glitchlings 0.2.5__tar.gz → 0.2.6__tar.gz

{glitchlings-0.2.5 → glitchlings-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: glitchlings
-Version: 0.2.5
+Version: 0.2.6
 Summary: Monsters for your language games.
 Author: osoleve
 License:                                  Apache License
@@ -296,7 +296,7 @@ print(gaggle(SAMPLE_TEXT))
 
 Consult the [Glitchlings Usage Guide](docs/index.md)
 for end-to-end instructions spanning the Python API, CLI, HuggingFace and Prime Intellect
-integrations, and the feature-flagged Rust pipeline.
+integrations, and the autodetected Rust pipeline (enabled whenever the extension is present).
 
 ## Motivation
 

{glitchlings-0.2.5 → glitchlings-0.2.6}/README.md

@@ -55,7 +55,7 @@ print(gaggle(SAMPLE_TEXT))
 
 Consult the [Glitchlings Usage Guide](docs/index.md)
 for end-to-end instructions spanning the Python API, CLI, HuggingFace and Prime Intellect
-integrations, and the feature-flagged Rust pipeline.
+integrations, and the autodetected Rust pipeline (enabled whenever the extension is present).
 
 ## Motivation
 

{glitchlings-0.2.5 → glitchlings-0.2.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "glitchlings"
-version = "0.2.5"
+version = "0.2.6"
 description = "Monsters for your language games."
 readme = "README.md"
 requires-python = ">=3.10"

{glitchlings-0.2.5 → glitchlings-0.2.6}/src/glitchlings/dlc/prime.py

@@ -49,7 +49,24 @@ def _resolve_columns(dataset: Dataset, columns: Sequence[str] | None) -> list[st
         if candidate in available:
             return [candidate]
 
-    sample = dataset[0] if len(dataset) else {}
+    try:
+        dataset_length = len(dataset)  # type: ignore[arg-type]
+    except TypeError:
+        preview_rows: list[dict[str, Any]]
+        take_fn = getattr(dataset, "take", None)
+        if callable(take_fn):
+            preview_rows = list(take_fn(1))
+        else:
+            iterator = iter(dataset)
+            try:
+                first_row = next(iterator)
+            except StopIteration:
+                preview_rows = []
+            else:
+                preview_rows = [first_row]
+        sample = dict(preview_rows[0]) if preview_rows else {}
+    else:
+        sample = dataset[0] if dataset_length else {}
     inferred = [
         name
         for name in dataset.column_names

{glitchlings-0.2.5 → glitchlings-0.2.6}/src/glitchlings/zoo/core.py

@@ -27,17 +27,25 @@ log = logging.getLogger(__name__)
 
 
 _PIPELINE_FEATURE_FLAG_ENV = "GLITCHLINGS_RUST_PIPELINE"
+_PIPELINE_ENABLE_VALUES = {"1", "true", "yes", "on"}
+_PIPELINE_DISABLE_VALUES = {"0", "false", "no", "off"}
 
 
 def _pipeline_feature_flag_enabled() -> bool:
-    """Return ``True`` when the environment explicitly opts into the Rust pipeline."""
+    """Return ``True`` when the environment does not explicitly disable the Rust pipeline."""
 
     value = os.environ.get(_PIPELINE_FEATURE_FLAG_ENV)
     if value is None:
-        return False
+        return True
 
     normalized = value.strip().lower()
-    return normalized in {"1", "true", "yes", "on"}
+    if normalized in _PIPELINE_DISABLE_VALUES:
+        return False
+
+    if normalized in _PIPELINE_ENABLE_VALUES:
+        return True
+
+    return True
 
 if TYPE_CHECKING:  # pragma: no cover - typing only
     from datasets import Dataset  # type: ignore
@@ -356,7 +364,7 @@ class Gaggle(Glitchling):
 
     @staticmethod
     def rust_pipeline_enabled() -> bool:
-        """Return ``True`` when the Rust pipeline is available and opted in."""
+        """Return ``True`` when the Rust pipeline is available and not explicitly disabled."""
 
