mirror of
https://github.com/safishamsi/graphify.git
synced 2026-10-15 08:29:26 +00:00
AndresDiagoM
61ce767dab
- Add to_json() and watch._check_shrink() regression tests that build a real
graph where build_from_json folds a semantic ghost into its AST twin (#4389),
plus a negative control showing a genuine semantic loss is still refused.
Both positive tests fail on v8 and pass with this branch.
- Update _fold_node_aliases / _fold_edge_aliases docstrings and the build()
dedup comment to list the new file/filepath and relationship aliases.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 946333b2ab)
143 lines
5.0 KiB
Python
143 lines
5.0 KiB
Python
"""_lost_saved_semantic_nodes: a semantic ghost deduplicated into its AST twin
|
|
(#1145 extended) is preserved, not lost."""
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
from pathlib import Path
|
|
|
|
from graphify.build import _lost_saved_semantic_nodes, build_from_json
|
|
from graphify.export import to_json
|
|
from graphify.watch import _check_shrink
|
|
|
|
|
|
def _sem(node_id: str, label: str, sf: str) -> dict:
|
|
return {
|
|
"id": node_id,
|
|
"label": label,
|
|
"file_type": "code",
|
|
"source_file": sf,
|
|
"source_location": None,
|
|
"_origin": "semantic",
|
|
}
|
|
|
|
|
|
def _ast(node_id: str, label: str, sf: str) -> dict:
|
|
return {
|
|
"id": node_id,
|
|
"label": label,
|
|
"file_type": "code",
|
|
"source_file": sf,
|
|
"source_location": "L10",
|
|
"_origin": "ast",
|
|
}
|
|
|
|
|
|
def test_same_id_present_is_not_lost():
|
|
saved = [_sem("foo_bar", "bar", "src/foo.py")]
|
|
new = [_sem("foo_bar", "bar", "src/foo.py")]
|
|
assert _lost_saved_semantic_nodes(saved, new) == []
|
|
|
|
|
|
def test_deduplicated_into_ast_node_is_not_lost():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
new = [_ast("foo_py_bar", "bar", "src/foo.py")]
|
|
assert _lost_saved_semantic_nodes(saved, new) == []
|
|
|
|
|
|
def test_non_ast_replacement_is_not_excused():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
new = [_sem("other_bar", "bar", "src/foo.py")]
|
|
lost = _lost_saved_semantic_nodes(saved, new)
|
|
assert [n["id"] for n in lost] == ["ghost_bar"]
|
|
|
|
|
|
def test_ast_label_collision_is_not_excused():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
new = [
|
|
_ast("foo_py_bar_1", "bar", "src/foo.py"),
|
|
_ast("foo_py_bar_2", "bar", "src/foo.py"),
|
|
]
|
|
lost = _lost_saved_semantic_nodes(saved, new)
|
|
assert [n["id"] for n in lost] == ["ghost_bar"]
|
|
|
|
|
|
def test_ast_collision_after_normalization_is_not_excused():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
new = [
|
|
_ast("foo_py_bar_1", "bar", "src/foo.py"),
|
|
_ast("foo_py_bar_2", "bar", "src\\foo.py"),
|
|
]
|
|
lost = _lost_saved_semantic_nodes(saved, new)
|
|
assert [n["id"] for n in lost] == ["ghost_bar"]
|
|
|
|
|
|
def test_dedup_match_normalizes_source_file():
|
|
saved = [_sem("ghost_bar", "bar", "src\\foo.py")]
|
|
new = [_ast("foo_py_bar", "bar", "src/foo.py")]
|
|
assert _lost_saved_semantic_nodes(saved, new) == []
|
|
|
|
|
|
def test_genuinely_lost_node_is_reported():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
new = [
|
|
_ast("foo_py_baz", "baz", "src/foo.py"),
|
|
_ast("other_py_bar", "bar", "src/other.py"),
|
|
]
|
|
lost = _lost_saved_semantic_nodes(saved, new)
|
|
assert [n["id"] for n in lost] == ["ghost_bar"]
|
|
|
|
|
|
def test_node_in_deleted_source_is_not_lost():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
assert _lost_saved_semantic_nodes(saved, [], {"src/foo.py"}) == []
|
|
|
|
|
|
def test_node_in_other_file_not_excused_by_deleted_source():
|
|
saved = [_sem("ghost_bar", "bar", "src/foo.py")]
|
|
lost = _lost_saved_semantic_nodes(saved, [], {"src/other.py"})
|
|
assert [n["id"] for n in lost] == ["ghost_bar"]
|
|
|
|
|
|
def test_malformed_non_string_source_file_does_not_crash():
|
|
# Truthy non-string source_file in new_nodes or saved_nodes must not raise AttributeError
|
|
saved = [{"id": "s1", "label": "S1", "source_file": 123, "_origin": "semantic"}]
|
|
new = [{"id": "n1", "label": "N1", "source_file": True}]
|
|
lost = _lost_saved_semantic_nodes(saved, new)
|
|
assert [n["id"] for n in lost] == ["s1"]
|
|
|
|
|
|
# --- end-to-end: the write paths that call the guard (#4389) ---
|
|
|
|
|
|
def _saved_graph(tmp_path: Path, sem: dict) -> tuple[Path, dict]:
|
|
out = tmp_path / "graph.json"
|
|
assert to_json(build_from_json({"nodes": [sem], "edges": []}), {}, str(out), force=True)
|
|
return out, json.loads(out.read_text(encoding="utf-8"))
|
|
|
|
|
|
def _node_data(G) -> dict:
|
|
return {"nodes": [{"id": nid, **attrs} for nid, attrs in G.nodes(data=True)], "links": []}
|
|
|
|
|
|
def test_to_json_writes_after_ghost_dedup(tmp_path):
|
|
sem = _sem("src_models_user_ts_user", "User", "src/models/user.ts")
|
|
out, _ = _saved_graph(tmp_path, sem)
|
|
G = build_from_json({"nodes": [sem, _ast("user_user", "User", "src/models/user.ts")], "edges": []})
|
|
assert list(G.nodes) == ["user_user"] # ghost id folded into its AST twin
|
|
assert to_json(G, {}, str(out)) is True
|
|
|
|
|
|
def test_to_json_still_refuses_genuine_loss(tmp_path, capsys):
|
|
sem = _sem("src_models_user_ts_user", "User", "src/models/user.ts")
|
|
out, _ = _saved_graph(tmp_path, sem)
|
|
G = build_from_json({"nodes": [_ast("user_account", "Account", "src/models/user.ts")], "edges": []})
|
|
assert to_json(G, {}, str(out)) is False
|
|
assert "src_models_user_ts_user" in capsys.readouterr().err
|
|
|
|
|
|
def test_check_shrink_allows_ghost_dedup(tmp_path):
|
|
sem = _sem("src_models_user_ts_user", "User", "src/models/user.ts")
|
|
_, existing = _saved_graph(tmp_path, sem)
|
|
G = build_from_json({"nodes": [sem, _ast("user_user", "User", "src/models/user.ts")], "edges": []})
|
|
assert _check_shrink(force=False, existing_data=existing, new_data=_node_data(G)) is True
|