Upgrading deps to the latest one (#241)

* * update slack to MLSecOps

* * upgrade deps

* * all ruff issues fixed

* * fixing issues with upgrades of presidio

Author: asofter

Makefile

@@ -3,7 +3,7 @@
 .PHONY: install-dev
 install-dev: ## Install development dependencies.
 	@echo "Installing development dependencies..."
-	@python -m pip install ".[dev, onnxruntime]"
+	@python -m pip install ".[dev, onnxruntime]" -U
 	@pre-commit install
 
 .PHONY: lint

README.md

@@ -12,7 +12,7 @@ stars](https://img.shields.io/github/stars/protectai/llm-guard.svg?style=social&
 [![Downloads](https://static.pepy.tech/badge/llm-guard)](https://pepy.tech/project/llm-guard)
 [![Downloads](https://static.pepy.tech/badge/llm-guard/month)](https://pepy.tech/project/llm-guard)
 
-<a href="https://join.slack.com/t/laiyerai/shared_invite/zt-28jv3ci39-sVxXrLs3rQdaN3mIl9IT~w"><img src="https://github.com/protectai/llm-guard/blob/main/docs/assets/join-our-slack-community.png?raw=true" width="200" alt="Join Our Slack Community"></a>
+<a href="https://mlsecops.com/slack"><img src="https://github.com/protectai/llm-guard/blob/main/docs/assets/join-our-slack-community.png?raw=true" width="200" alt="Join Our Slack Community"></a>
 
 ## What is LLM Guard?
 
@@ -108,7 +108,7 @@ we would love to have you as part of our community.
 Join our Slack to give us feedback, connect with the maintainers and fellow users, ask questions,
 get help for package usage or contributions, or engage in discussions about LLM security!
 
-<a href="https://join.slack.com/t/laiyerai/shared_invite/zt-28jv3ci39-sVxXrLs3rQdaN3mIl9IT~w"><img src="https://github.com/protectai/llm-guard/blob/main/docs/assets/join-our-slack-community.png?raw=true" width="200" alt="Join Our Slack Community"></a>
+<a href="https://mlsecops.com/slack"><img src="https://github.com/protectai/llm-guard/blob/main/docs/assets/join-our-slack-community.png?raw=true" width="200" alt="Join Our Slack Community"></a>
 
 ### Production Support
 

docs/index.md

@@ -4,7 +4,7 @@ LLM Guard by [Protect AI](https://protectai.com/llm-guard) is a comprehensive to
 
 [**Playground**](https://huggingface.co/spaces/ProtectAI/llm-guard-playground) | [**Changelog**](./changelog.md)
 
-<a href="https://join.slack.com/t/laiyerai/shared_invite/zt-28jv3ci39-sVxXrLs3rQdaN3mIl9IT~w"><img src="./assets/join-our-slack-community.png" width="200" alt="Join Our Slack Community"></a>
+<a href="https://mlsecops.com/slack"><img src="./assets/join-our-slack-community.png" width="200" alt="Join Our Slack Community"></a>
 
 ## What is LLM Guard?
 
@@ -55,4 +55,4 @@ we would love to have you as part of our community.
 Join our Slack to give us feedback, connect with the maintainers and fellow users, ask questions,
 get help for package usage or contributions, or engage in discussions about LLM security!
 
-<a href="https://join.slack.com/t/laiyerai/shared_invite/zt-28jv3ci39-sVxXrLs3rQdaN3mIl9IT~w"><img src="./assets/join-our-slack-community.png" width="200" alt="Join Our Slack Community"></a>
+<a href="https://mlsecops.com/slack"><img src="./assets/join-our-slack-community.png" width="200" alt="Join Our Slack Community"></a>

llm_guard/input_scanners/anonymize_helpers/analyzer.py

@@ -93,7 +93,7 @@ def _add_recognizers(
     return registry
 
 
-def _get_nlp_engine(languages: list[str] = ["en"]) -> NlpEngine:
+def _get_nlp_engine(languages: list[str]) -> NlpEngine:
     models = []
 
     for language in languages:
@@ -143,7 +143,7 @@ def get_analyzer(
 ) -> AnalyzerEngine:
     nlp_engine = _get_nlp_engine(languages=supported_languages)
 
-    registry = RecognizerRegistry()
+    registry = RecognizerRegistry(supported_languages=supported_languages)
     registry.load_predefined_recognizers(nlp_engine=nlp_engine)
     registry = _add_recognizers(registry, regex_groups, custom_names, supported_languages)
     registry.add_recognizer(recognizer)

llm_guard/input_scanners/anonymize_helpers/transformers_recognizer.py

@@ -5,7 +5,7 @@
 
 from presidio_analyzer import AnalysisExplanation, EntityRecognizer, RecognizerResult
 from presidio_analyzer.nlp_engine import NlpArtifacts
-from transformers import TokenClassificationPipeline
+from transformers.pipelines.token_classification import TokenClassificationPipeline
 
