grammar-inference-engine/tests/test_analyze.py
tobjend 2620b6e49f
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
feat: method-level sequence extraction via child_by_field_name('body')
- _find_method_bodies uses tree-sitter's universal body field (9/10 grammars)
- Kotlin fallback: scan children for body-like types
- preprocess_by_method groups highlight captures by enclosing method body
- Returns per-method sequences for k-ORE ordering analysis
- analyze.py infer() now uses preprocess_by_method
- Document findings in ANALYSIS.md
- 97 tests pass
2026-07-03 21:34:25 +02:00

189 lines
6.5 KiB
Python

"""Tests for tag-preprocessor orchestrator (analyze.py)."""
from pathlib import Path
import tempfile
import sys
sys.path.insert(0, str(Path(__file__).parent.parent))
from bex.tag_preprocessor.analyze import (
scan_directory, frequency_filter, infer, analyze_directory, _match_glob,
)
def test_scan_directory_empty():
with tempfile.TemporaryDirectory() as td:
result = scan_directory(td)
assert result == {}
print(" PASS test_scan_directory_empty")
def test_scan_directory_groups_by_extension():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "a.py").write_text("x = 1")
(d / "b.py").write_text("y = 2")
(d / "c.js").write_text("let x = 1;")
(d / "d.rs").write_text("fn main() {}")
result = scan_directory(td)
assert ".py" in result
assert ".js" in result
assert ".rs" in result
assert len(result[".py"]) == 2
assert len(result[".js"]) == 1
assert len(result[".rs"]) == 1
print(" PASS test_scan_directory_groups_by_extension")
def test_scan_directory_skips_unsupported():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "f.py").write_text("x = 1")
(d / "f.txt").write_text("hello")
(d / "f.md").write_text("# doc")
result = scan_directory(td)
assert ".py" in result
assert ".txt" not in result
assert ".md" not in result
print(" PASS test_scan_directory_skips_unsupported")
def test_scan_directory_nested():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "sub").mkdir()
(d / "sub" / "a.py").write_text("x = 1")
(d / "b.py").write_text("y = 2")
result = scan_directory(td)
assert len(result[".py"]) == 2
print(" PASS test_scan_directory_nested")
def test_scan_directory_skips_build_dirs():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "src" / "main").mkdir(parents=True)
(d / "build" / "reports").mkdir(parents=True)
(d / "node_modules" / "pkg").mkdir(parents=True)
(d / "src" / "main" / "app.py").write_text("x = 1")
(d / "build" / "reports" / "report.js").write_text("let x = 1;")
(d / "node_modules" / "pkg" / "index.js").write_text("let y = 2;")
result = scan_directory(td)
assert ".py" in result
assert ".js" not in result
assert len(result[".py"]) == 1
print(" PASS test_scan_directory_skips_build_dirs")
def test_frequency_filter_nothing_to_filter():
sequences = [
[("function", "foo", 1), ("keyword.return", "return", 2)],
[("function", "bar", 1), ("keyword.return", "return", 2)],
]
filtered = frequency_filter(sequences, min_coverage=0.5)
assert len(filtered) == 2
assert filtered == sequences
print(" PASS test_frequency_filter_nothing_to_filter")
def test_frequency_filter_removes_infrequent_symbol():
sequences = [
[("keyword.return", "return", 1)],
[("keyword.return", "return", 1)],
[("function", "rare_fn", 1)],
]
filtered = frequency_filter(sequences, min_coverage=0.67)
assert len(filtered) == 3
assert len(filtered[0]) == 1 # "return" kept
assert len(filtered[1]) == 1 # "return" kept
assert len(filtered[2]) == 0 # "rare_fn" removed
print(" PASS test_frequency_filter_removes_infrequent_symbol")
def test_frequency_filter_edge_empty_sequences():
assert frequency_filter([], min_coverage=0.5) == []
assert frequency_filter([[], []], min_coverage=0.5) == [[], []]
print(" PASS test_frequency_filter_edge_empty_sequences")
def test_infer_returns_ensemble_dict():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "a.py").write_text("def foo():\n return 1")
(d / "b.py").write_text("def bar():\n return 2")
result = infer([str(d / "a.py"), str(d / "b.py")], ".py", min_coverage=0.5)
assert isinstance(result, dict)
assert "best" in result
assert "all" in result
assert "why" in result
print(" PASS test_infer_returns_ensemble_dict")
def test_match_glob():
assert _match_glob("/repo/src/main/app.kt", "**/src/main/**")
assert _match_glob("/repo/src/main/org/app.kt", "**/src/main/**")
assert _match_glob("/repo/src/main/deep/nested/app.kt", "**/src/main/**")
assert not _match_glob("/repo/src/test/app.kt", "**/src/main/**")
assert not _match_glob("/repo/build/app.kt", "**/src/main/**")
print(" PASS test_match_glob")
def test_analyze_directory_include_glob():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "src" / "main").mkdir(parents=True)
(d / "src" / "test").mkdir(parents=True)
(d / "src" / "main" / "prod.py").write_text("x = 1")
(d / "src" / "test" / "test_prod.py").write_text("y = 2")
results = analyze_directory(td, include="**/src/main/**")
assert ".py" in results
assert len(results[".py"]["all"]) >= 1
print(" PASS test_analyze_directory_include_glob")
def test_infer_low_coverage_filters_noise():
with tempfile.TemporaryDirectory() as td:
d = Path(td)
(d / "common.py").write_text(
"def setup():\n pass\ndef teardown():\n pass"
)
(d / "rare.py").write_text(
"def setup():\n pass\ndef one_off():\n raise Exception('boom')"
)
result = infer(
[str(d / "common.py"), str(d / "rare.py")], ".py", min_coverage=0.6
)
assert result["best"] is not None
assert result["best"]["grammar"] is not None
print(" PASS test_infer_low_coverage_filters_noise")
def run_all():
tests = [
test_scan_directory_empty,
test_scan_directory_groups_by_extension,
test_scan_directory_skips_unsupported,
test_scan_directory_nested,
test_scan_directory_skips_build_dirs,
test_match_glob,
test_analyze_directory_include_glob,
test_frequency_filter_nothing_to_filter,
test_frequency_filter_removes_infrequent_symbol,
test_frequency_filter_edge_empty_sequences,
test_infer_returns_ensemble_dict,
test_infer_low_coverage_filters_noise,
]
passed = 0
failed = 0
for t in tests:
try:
t()
passed += 1
except Exception as e:
print(f" FAIL {t.__name__}: {e}")
failed += 1
print(f"\n{passed} passed, {failed} failed")
if __name__ == "__main__":
run_all()