glitchlings 0.4.4__cp310-cp310-win_amd64.whl → 0.4.5__cp310-cp310-win_amd64.whl

glitchlings/zoo/hokey.py

@@ -0,0 +1,173 @@
+"""Hokey glitchling that performs expressive lengthening."""
+
+from __future__ import annotations
+
+import random
+from typing import Any, cast
+
+from ..util.hokey_generator import HokeyConfig, HokeyGenerator, StretchEvent
+from ..util.stretchability import StretchabilityAnalyzer
+from ._rust_extensions import get_rust_operation
+from .core import AttackOrder, AttackWave, Gaggle
+from .core import Glitchling as GlitchlingBase
+
+_hokey_rust = get_rust_operation("hokey")
+_ANALYZER = StretchabilityAnalyzer()
+_GENERATOR = HokeyGenerator(analyzer=_ANALYZER)
+
+
+def _python_extend_vowels(
+    text: str,
+    *,
+    rate: float,
+    extension_min: int,
+    extension_max: int,
+    word_length_threshold: int,
+    base_p: float,
+    rng: random.Random,
+    return_trace: bool = False,
+) -> str | tuple[str, list[StretchEvent]]:
+    config = HokeyConfig(
+        rate=rate,
+        extension_min=extension_min,
+        extension_max=extension_max,
+        word_length_threshold=word_length_threshold,
+        base_p=base_p,
+    )
+    result, events = _GENERATOR.generate(text, rng=rng, config=config)
+    return (result, events) if return_trace else result
+
+
+def extend_vowels(
+    text: str,
+    rate: float = 0.3,
+    extension_min: int = 2,
+    extension_max: int = 5,
+    word_length_threshold: int = 6,
+    seed: int | None = None,
+    rng: random.Random | None = None,
+    *,
+    return_trace: bool = False,
+    base_p: float | None = None,
+) -> str | tuple[str, list[StretchEvent]]:
+    """Extend expressive segments of words for emphasis.
+
+    Parameters
+    ----------
+    text : str
+        Input text to transform.
+    rate : float, optional
+        Global selection rate for candidate words.
+    extension_min : int, optional
+        Minimum number of extra repetitions for the stretch unit.
+    extension_max : int, optional
+        Maximum number of extra repetitions for the stretch unit.
+    word_length_threshold : int, optional
+        Preferred maximum alphabetic length; longer words are de-emphasised but not
+        excluded.
+    seed : int, optional
+        Deterministic seed when ``rng`` is not supplied.
+    rng : random.Random, optional
+        Random number generator to drive sampling.
+    return_trace : bool, optional
+        When ``True`` also return the stretch events for introspection.
+    base_p : float, optional
+        Base probability for the negative-binomial sampler (heavier tails for smaller
+        values). Defaults to ``0.45``.
+    """
+    if not text:
+        empty_trace: list[StretchEvent] = []
+        return (text, empty_trace) if return_trace else text
+
+    if rng is None:
+        rng = random.Random(seed)
+    base_probability = base_p if base_p is not None else 0.45
+
+    if return_trace or _hokey_rust is None:
+        return _python_extend_vowels(
+            text,
+            rate=rate,
+            extension_min=extension_min,
+            extension_max=extension_max,
+            word_length_threshold=word_length_threshold,
+            base_p=base_probability,
+            rng=rng,
+            return_trace=return_trace,
+        )
+
+    return cast(
+        str,
+        _hokey_rust(
+            text,
+            rate,
+            extension_min,
+            extension_max,
+            word_length_threshold,
+            base_probability,
+            rng,
+        ),
+    )
+
+
+class Hokey(GlitchlingBase):
+    """Glitchling that stretches words using linguistic heuristics."""