         return Gaggle.rust_pipeline_supported() and _pipeline_feature_flag_enabled()
 

{glitchlings-0.2.5 → glitchlings-0.2.6}/src/glitchlings/zoo/redactyl.py

@@ -86,7 +86,10 @@ def _python_redact_words(
         if core_length <= 0:
             core_length = 1
         weights.append(1.0 if unweighted else float(core_length))
-    num_to_redact = max(1, int(len(word_indices) * rate))
+    raw_quota = len(word_indices) * rate
+    num_to_redact = int(raw_quota)
+    if rate > 0:
+        num_to_redact = max(1, num_to_redact)
     if num_to_redact > len(word_indices):
         raise ValueError("Sample larger than population or is negative")
     indices_to_redact = _weighted_sample_without_replacement(

{glitchlings-0.2.5 → glitchlings-0.2.6}/src/glitchlings.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: glitchlings
-Version: 0.2.5
+Version: 0.2.6
 Summary: Monsters for your language games.
 Author: osoleve
 License:                                  Apache License
@@ -296,7 +296,7 @@ print(gaggle(SAMPLE_TEXT))
 
 Consult the [Glitchlings Usage Guide](docs/index.md)
 for end-to-end instructions spanning the Python API, CLI, HuggingFace and Prime Intellect
-integrations, and the feature-flagged Rust pipeline.
+integrations, and the autodetected Rust pipeline (enabled whenever the extension is present).
 
 ## Motivation
 

{glitchlings-0.2.5 → glitchlings-0.2.6}/src/glitchlings.egg-info/SOURCES.txt

@@ -41,6 +41,7 @@ src/glitchlings/zoo/rushmore.py
 src/glitchlings/zoo/scannequin.py
 src/glitchlings/zoo/typogre.py
 src/glitchlings/zoo/zeedub.py
+tests/test_benchmarks.py
 tests/test_cli.py
 tests/test_dataset_corruption.py
 tests/test_gaggle.py

glitchlings-0.2.6/tests/test_benchmarks.py

@@ -0,0 +1,60 @@
+"""Regression checks for the benchmarking utilities."""
+
+from __future__ import annotations
+
+from collections.abc import Mapping
+from pathlib import Path
+import sys
+
+import pytest
+
+ROOT = Path(__file__).resolve().parents[1]
+if str(ROOT) not in sys.path:
+    sys.path.insert(0, str(ROOT))
+
+from benchmarks.pipeline_benchmark import (
+    BenchmarkResult,
+    collect_benchmark_results,
+)
+
+
+@pytest.fixture(scope="module")
+def benchmark_results() -> Mapping[str, BenchmarkResult]:
+    """Collect a small sample of benchmark data once per test run."""
+
+    results = collect_benchmark_results(iterations=5)
+    return {result.label: result for result in results}
+
+
+def test_collect_benchmark_results_structure(
+    benchmark_results: Mapping[str, BenchmarkResult],
+) -> None:
+    """Top-level sanity check that the benchmark harness returns populated results."""
+
+    assert benchmark_results
+    assert {"short", "medium", "long"}.issubset(benchmark_results.keys())
+    for result in benchmark_results.values():
+        assert result.char_count > 0
+        assert result.python.mean_seconds >= 0
+        assert result.python.stdev_seconds >= 0
+
+
+@pytest.mark.parametrize(
+    ("label", "threshold"),
+    [
+        ("short", 0.01),
+        ("medium", 0.03),
+        ("long", 0.1),
+    ],
+)
+def test_python_pipeline_regression_guard(
+    benchmark_results: Mapping[str, BenchmarkResult],
+    label: str,
+    threshold: float,
+) -> None:
+    """Fail fast if the Python pipeline slows down dramatically on canonical samples."""
+
+    mean_seconds = benchmark_results[label].python.mean_seconds
+    assert mean_seconds <= threshold, (
+        f"Python pipeline mean for '{label}' text exceeded {threshold:.3f}s: {mean_seconds:.3f}s"
+    )

