Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
Python aicode-complexitycode-healthcode-intelligencecode-qualitydead-codedeveloper-toolsdocumentationgit-analyticsmcpopen-sourcepythonrefactoringstatic-analysistechnical-debt
"""The C4 verb map must stay total over the edge-type vocabulary.
`relation_label` falls through to "depends on" for an unknown type, and that
fall-through is silent: `framework` and `dynamic_uses` rendered that way for
their whole life, covering 10% of file-level edges on a live index, and
`type_use` (6,596 rows across 42 local indexes) still did until this test
existed.
"""
from __future__ import annotations
from repowise.core.analysis.c4.labels import _EDGE_VERB, _VERB_PRIORITY
from repowise.core.ingestion.models import EDGE_TYPE_VALUES
def test_every_edge_type_has_a_verb() -> None:
missing = sorted(EDGE_TYPE_VALUES - _EDGE_VERB.keys())
assert not missing, (
f"edge type(s) with no C4 verb — they render as the vague 'depends on': {missing}"
)
def test_no_verb_is_keyed_on_a_type_nothing_emits() -> None:
dead = sorted(_EDGE_VERB.keys() - EDGE_TYPE_VALUES)
assert not dead, f"C4 verb(s) keyed on a type nothing emits: {dead}"
def test_every_verb_can_be_ranked() -> None:
"""An aggregated edge picks the highest-priority verb, so an unranked one never wins."""
unranked = sorted(set(_EDGE_VERB.values()) - set(_VERB_PRIORITY))
assert not unranked, f"verb(s) missing from _VERB_PRIORITY: {unranked}"