+
+    seed: int | None
+
+    def __init__(
+        self,
+        *,
+        rate: float = 0.3,
+        extension_min: int = 2,
+        extension_max: int = 5,
+        word_length_threshold: int = 6,
+        base_p: float = 0.45,
+        seed: int | None = None,
+    ) -> None:
+        self._master_seed: int | None = seed
+
+        def _corruption_wrapper(text: str, **kwargs: Any) -> str:
+            result = extend_vowels(text, **kwargs)
+            return result if isinstance(result, str) else result[0]
+
+        super().__init__(
+            name="Hokey",
+            corruption_function=_corruption_wrapper,
+            scope=AttackWave.CHARACTER,
+            order=AttackOrder.FIRST,
+            seed=seed,
+            rate=rate,
+            extension_min=extension_min,
+            extension_max=extension_max,
+            word_length_threshold=word_length_threshold,
+            base_p=base_p,
+        )
+
+    def pipeline_operation(self) -> dict[str, Any] | None:
+        return {
+            "type": "hokey",
+            "rate": self.kwargs.get("rate", 0.3),
+            "extension_min": self.kwargs.get("extension_min", 2),
+            "extension_max": self.kwargs.get("extension_max", 5),
+            "word_length_threshold": self.kwargs.get("word_length_threshold", 6),
+            "base_p": self.kwargs.get("base_p", 0.45),
+        }
+
+    def reset_rng(self, seed: int | None = None) -> None:
+        if seed is not None:
+            self._master_seed = seed
+            super().reset_rng(seed)
+            if self.seed is None:
+                return
+            derived = Gaggle.derive_seed(int(seed), self.name, 0)
+            self.seed = int(derived)
+            self.rng = random.Random(self.seed)
+            self.kwargs["seed"] = self.seed
+        else:
+            super().reset_rng(None)
+
+
+hokey = Hokey()
+
+
+__all__ = ["Hokey", "hokey", "extend_vowels"]

{glitchlings-0.4.4.dist-info → glitchlings-0.4.5.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: glitchlings
-Version: 0.4.4
+Version: 0.4.5
 Summary: Monsters for your language games.
 Author: osoleve
 License:                                  Apache License
@@ -294,6 +294,12 @@ Dynamic: license-file
                         Every language game breeds monsters.
 ```
 
+[![PyPI version](https://img.shields.io/pypi/v/glitchlings.svg)](https://pypi.org/project/glitchlings/)
+[![PyPI Status](https://github.com/osoleve/glitchlings/actions/workflows/publish.yml/badge.svg)](https://github.com/osoleve/glitchlings/actions/workflows/publish.yml)
+[![Lint and Type](https://github.com/osoleve/glitchlings/actions/workflows/ci.yml/badge.svg)](https://github.com/osoleve/glitchlings/actions/workflows/ci.yml)
+[![Website status](https://img.shields.io/website-up-down-green-red/https/osoleve.github.io/glitchlings)](https://osoleve.github.io/glitchlings/)
+[![License](https://img.shields.io/github/license/osoleve/glitchlings.svg)](https://github.com/osoleve/glitchlings/blob/main/LICENSE)
+
 `Glitchlings` are **utilities for corrupting the text inputs to your language models in deterministic, _linguistically principled_** ways.  
 Each embodies a different way that documents can be compromised in the wild.
 
@@ -382,6 +388,7 @@ glitchlings --list
 ```text
    Typogre — scope: Character, order: early
 Apostrofae — scope: Character, order: normal
+     Hokey — scope: Character, order: first
      Mim1c — scope: Character, order: last
   Jargoyle — scope: Word, order: normal
      Adjax — scope: Word, order: normal
@@ -429,10 +436,6 @@ options:
 ```
 <!-- END: CLI_USAGE -->
 
-Run `python docs/build_cli_reference.py` whenever you tweak the CLI so the README stays in sync with the actual output. The script executes the commands above and replaces the block between the markers automatically.
-
-Prefer inline tweaks? You can still configure glitchlings directly in the shell:
-
 ```bash
 # Run Typogre against the contents of a file and inspect the diff.
 glitchlings -g typogre --file documents/report.txt --diff
@@ -507,6 +510,22 @@ _Wait, was that...?_
 > - `banned_characters (Collection[str])`: Characters that must never appear as replacements (default: none).
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
+### Hokey
+
+_She's soooooo coooool!_
+
+> _**Passionista.**_ Hokey sometimes gets a little excited and elongates words for emphasis.
+>
+> Args
+>
+> - `rate (float)`: Share of high-scoring tokens to stretch (default: 0.3).
