudgetComposed, ContextWindow) plus a CompactionProvider(ContextProvider) that wires a strategy into the client’s get_response loop.

This module intentionally delivers a smaller, dependency-free surface: the Tokenizer and CompactionStrategy abstractions upstream defines, plus four concrete, non-LLM strategies that mirror upstream’s Truncation, SlidingWindow, ContextWindow/TokenBudget, and ToolResult (renamed SelectiveToolResult here to avoid confusion with Content::FunctionResult… “tool result” is the plain-English name). Compaction here works by returning a reduced list rather than annotating messages in place — simpler, and sufficient for the strategies included. Wiring a strategy into the client’s get_response loop (upstream’s CompactionProvider) is intentionally out of scope for this change; see UPSTREAM_DRIFT.md §9.

Compaction never errors on content: given any message list it returns a (possibly unchanged) retained subset that satisfies the strategy’s constraint.

Structs

ApproxTokenizer

A dependency-free default tokenizer using a ~4-characters-per-token heuristic. Mirrors upstream’s CharacterEstimatorTokenizer.

CompactionProvider

A ContextProvider that compacts the accumulated message list — typically the run’s history, once a HistoryProvider has prepended it in before_run — down to fit a CompactionStrategy’s constraint before it reaches the model. Rust equivalent of (a subset of) upstream’s CompactionProvider (see module docs and UPSTREAM_DRIFT.md §9).

SelectiveToolResult

Replace the payload of Content::FunctionResult (tool-result) content in all but the last keep_last messages that carry tool results — they are the bulkiest and least useful once stale. Text and other content is left intact.

SlidingWindow

Keep leading system message(s) + the last window non-system messages. Mirrors upstream’s SlidingWindow strategy.

TokenBudget

Keep leading system message(s), then walk from the newest message backward accumulating token counts, keeping messages until adding the next would exceed max_tokens. Returns the kept messages in original order. Mirrors upstream’s ContextWindow/token-budget strategy.

Truncation

Keep the most recent max_messages, always preserving any leading system message(s) at the front. Mirrors upstream’s Truncation strategy.

Constants

OMITTED_TOOL_RESULT

Stand-in payload left in place of a tool result this strategy compacts.

Traits

CompactionStrategy

A strategy that reduces a message list to fit some constraint.

Tokenizer

Counts tokens for a piece of text. Rust equivalent of upstream TokenizerProtocol.

Functions

compact

Compact messages with strategy and tokenizer.

count_message_tokens

Sum the token counts of a message’s text content (text and reasoning content items) using tokenizer.