glitchlings 0.2.6__tar.gz → 0.4.0__tar.gz

{glitchlings-0.2.6 → glitchlings-0.4.0}/MANIFEST.in

@@ -1,4 +1,6 @@
 recursive-include rust *.rs *.toml *.lock *.tsv
 recursive-include src/glitchlings/zoo *.tsv
+recursive-include src/glitchlings/lexicon/data *.json
+include src/glitchlings/config.toml
 prune rust/target
 prune rust/zoo/target

{glitchlings-0.2.6/src/glitchlings.egg-info → glitchlings-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: glitchlings
-Version: 0.2.6
+Version: 0.4.0
 Summary: Monsters for your language games.
 Author: osoleve
 License:                                  Apache License
@@ -209,7 +209,7 @@ Project-URL: Homepage, https://github.com/osoleve/glitchlings
 Project-URL: Repository, https://github.com/osoleve/glitchlings.git
 Project-URL: Issues, https://github.com/osoleve/glitchlings/issues
 Project-URL: Changelog, https://github.com/osoleve/glitchlings/releases
-Keywords: nlp,text,adversarial augmentation,text augmentation,large language models,llms,data augmentation,confusables,typo,
+Keywords: nlp,adversarial augmentation,text augmentation,data augmentation,domain randomization
 Classifier: Development Status :: 3 - Alpha
 Classifier: Intended Audience :: Developers
 Classifier: Programming Language :: Python
@@ -224,18 +224,20 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: confusable-homoglyphs>=3.3.1
+Requires-Dist: tomli>=2.0.1; python_version < "3.11"
+Requires-Dist: pyyaml>=6.0.0
 Provides-Extra: hf
 Requires-Dist: datasets>=4.0.0; extra == "hf"
-Provides-Extra: wordnet
-Requires-Dist: nltk>=3.9.1; extra == "wordnet"
-Requires-Dist: numpy<=2.0,>=1.24; extra == "wordnet"
+Provides-Extra: vectors
+Requires-Dist: numpy<=2.0,>=1.24; extra == "vectors"
+Requires-Dist: spacy>=3.7.2; extra == "vectors"
+Requires-Dist: gensim>=4.3.2; extra == "vectors"
 Provides-Extra: prime
 Requires-Dist: verifiers>=0.1.3.post0; extra == "prime"
 Requires-Dist: jellyfish>=1.2.0; extra == "prime"
 Provides-Extra: dev
 Requires-Dist: pytest>=8.0.0; extra == "dev"
 Requires-Dist: hypothesis>=6.140.0; extra == "dev"
-Requires-Dist: nltk>=3.9.1; extra == "dev"
 Requires-Dist: numpy<=2.0,>=1.24; extra == "dev"
 Dynamic: license-file
 
@@ -348,10 +350,30 @@ glitchlings -g "Typogre(rate=0.05)" "Ghouls just wanna have fun"
 
