contentintelpy 0.1.0__tar.gz → 0.1.1__tar.gz

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: contentintelpy
-Version: 0.1.0
+Version: 0.1.1
 Summary: Production-grade NLP library for unified content intelligence.
 Author-email: Ronit Fulari <ronitfulari31@gmail.com>
 License: MIT
@@ -16,6 +16,7 @@ Provides-Extra: core
 Requires-Dist: transformers<5.0.0,>=4.30.0; extra == "core"
 Requires-Dist: torch<3.0.0,>=2.0.0; extra == "core"
 Requires-Dist: sentence-transformers>=2.2.0; extra == "core"
+Requires-Dist: scikit-learn>=1.0.0; extra == "core"
 Provides-Extra: ner
 Requires-Dist: spacy>=3.7.0; extra == "ner"
 Requires-Dist: gliner>=0.1.0; extra == "ner"

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy/nodes/keyword_extract_node.py

@@ -1,8 +1,7 @@
 from ..pipeline.base_node import Node
 from ..pipeline.context import PipelineContext
 from ..utils.model_registry import registry
-from sklearn.metrics.pairwise import cosine_similarity
-from sklearn.feature_extraction.text import CountVectorizer
+from ..utils.lazy_import import ensure_dependency
 import numpy as np
 import logging
 import itertools
@@ -30,6 +29,10 @@ class KeywordExtractionNode(Node):
             return context
 
         try:
+            ensure_dependency("sklearn", "core")
+            from sklearn.metrics.pairwise import cosine_similarity
+            from sklearn.feature_extraction.text import CountVectorizer
+
             model = registry.get_embedding_model()
 
             # 1. Candidate Generation (using simple CountVectorizer)

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy/nodes/summarization_node.py

@@ -3,12 +3,7 @@ from ..pipeline.context import PipelineContext
 from ..utils.model_registry import registry
 import logging
 
-# Sumy imports
-from sumy.parsers.plaintext import PlaintextParser
-from sumy.nlp.tokenizers import Tokenizer
-from sumy.summarizers.lsa import LsaSummarizer as Summarizer
-from sumy.nlp.stemmers import Stemmer
-from sumy.utils import get_stop_words
+from ..utils.lazy_import import ensure_dependency
 
 logger = logging.getLogger("contentintelpy.nodes.summarization")
 
@@ -48,6 +43,13 @@ class SummarizationNode(Node):
         # 2. Fallback: Sumy (LSA)
         if not summary_text:
             try:
+                ensure_dependency("sumy", "summarization")
+                from sumy.parsers.plaintext import PlaintextParser
+                from sumy.nlp.tokenizers import Tokenizer
+                from sumy.summarizers.lsa import LsaSummarizer as Summarizer
+                from sumy.nlp.stemmers import Stemmer
+                from sumy.utils import get_stop_words
+
                 parser = PlaintextParser.from_string(text, Tokenizer("english"))
                 stemmer = Stemmer("english")
                 summarizer = Summarizer(stemmer)

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy/nodes/translation_node.py

@@ -1,9 +1,8 @@
 from ..pipeline.base_node import Node
 from ..pipeline.context import PipelineContext
 from ..utils.model_registry import registry
+from ..utils.lazy_import import ensure_dependency
 import logging
-import argostranslate.package
-import argostranslate.translate
 
 logger = logging.getLogger("contentintelpy.nodes.translation")
 
@@ -59,6 +58,10 @@ class TranslationNode(Node):
         # 3. Fallback: ArgosTranslate (Offline)
         if not translated_text:
             try:
+                argostranslate = ensure_dependency("argostranslate", "translation")
+                import argostranslate.package
+                import argostranslate.translate
+                
                 logger.info("Attempting translation with ArgosTranslate...")
                 # Argos requires ensuring packages are installed
                 # This is a blocking network call on first run if not present

contentintelpy-0.1.1/contentintelpy/utils/lazy_import.py

@@ -0,0 +1,16 @@
+import importlib
+import logging
+
+logger = logging.getLogger("contentintelpy.utils")
+
+def ensure_dependency(module_name: str, extra_name: str):
+    """
+    Attempts to import a module. If it fails, raises an ImportError with a 
+    helpful message directing the user to install the appropriate optional extra.
+    """
+    try:
+        return importlib.import_module(module_name)
+    except (ImportError, ModuleNotFoundError):
+        msg = f"Missing optional dependency '{module_name}'. To use this feature, please install it using: pip install \"contentintelpy[{extra_name}]\""
+        logger.error(msg)
+        raise ImportError(msg) from None