+> - `extension_min` / `extension_max (int)`: Bounds for extra repetitions (defaults: 2 / 5).
+> - `word_length_threshold (int)`: Preferred maximum alphabetic length; longer words are damped instead of excluded (default: 6).
+> - `base_p (float)`: Base probability for the heavy-tailed sampler (default: 0.45).
+> - `seed (int)`: The random seed for reproducibility (default: 151).
+
+_Apocryphal Glitchling contributed by Chloé Nunes_
+
 ### Scannequin
 
 _How can a computer need reading glasses?_
@@ -539,7 +558,9 @@ _Uh oh. The worst person you know just bought a thesaurus._
 > Args
 >
 > - `rate (float)`: The maximum proportion of words to replace (default: 0.01, 1%).
+>
 - `part_of_speech`: The WordNet-style part(s) of speech to target (default: nouns). Accepts `wn.NOUN`, `wn.VERB`, `wn.ADJ`, `wn.ADV`, any iterable of those tags, or the string `"any"` to include them all. Vector/graph backends ignore this filter while still honouring deterministic sampling.
+>
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ### Reduple

{glitchlings-0.4.4.dist-info → glitchlings-0.4.5.dist-info}/RECORD

@@ -1,25 +1,30 @@
-glitchlings/__init__.py,sha256=jyfvslkaFlYVefDFq0CQKkc52F8p8I1UrbFbcyihFCU,1249
+glitchlings/__init__.py,sha256=pNL6Vx6ggpRq-vW0CXJjnWOe_LktP6Zz9xNhqDHj3Dw,1301
 glitchlings/__main__.py,sha256=nB7btO_T4wBFOcyawfWpjEindVrUfTqqV5hdeeS1HT8,128
-glitchlings/_zoo_rust.cp310-win_amd64.pyd,sha256=o5tlRkqbmzbnWqe6r0k5Ltk5lfo-WZExCJTz_s1IjnE,2215936
-glitchlings/compat.py,sha256=xM5fT5RELgIdQTmgKQFZaPZJJpOg0noC-gLDcO390Ro,9192
-glitchlings/config.py,sha256=pKpM5onr9UG8sHKVIZKj2Ti2gD5bZ6N1wyxOhzgrviQ,13341
+glitchlings/_zoo_rust.cp310-win_amd64.pyd,sha256=HXlrBd0WtqjwzuxEYFpvVxEHnylaGfOetxxZDIlncU0,2350080
+glitchlings/compat.py,sha256=j4lkWNtyox5sen5j7u0SHfnk8QUn-yicaqvuLlZp1-s,9174
+glitchlings/config.py,sha256=rBBHRmkIyoWRPcUFrsxmSm08qxZ07wDAuO5AAF6sjAk,13323
 glitchlings/config.toml,sha256=OywXmEpuOPtyJRbcRt4cwQkHiZ__5axEHoCaX9ye-uA,102
 glitchlings/main.py,sha256=eCUEFsu8-NLDz1xyKNDIucVm975HNQZJYm6YCv8RIyg,10987
+glitchlings/data/__init__.py,sha256=kIT4a2EDNo_R-iL8kVisJ8PxR_BrygNYegfG_Ua6DcE,69
+glitchlings/data/hokey_assets.json,sha256=1GaSEzXwtT1nvf0B9mFyLzHOcqzKbPreibsC6iBWAHA,3083
 glitchlings/dlc/__init__.py,sha256=iFDTwkaWl2C0_QUYykIXfmOUzy__oURX_BiJhexf-8o,312
