freemo
f2232eec09
CI / push-validation (pull_request) Successful in 32s
CI / helm (pull_request) Successful in 42s
CI / build (pull_request) Successful in 45s
CI / lint (pull_request) Successful in 1m13s
CI / quality (pull_request) Successful in 1m14s
CI / typecheck (pull_request) Successful in 1m27s
CI / security (pull_request) Successful in 1m28s
CI / unit_tests (pull_request) Successful in 7m6s
CI / docker (pull_request) Successful in 1m53s
CI / integration_tests (pull_request) Successful in 21m26s
CI / coverage (pull_request) Successful in 13m44s
CI / status-check (pull_request) Successful in 3s
- Updated SkeletonCompressorService.compress() to accept (fragments: tuple[ContextFragment, ...], skeleton_budget: int) -> tuple[ContextFragment, ...], matching the SkeletonCompressor protocol - Removed skeleton_ratio and CompressionResult from the public API - Added @runtime_checkable to SkeletonCompressor protocol in acms_service.py - Added structural subtype assertion at module level to prevent future protocol drift - Rewrote all Behave feature scenarios and step definitions to use skeleton_budget - Updated benchmarks and robot helpers to use absolute token budgets - Removed CompressionResult export from services __init__.py and vulture_whitelist.py - The depth_breadth_projection.py call site already correctly computed and passed skeleton_budget ISSUES CLOSED: #2925
124 lines
5.1 KiB
Gherkin
124 lines
5.1 KiB
Gherkin
Feature: Skeleton compressor
|
|
As an ACMS subsystem
|
|
I want to compress context fragments for subplan inheritance
|
|
So that child plans receive relevant context within a token budget
|
|
|
|
Background:
|
|
Given a skeleton compressor service
|
|
|
|
# --- budget validation -------------------------------------------------
|
|
|
|
Scenario: Reject negative skeleton_budget
|
|
Given context fragments with total tokens 1000
|
|
When I compress with skeleton_budget -1
|
|
Then the compressor should raise a ValueError for invalid budget
|
|
|
|
Scenario: Accept budget 0 returns empty
|
|
Given context fragments with total tokens 1000
|
|
When I compress with skeleton_budget 0
|
|
Then the result should contain zero fragments
|
|
|
|
Scenario: Accept large budget returns all fragments
|
|
Given context fragments with total tokens 1000
|
|
When I compress with skeleton_budget 1000
|
|
Then all fragments should be returned unchanged
|
|
|
|
Scenario: Accept budget at boundary 500
|
|
Given context fragments with total tokens 1000
|
|
When I compress with skeleton_budget 500
|
|
Then compressed tokens should be at most 500
|
|
|
|
# --- stable ordering --------------------------------------------------
|
|
|
|
Scenario: Fragments with equal relevance are ordered by id
|
|
Given three fragments with equal relevance 0.5
|
|
When I compress with skeleton_budget 10000
|
|
Then fragments should be ordered by fragment_id ascending
|
|
|
|
Scenario: Fragments are ordered by relevance descending
|
|
Given fragments with relevances 0.9, 0.3, and 0.7
|
|
When I compress with skeleton_budget 10000
|
|
Then the first fragment should have relevance 0.9
|
|
And the last fragment should have relevance 0.3
|
|
|
|
# --- edge cases -------------------------------------------------------
|
|
|
|
Scenario: Empty fragment list compresses to empty
|
|
Given an empty fragment list
|
|
When I compress with skeleton_budget 500
|
|
Then the result should contain zero fragments
|
|
|
|
Scenario: Single fragment fits within budget
|
|
Given a single fragment with 100 tokens
|
|
When I compress with skeleton_budget 200
|
|
Then the result should contain one fragment
|
|
|
|
Scenario: Single fragment kept even when it exceeds budget
|
|
Given a single fragment with 100 tokens
|
|
When I compress with skeleton_budget 50
|
|
Then the result should contain one fragment
|
|
|
|
# --- argument validation ----------------------------------------------
|
|
|
|
Scenario: Reject non-tuple fragments argument
|
|
When I compress with a non-tuple fragments argument
|
|
Then the compressor should raise a TypeError
|
|
|
|
Scenario: Reject fragment with negative token count
|
|
Given a fragment with negative token count
|
|
When I compress with skeleton_budget 500
|
|
Then the compressor should raise a ValueError for invalid fragment
|
|
|
|
Scenario: Reject fragment with empty id
|
|
Given a fragment with empty fragment_id
|
|
When I compress with skeleton_budget 500
|
|
Then the compressor should raise a ValueError for invalid fragment
|
|
|
|
Scenario: Reject fragment with relevance out of range
|
|
Given a fragment with relevance 1.5
|
|
When I compress with skeleton_budget 500
|
|
Then the compressor should raise a ValueError for invalid fragment
|
|
|
|
Scenario: Reject non-ContextFragment item in tuple
|
|
When I compress with a tuple containing a non-fragment item
|
|
Then the compressor should raise a TypeError for invalid item
|
|
|
|
Scenario: Reject non-integer skeleton_budget
|
|
Given context fragments with total tokens 1000
|
|
When I compress with a non-integer skeleton_budget
|
|
Then the compressor should raise a TypeError for invalid budget type
|
|
|
|
Scenario: Reject skeleton metadata with compressed exceeding original
|
|
When I create skeleton metadata with compressed exceeding original
|
|
Then a validation error should be raised for compressed exceeding original
|
|
|
|
# --- skeleton_budget integration with plan model -----------------------
|
|
|
|
Scenario: Plan model accepts skeleton_metadata
|
|
Given a skeleton metadata with ratio 0.5 and 1000 original tokens and 500 compressed
|
|
When I attach skeleton_metadata to a plan
|
|
Then the plan should expose skeleton metadata in cli dict
|
|
|
|
# --- _validate_fragments edge-cases ------------------------------------
|
|
|
|
Scenario: Reject fragment with negative token_count
|
|
Given a context fragment constructed with negative token_count
|
|
When I validate the invalid fragments
|
|
Then the compressor should raise a ValueError mentioning "token_count must be >= 0"
|
|
|
|
Scenario: Reject fragment with out-of-range relevance_score
|
|
Given a context fragment constructed with relevance_score 1.5
|
|
When I validate the invalid fragments
|
|
Then the compressor should raise a ValueError mentioning "relevance_score must be in"
|
|
|
|
Scenario: Reject fragment with empty fragment_id
|
|
Given a context fragment constructed with empty fragment_id
|
|
When I validate the invalid fragments
|
|
Then the compressor should raise a ValueError mentioning "fragment_id must be non-empty"
|
|
|
|
# --- structural subtype assertion -------------------------------------
|
|
|
|
Scenario: SkeletonCompressorService satisfies SkeletonCompressor protocol
|
|
When I check that SkeletonCompressorService satisfies the SkeletonCompressor protocol
|
|
Then the structural subtype assertion should pass
|