 from llm_guard.model import Model
 from llm_guard.transformers_helpers import get_tokenizer_and_model_for_ner
@@ -60,7 +60,7 @@ class TransformersRecognizer(EntityRecognizer):
     chunk_length: int
     id_entity_name: str
     id_score_reduction: float
-    pipeline: transformers.Pipeline | None
+    pipeline: transformers.pipelines.base.Pipeline | None
 
     def load(self) -> None:
         pass
@@ -139,7 +139,7 @@ def _load_pipeline(
         self.model.pipeline_kwargs["ignore_labels"] = self.ignore_labels
 
         transformers = cast("transformers", lazy_load_dep("transformers"))
-        self.pipeline = transformers.pipeline(
+        self.pipeline = transformers.pipelines.pipeline(
             "ner",
             model=tf_model,
             tokenizer=tf_tokenizer,

llm_guard/input_scanners/prompt_injection.py

@@ -2,7 +2,8 @@
 
 from enum import Enum
 
-from transformers import PreTrainedTokenizer, PreTrainedTokenizerFast
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
 
 from llm_guard.model import Model
 from llm_guard.transformers_helpers import get_tokenizer_and_model_for_classification, pipeline

llm_guard/transformers_helpers.py

@@ -4,13 +4,6 @@
 from functools import lru_cache
 from typing import Literal, get_args
 
-from transformers import (
-    PreTrainedModel,
-    PreTrainedTokenizer,
-    PreTrainedTokenizerFast,
-    TFPreTrainedModel,
-)
-
 from .exception import LLMGuardValidationError
 from .model import Model
 from .util import device, get_logger, lazy_load_dep
@@ -155,8 +148,8 @@ def get_tokenizer_and_model_for_ner(
 
 def pipeline(
     task: str,
-    model: PreTrainedModel | TFPreTrainedModel,
-    tokenizer: PreTrainedTokenizer | PreTrainedTokenizerFast,
+    model,
+    tokenizer,
     **kwargs,
 ):
     if task not in get_args(ClassificationTask):

mkdocs.yml

@@ -88,7 +88,7 @@ extra:
     - icon: fontawesome/brands/linkedin
       link: https://www.linkedin.com/company/protect-ai
     - icon: fontawesome/brands/slack
-      link: https://join.slack.com/t/laiyerai/shared_invite/zt-28jv3ci39-sVxXrLs3rQdaN3mIl9IT~w
+      link: https://mlsecops.com/slack
     - icon: fontawesome/solid/globe
       link: https://protectai.com/llm-guard
   analytics:

pyproject.toml

@@ -20,46 +20,44 @@ classifiers = [
 requires-python = ">=3.10,<3.13"
 
 dependencies = [
-  "bc-detect-secrets==1.5.36",
-  "faker>=36,<37",
+  "bc-detect-secrets==1.5.43",
+  "faker>=37,<38",
   "fuzzysearch>=0.7,<0.9",
-  "json-repair>=0.39,<0.40",
+  "json-repair==0.44.1",
   "nltk>=3.9.1,<4",
-  "presidio-analyzer==2.2.354",
-  "presidio-anonymizer==2.2.354",
+  "presidio-analyzer==2.2.358",
+  "presidio-anonymizer==2.2.358",
   "regex==2024.11.6",
   "tiktoken>=0.9,<1.0",
   "torch>=2.4.0",
-  "transformers==4.48.3",
-  "structlog>=24",
-  "oldest-supported-numpy",
-  "spacy-pkuseg==1.0.0"
+  "transformers==4.51.3",
+  "structlog>=24"
 ]
 
 [project.optional-dependencies]
 onnxruntime = [
-  "optimum[onnxruntime]==1.24.0",
+  "optimum[onnxruntime]==1.25.2",
 ]
 onnxruntime-gpu = [
-  "optimum[onnxruntime-gpu]==1.24.0",
+  "optimum[onnxruntime-gpu]==1.25.2",
 ]
 docs-dev = [
   "mkdocs>=1.6,<2",
-  "mkdocs-autorefs==1.3.1",
-  "mkdocs-git-revision-date-localized-plugin>=1.3.0",
+  "mkdocs-autorefs==1.4.1",
+  "mkdocs-git-revision-date-localized-plugin>=1.4.0",
   "mkdocs-jupyter>=0.25.1",
-  "mkdocs-material>=9.6.5",
+  "mkdocs-material>=9.6.14",
   "mkdocs-material-extensions>=1.3.1",
-  "mkdocs-swagger-ui-tag>=0.6.11",
+  "mkdocs-swagger-ui-tag>=0.7.1",
 ]
 dev = [
   "llm_guard[docs-dev]",
   "autoflake>=2,<3",
-  "pytest>=8.0.0,<9",
-  "pytest-cov>=5.0.0,<6",
-  "pre-commit>=3.8,<5",
-  "pyright~=1.1.394",
-  "ruff==0.9.6",
+  "pytest>=8.3.5,<9",
+  "pytest-cov>=6.1.1,<7",
+  "pre-commit>=4.2.0,<5",
+  "pyright~=1.1.400",
+  "ruff==0.11.10",
 ]
 
 [project.urls]