-glitchlings/dlc/_shared.py,sha256=L6dc4Xi0q2K4ZwkyytXIU8Zu3MPugIDXhseZNuJNqyY,4560
-glitchlings/dlc/huggingface.py,sha256=Ym8dTArb-43AnCyukOO1m66iAbs8al9YkIWB3rGdhTk,2657
-glitchlings/dlc/prime.py,sha256=KY3so8WOwksbsKhZXfWorsZSYvIdPiUtc8e9apHabkM,8861
-glitchlings/dlc/pytorch.py,sha256=FZZbWHVX94FA9Ab77twZPUwQNno4IKTAjuHPXNhdY6M,6399
-glitchlings/dlc/pytorch_lightning.py,sha256=cKLFn3cFcOhFSN3QFlcvRPP5sLQts0M6DQ1J1cQuqRM,8289
+glitchlings/dlc/_shared.py,sha256=q1xMclEsbR0KIEn9chwZXRubMnIvU-uIc_0PxCFpmqE,4560
+glitchlings/dlc/huggingface.py,sha256=6wD4Vu2jp1d8GYSUiAvOAXqCO9w4HBxCOSfbJM8Ylzw,2657
+glitchlings/dlc/prime.py,sha256=8-Ix8bjKyDKDMyXHDESE-gFDwC6blAP3mPXm1YiP3_U,8861
+glitchlings/dlc/pytorch.py,sha256=xR-uoeo8f6g-aM2CmQ1cfHKfqn12uwRA9eCebEO5wXA,6399
+glitchlings/dlc/pytorch_lightning.py,sha256=EQq7SMlvNoyBLYVW2Vi19H5k8VUqoAiNx8IqkDWqy5I,8214
 glitchlings/lexicon/__init__.py,sha256=oSLI1bOMfTpqiI7bMp9beeQ7Vp81G5coXxwdmCIQfV0,6199
-glitchlings/lexicon/_cache.py,sha256=KWqJ__WrM2ccIlplaaqoVT0ns65uU5WHewlJd4BvnJE,4196
+glitchlings/lexicon/_cache.py,sha256=MRvomTi2Rx0l9FzHm91VrfRkrHqACjS5LG4a4OOCvLY,4180
 glitchlings/lexicon/metrics.py,sha256=TZAafSKgHpUS4h6vCuhTKGsvu_fru9kMqsXqLID6BTM,4734
-glitchlings/lexicon/vector.py,sha256=x9iT1O8Osolwt08g41V_70WHZt_b4OGzHBU72YHkmwg,23181
-glitchlings/lexicon/wordnet.py,sha256=fJi5SNa-sLpQiTIoXorkYzc2ZArejIms6zhoe8TPIOg,7840
+glitchlings/lexicon/vector.py,sha256=1YivmCdb9IFqVNQTJiTztxRoFQi7Fmmg-8qKO7FUKzM,23252
+glitchlings/lexicon/wordnet.py,sha256=Kdj0k1m9GJAdH41zVw0MkEAoLy0-KG0ZyS6hGFq_raY,7804
 glitchlings/lexicon/data/default_vector_cache.json,sha256=bnMV4tHIVOQtK7FDH81yqSLRkeViEzclGKXrrS8fEJ8,1079
 glitchlings/util/__init__.py,sha256=Q5lkncOaM6f2eJK3HAtZyxpCjGnekCpwPloqasS3JDo,4869
 glitchlings/util/adapters.py,sha256=mFhPlE8JaFuO_C-3_aqhgwkqa6isV8Y2ifqVh3Iv9JM,720
-glitchlings/zoo/__init__.py,sha256=j21naQtFunJYgdgYsyTNYUSa7sl88yNQWaGP8sWyw5U,5411
+glitchlings/util/hokey_generator.py,sha256=hNWVbscVeKvcqwFtJ1oaWYf2Z0qHSxy0-iMLjht6zuM,4627
+glitchlings/util/stretch_locator.py,sha256=tM4XsQ_asNXQq2Yee8STybFMCC2HbSKCu9I49G0yJ3c,4334
+glitchlings/util/stretchability.py,sha256=ja1PJKVAKySUP5G2T0Q_THwIvydJn49xYFaQn58cQYw,13194
+glitchlings/zoo/__init__.py,sha256=Ab69SD_RUXfiWUEzqU3wteGWobpm2kkbmwndYLEbv_0,5511
