Module compaction 

Structured conversation compaction for context overflow handling

When conversation history exceeds context limits, this module provides tools to compact the history while preserving key information like:

• User’s original intent and requirements
• Important decisions made
• Code changes and their rationale
• Error messages and debugging information

Structs

CodeChange

A code change with metadata

CompactionResult

Compaction result with statistics

CompactionStrategy

Compaction strategy for preserving different types of information

ErrorSolution

An error and its solution

ParsedCompactionSummary

Parse a structured compaction summary

Functions

build_compaction_prompt

Build a structured compaction prompt for the LLM

compact_messages

Compact messages based on the given strategy

parse_compaction_summary

Parse a structured compaction summary from LLM output

summary_to_message

Convert a parsed compaction summary back to a message