{glitchlings-0.2.5 → glitchlings-0.2.6}/tests/test_prime_echo_chamber.py

@@ -179,6 +179,30 @@ class _RecordingGaggle:
         return dataset
 
 
+
+def test_prime_resolve_columns_handles_streaming_dataset():
+    row = {"context": "alpha", "score": 1, "response": "beta"}
+
+    class StreamingDataset:
+        def __init__(self):
+            self.column_names = ["context", "score", "response"]
+
+        def __len__(self):
+            raise TypeError("Streaming dataset does not define __len__.")
+
+        def __getitem__(self, index):
+            raise TypeError("Streaming dataset does not support indexing.")
+
+        def take(self, n):
+            return [row][:n]
+
+        def __iter__(self):
+            return iter([row])
+
+    inferred = prime._resolve_columns(StreamingDataset(), None)
+
+    assert inferred == ["context", "response"]
+
 def test_load_environment_respects_explicit_columns(monkeypatch):
     dataset = Dataset.from_dict({"prompt": ["alpha"], "extra": ["beta"]})
     stub = _RecordingGaggle()

{glitchlings-0.2.5 → glitchlings-0.2.6}/tests/test_rust_backed_glitchlings.py

@@ -249,6 +249,23 @@ def test_redactyl_merge_adjacent_blocks():
     assert result == expected == "█████████████████"
 
 
+
+def test_redactyl_zero_rate_is_noop(monkeypatch):
+    text = "alpha beta gamma"
+    monkeypatch.setattr(redactyl_module, "_redact_words_rust", None, raising=False)
+
+    result = redactyl_module.redact_words(text, rate=0.0, seed=42)
+    assert result == text
+
+    python_result = redactyl_module._python_redact_words(
+        text,
+        replacement_char=redactyl_module.FULL_BLOCK,
+        rate=0.0,
+        merge_adjacent=False,
+        rng=random.Random(42),
+    )
+    assert python_result == text
+
 def test_redactyl_empty_text_raises_value_error():
     message = "contains no redactable words"
     with pytest.raises(ValueError, match=message):
@@ -425,9 +442,9 @@ def test_gaggle_python_fallback_when_pipeline_disabled(monkeypatch):
     pytest.importorskip("glitchlings._zoo_rust")
 
     def _fail(*_args: object, **_kwargs: object) -> str:
-        raise AssertionError("Rust pipeline should not run when feature flag is disabled")
+        raise AssertionError("Rust pipeline should not run when explicitly disabled")
 
-    monkeypatch.delenv("GLITCHLINGS_RUST_PIPELINE", raising=False)
+    monkeypatch.setenv("GLITCHLINGS_RUST_PIPELINE", "0")
     monkeypatch.setattr(core_module, "_compose_glitchlings_rust", _fail, raising=False)
 
     gaggle = core_module.Gaggle(
@@ -500,10 +517,14 @@ def test_pipeline_falls_back_for_incomplete_operation(monkeypatch):
 
 def test_rust_pipeline_feature_flag_introspection(monkeypatch):
     monkeypatch.delenv("GLITCHLINGS_RUST_PIPELINE", raising=False)
-    assert not core_module._pipeline_feature_flag_enabled()
+    assert core_module._pipeline_feature_flag_enabled()
     assert core_module.Gaggle.rust_pipeline_supported() is (
         core_module._compose_glitchlings_rust is not None
     )
+    assert core_module.Gaggle.rust_pipeline_enabled() is core_module.Gaggle.rust_pipeline_supported()
+
+    monkeypatch.setenv("GLITCHLINGS_RUST_PIPELINE", "0")
+    assert not core_module._pipeline_feature_flag_enabled()
     assert not core_module.Gaggle.rust_pipeline_enabled()
 
     monkeypatch.setenv("GLITCHLINGS_RUST_PIPELINE", "1")