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy/utils/model_registry.py

@@ -1,3 +1,4 @@
+from .lazy_import import ensure_dependency
 import logging
 import threading
 from typing import Any, Optional
@@ -50,6 +51,8 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_sentiment_pipeline(self):
         def _loader():
+            ensure_dependency("transformers", "core")
+            ensure_dependency("torch", "core")
             from transformers import pipeline
             # Use a high-quality multilingual or English sentiment model
             # CardiffNLP is standard for Twitter-like text, widely used
@@ -63,6 +66,7 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_translation_pipeline(self):
         def _loader():
+            ensure_dependency("transformers", "core")
             from transformers import pipeline
             # NLLB-200 Distilled (600M) is a good balance of size/quality
             model_name = "facebook/nllb-200-distilled-600M"
@@ -75,6 +79,7 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_gliner_model(self):
         def _loader():
+            ensure_dependency("gliner", "ner")
             from gliner import GLiNER
             # Standard GLiNER model
             return GLiNER.from_pretrained("urchade/gliner_large-v2.1")
@@ -86,6 +91,7 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_classifier_pipeline(self):
         def _loader():
+            ensure_dependency("transformers", "core")
             from transformers import pipeline
             return pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
         
@@ -96,6 +102,7 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_summarization_pipeline(self):
         def _loader():
+            ensure_dependency("transformers", "core")
             from transformers import pipeline
             return pipeline("summarization", model="facebook/bart-large-cnn")
         
@@ -106,6 +113,7 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_language_detector(self):
         def _loader():
+            ensure_dependency("transformers", "core")
             from transformers import pipeline
             return pipeline("text-classification", model="qanastek/51-languages-classifier")
         
@@ -116,6 +124,7 @@ class ModelRegistry:
     # --------------------------------------------------------------------------
     def get_embedding_model(self):
         def _loader():
+            ensure_dependency("sentence_transformers", "core")
             from sentence_transformers import SentenceTransformer
             # Fast, effective embedding model for semantic similarity
             return SentenceTransformer('all-MiniLM-L6-v2')

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: contentintelpy
-Version: 0.1.0
+Version: 0.1.1
 Summary: Production-grade NLP library for unified content intelligence.
 Author-email: Ronit Fulari <ronitfulari31@gmail.com>
 License: MIT
@@ -16,6 +16,7 @@ Provides-Extra: core
 Requires-Dist: transformers<5.0.0,>=4.30.0; extra == "core"
 Requires-Dist: torch<3.0.0,>=2.0.0; extra == "core"
 Requires-Dist: sentence-transformers>=2.2.0; extra == "core"
+Requires-Dist: scikit-learn>=1.0.0; extra == "core"
 Provides-Extra: ner
 Requires-Dist: spacy>=3.7.0; extra == "ner"
 Requires-Dist: gliner>=0.1.0; extra == "ner"

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy.egg-info/SOURCES.txt

@@ -22,4 +22,5 @@ contentintelpy/services/ner_service.py
 contentintelpy/services/sentiment_service.py
 contentintelpy/services/summarization_service.py
 contentintelpy/services/translation_service.py
+contentintelpy/utils/lazy_import.py
 contentintelpy/utils/model_registry.py

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/contentintelpy.egg-info/requires.txt

@@ -5,6 +5,7 @@ tqdm>=4.66.0
 transformers<5.0.0,>=4.30.0
 torch<3.0.0,>=2.0.0
 sentence-transformers>=2.2.0
+scikit-learn>=1.0.0
 
 [dev]
 pytest

{contentintelpy-0.1.0 → contentintelpy-0.1.1}/pyproject.toml

@@ -4,10 +4,10 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "contentintelpy"
-version = "0.1.0"
+version = "0.1.1"
 description = "Production-grade NLP library for unified content intelligence."
 authors = [
-  { name = "Ronit Fulari", email = "ronitfulari31@gmail.com" },
+   { name = "Ronit Fulari", email = "ronitfulari31@gmail.com" },
 ]
 license = { text = "MIT" }
 readme = "README.md"
@@ -26,7 +26,8 @@ dependencies = [
 core = [
   "transformers>=4.30.0,<5.0.0",
   "torch>=2.0.0,<3.0.0",
-  "sentence-transformers>=2.2.0"
+  "sentence-transformers>=2.2.0",
+  "scikit-learn>=1.0.0"
 ]
 
 ner = [