 # Pipe text straight into the CLI for an on-the-fly corruption.
 echo "Beware LLM-written flavor-text" | glitchlings -g mim1c
+
+# Load a roster from a YAML attack configuration.
+glitchlings --config experiments/chaos.yaml "Let slips the glitchlings of war"
 ```
 
 Use `--help` for a complete breakdown of available options, including support for parameterised glitchlings via `-g "Name(arg=value, ...)"` to mirror the Python API.
 
+Attack configurations live in plain YAML files so you can version-control experiments without touching code:
+
+```yaml
+# experiments/chaos.yaml
+seed: 31337
+glitchlings:
+  - name: Typogre
+    rate: 0.04
+  - "Rushmore(rate=0.12, unweighted=True)"
+  - name: Zeedub
+    parameters:
+      rate: 0.02
+      characters: ["\u200b", "\u2060"]
+```
+
+Pass the file to `glitchlings --config` or load it from Python with `glitchlings.load_attack_config` and `glitchlings.build_gaggle`.
+
 ## Development
 
 Follow the [development setup guide](docs/development.md) for editable installs, automated tests, and tips on enabling the Rust pipeline while you hack on new glitchlings.
@@ -416,8 +438,8 @@ _Uh oh. The worst person you know just bought a thesaurus._
 >
 > Args
 >
-> - `rate (float)`: The maximum proportion of words to replace (default: 0.1, 10%).
-> - `part_of_speech`: The WordNet 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.
+> - `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
@@ -428,7 +450,8 @@ _Did you say that or did I?_
 >
 > Args
 >
-> - `rate (float)`: The maximum proportion of words to reduplicate (default: 0.05, 5%).
+> - `rate (float)`: The maximum proportion of words to reduplicate (default: 0.01, 1%).
+> - `unweighted (bool)`: Sample words uniformly instead of favouring shorter tokens (default: False).
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ### Rushmore
@@ -440,6 +463,19 @@ _I accidentally an entire word._
 > Args
 >
 > - `rate (float)`: The maximum proportion of words to delete (default: 0.01, 1%).
+> - `unweighted (bool)`: Sample words uniformly instead of favouring shorter tokens (default: False).
+> - `seed (int)`: The random seed for reproducibility (default: 151).
+
+### Adjax
+
+_Keep your hands and punctuation where I can see them._
+
+> _**Perfect Shuffle.**_ Adjax trades the cores of neighbouring words while leaving punctuation, casing, and surrounding whitespace untouched, turning fluent prose into locally scrambled tongue-twisters.
+>
+> Args
+>
+> - `rate (float)`: Probability that each adjacent pair swaps cores (default: 0.5, 50%).
+> - `swap_rate (float)`: Alias for `rate`, retained for backward compatibility.
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ### Redactyl
@@ -450,9 +486,10 @@ _Oops, that was my black highlighter._
 >
 > ### Args
 >
-> - `replacement_char (str)`: The character to use for redaction (default: █).
-> - `rate (float)`: The maximum proportion of words to redact (default: 0.05, 5%).
+> - `replacement_char (str)`: The character to use for redaction (default: FULL_BLOCK).
+> - `rate (float)`: The maximum proportion of words to redact (default: 0.025, 2.5%).
 > - `merge_adjacent (bool)`: Whether to redact the space between adjacent redacted words (default: False).
+> - `unweighted (bool)`: Sample words uniformly instead of biasing toward longer tokens (default: False).
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ## Field Report: Uncontained Specimens

{glitchlings-0.2.6 → glitchlings-0.4.0}/README.md

@@ -107,10 +107,30 @@ glitchlings -g "Typogre(rate=0.05)" "Ghouls just wanna have fun"
 
 # Pipe text straight into the CLI for an on-the-fly corruption.
 echo "Beware LLM-written flavor-text" | glitchlings -g mim1c
+
+# Load a roster from a YAML attack configuration.
+glitchlings --config experiments/chaos.yaml "Let slips the glitchlings of war"
 ```
 
 Use `--help` for a complete breakdown of available options, including support for parameterised glitchlings via `-g "Name(arg=value, ...)"` to mirror the Python API.
 
+Attack configurations live in plain YAML files so you can version-control experiments without touching code:
+
+```yaml
+# experiments/chaos.yaml
+seed: 31337
+glitchlings:
+  - name: Typogre
+    rate: 0.04
+  - "Rushmore(rate=0.12, unweighted=True)"
+  - name: Zeedub
+    parameters:
+      rate: 0.02
+      characters: ["\u200b", "\u2060"]
+```
+
+Pass the file to `glitchlings --config` or load it from Python with `glitchlings.load_attack_config` and `glitchlings.build_gaggle`.
+
 ## Development
 