 glitchlings/zoo/_ocr_confusions.py,sha256=pPlvJOoan3ouwwGt8hATcO-9luIrGJl0vwUqssUMXD8,1236
 glitchlings/zoo/_rate.py,sha256=KxFDFJEGWsv76v1JcoHXIETj9kGdbbAiUqCPwAcWcDw,3710
 glitchlings/zoo/_rust_extensions.py,sha256=SdU06m-qjf-rRHnqM5OMophx1IacoI4KbyRu2HXrTUc,4354
@@ -28,6 +33,7 @@ glitchlings/zoo/_text_utils.py,sha256=LqCa33E-Qxbk6N5AVfxEmAz6C2u7_mCF0xPT9-404A
 glitchlings/zoo/adjax.py,sha256=8AqTfcJe50vPVxER5wREuyr4qqsmyKqi-klDW77HVYM,3646
 glitchlings/zoo/apostrofae.py,sha256=WB1zvCc1_YvTD9XdTNtFrK8H9FleaF-UNMCai_E7lqE,3949
 glitchlings/zoo/core.py,sha256=tkJFqGwpVa7qQxkUr9lsHOB8zS3lDCo987R6Nvz375U,21257
+glitchlings/zoo/hokey.py,sha256=h-yjCLqYAZFksIYw4fMniuTWUywFw9GU9yUFs_uECHo,5471
 glitchlings/zoo/jargoyle.py,sha256=kgKw_hsaJaKQjrxzt_CMlgtZApedM6I2-U3d3aeipXs,12014
 glitchlings/zoo/mim1c.py,sha256=GqUMErVAVcqMAZjx4hhJ0Af25CxA0Aorv3U_fTqLZek,3546
 glitchlings/zoo/ocr_confusions.tsv,sha256=S-IJEYCIXYKT1Uu7Id8Lnvg5pw528yNigTtWUdnMv9k,213
@@ -39,9 +45,9 @@ glitchlings/zoo/typogre.py,sha256=X6jcnCYifNtoSwe2ig7YS3QrODyrACirMZ9pjN5LLBA,69
 glitchlings/zoo/zeedub.py,sha256=KApSCSiTr3b412vsA8UHWFhYQDxe_jg0308wWK-GNtM,5025
 glitchlings/zoo/assets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 glitchlings/zoo/assets/apostrofae_pairs.json,sha256=lPLFLndzn_f7_5wZizxsLMnwBY4O63zsCvDjyJ56MLA,553
-glitchlings-0.4.4.dist-info/licenses/LICENSE,sha256=EFEP1evBfHaxsMTBjxm0sZVRp2wct8QLvHE1saII5FI,11538
-glitchlings-0.4.4.dist-info/METADATA,sha256=G5yU2jAph6LYY0Hvr5WTa3tnSdmDI890nTiecmP7P68,33015
-glitchlings-0.4.4.dist-info/WHEEL,sha256=KUuBC6lxAbHCKilKua8R9W_TM71_-9Sg5uEP3uDWcoU,101
-glitchlings-0.4.4.dist-info/entry_points.txt,sha256=kGOwuAsjFDLtztLisaXtOouq9wFVMOJg5FzaAkg-Hto,54
-glitchlings-0.4.4.dist-info/top_level.txt,sha256=VHFNBrLjtDwPCYXbGKi6o17Eueedi81eNbR3hBOoST0,12
-glitchlings-0.4.4.dist-info/RECORD,,
+glitchlings-0.4.5.dist-info/licenses/LICENSE,sha256=EFEP1evBfHaxsMTBjxm0sZVRp2wct8QLvHE1saII5FI,11538
+glitchlings-0.4.5.dist-info/METADATA,sha256=sULt2kvkK1lfhdqBbSvawNoRNyevwLsFrITBPMBrE9M,34146
+glitchlings-0.4.5.dist-info/WHEEL,sha256=KUuBC6lxAbHCKilKua8R9W_TM71_-9Sg5uEP3uDWcoU,101
+glitchlings-0.4.5.dist-info/entry_points.txt,sha256=kGOwuAsjFDLtztLisaXtOouq9wFVMOJg5FzaAkg-Hto,54
+glitchlings-0.4.5.dist-info/top_level.txt,sha256=VHFNBrLjtDwPCYXbGKi6o17Eueedi81eNbR3hBOoST0,12
+glitchlings-0.4.5.dist-info/RECORD,,