 Follow the [development setup guide](docs/development.md) for editable installs, automated tests, and tips on enabling the Rust pipeline while you hack on new glitchlings.
@@ -175,8 +195,8 @@ _Uh oh. The worst person you know just bought a thesaurus._
 >
 > Args
 >
-> - `rate (float)`: The maximum proportion of words to replace (default: 0.1, 10%).
-> - `part_of_speech`: The WordNet 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.
+> - `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
@@ -187,7 +207,8 @@ _Did you say that or did I?_
 >
 > Args
 >
-> - `rate (float)`: The maximum proportion of words to reduplicate (default: 0.05, 5%).
+> - `rate (float)`: The maximum proportion of words to reduplicate (default: 0.01, 1%).
+> - `unweighted (bool)`: Sample words uniformly instead of favouring shorter tokens (default: False).
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ### Rushmore
@@ -199,6 +220,19 @@ _I accidentally an entire word._
 > Args
 >
 > - `rate (float)`: The maximum proportion of words to delete (default: 0.01, 1%).
+> - `unweighted (bool)`: Sample words uniformly instead of favouring shorter tokens (default: False).
+> - `seed (int)`: The random seed for reproducibility (default: 151).
+
+### Adjax
+
+_Keep your hands and punctuation where I can see them._
+
+> _**Perfect Shuffle.**_ Adjax trades the cores of neighbouring words while leaving punctuation, casing, and surrounding whitespace untouched, turning fluent prose into locally scrambled tongue-twisters.
+>
+> Args
+>
+> - `rate (float)`: Probability that each adjacent pair swaps cores (default: 0.5, 50%).
+> - `swap_rate (float)`: Alias for `rate`, retained for backward compatibility.
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ### Redactyl
@@ -209,9 +243,10 @@ _Oops, that was my black highlighter._
 >
 > ### Args
 >
-> - `replacement_char (str)`: The character to use for redaction (default: █).
-> - `rate (float)`: The maximum proportion of words to redact (default: 0.05, 5%).
+> - `replacement_char (str)`: The character to use for redaction (default: FULL_BLOCK).
+> - `rate (float)`: The maximum proportion of words to redact (default: 0.025, 2.5%).
 > - `merge_adjacent (bool)`: Whether to redact the space between adjacent redacted words (default: False).
+> - `unweighted (bool)`: Sample words uniformly instead of biasing toward longer tokens (default: False).
 > - `seed (int)`: The random seed for reproducibility (default: 151).
 
 ## Field Report: Uncontained Specimens

{glitchlings-0.2.6 → glitchlings-0.4.0}/pyproject.toml

@@ -1,19 +1,27 @@
 [project]
 name = "glitchlings"
-version = "0.2.6"
+version = "0.4.0"
 description = "Monsters for your language games."
 readme = "README.md"
 requires-python = ">=3.10"
 
 dependencies = [
     "confusable-homoglyphs>=3.3.1",
+    "tomli>=2.0.1; python_version < '3.11'",
+    "pyyaml>=6.0.0",
 ]
 
 authors = [
     { name = "osoleve" }
 ]
 
-keywords = ["nlp", "text", "adversarial augmentation", "text augmentation", "large language models", "llms", "data augmentation", "confusables", "typo", ""]
+keywords = [
+    "nlp", 
+    "adversarial augmentation",
+    "text augmentation",
+    "data augmentation",
+    "domain randomization"
+]
 
 classifiers = [
     "Development Status :: 3 - Alpha",
@@ -42,12 +50,11 @@ glitchlings = "glitchlings.main:main"
 
 [project.optional-dependencies]
 hf = ["datasets>=4.0.0"]
-wordnet = ["nltk>=3.9.1", "numpy>=1.24,<=2.0"]
+vectors = ["numpy>=1.24,<=2.0", "spacy>=3.7.2", "gensim>=4.3.2"]
 prime = ["verifiers>=0.1.3.post0", "jellyfish>=1.2.0"]
 dev = [
     "pytest>=8.0.0",
     "hypothesis>=6.140.0",
-    "nltk>=3.9.1",
     "numpy>=1.24,<=2.0",
 ]
 
@@ -60,6 +67,8 @@ package-dir = {"" = "src"}
 include-package-data = true
 
 [tool.setuptools.package-data]
+"glitchlings" = ["config.toml"]
+"glitchlings.lexicon" = ["data/*.json"]
 "glitchlings.zoo" = ["ocr_confusions.tsv"]
 
 [tool.setuptools.packages.find]

{glitchlings-0.2.6 → glitchlings-0.4.0}/rust/Cargo.lock

@@ -407,4 +407,5 @@ dependencies = [
  "pyo3",
  "pyo3-build-config",
  "regex",
+ "smallvec",
 ]

{glitchlings-0.2.6 → glitchlings-0.4.0}/rust/zoo/Cargo.toml

@@ -13,6 +13,7 @@ pyo3 = { workspace = true }
 regex = { workspace = true }
 once_cell = { workspace = true }
 blake2 = { workspace = true }
+smallvec = "1"
 
 [package.metadata.maturin]
 module-name = "glitchlings._zoo_rust"