contentintelpy 0.1.0__py3-none-any.whl → 0.1.1__py3-none-any.whl

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/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/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/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/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.dist-info → contentintelpy-0.1.1.dist-info}/METADATA

@@ -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.dist-info → contentintelpy-0.1.1.dist-info}/RECORD

@@ -1,12 +1,12 @@
 contentintelpy/__init__.py,sha256=HSJcTru6PksguTH52iFj7ORhqSuHXIDHMQotj5KMmBc,1977
 contentintelpy/nodes/classification_node.py,sha256=ivfAHdYXZU-5eVbtgxD94_TiRHcq-mJg4ukOc7KqwXU,2116
-contentintelpy/nodes/keyword_extract_node.py,sha256=g_oERVXfE2VYVqFeqT7J2DIq-By1qXP7XsVOQc5R9mk,2897
+contentintelpy/nodes/keyword_extract_node.py,sha256=o-OTOdkdEIYnd_ZczdAjwJkWnjzlKLa28WnEZCKy_og,3024
 contentintelpy/nodes/language_node.py,sha256=sKRa65kLrb1IRYGrkT82tu8LgdhIXdN5EwhUrH6pSqI,1971
 contentintelpy/nodes/location_node.py,sha256=U3YQ31KclWNeoyrorodBAzAEd7zLmI31Deu72Viw1M0,1579
 contentintelpy/nodes/ner_node.py,sha256=8DRg7NVpz8ZXcobgwYZsWkNOvaFfIj_ZEWG8wJckqus,1632
 contentintelpy/nodes/sentiment_node.py,sha256=oFuw1Z0d6f4BWSYtnp8UN0gMCvL3nl5b4h68t6qv-cQ,2706
-contentintelpy/nodes/summarization_node.py,sha256=kGLM4ssVd01h5hm4Cc5xpg73feCumg6b_hQRAi5gos4,2626
-contentintelpy/nodes/translation_node.py,sha256=vd_RmkKmm07TynfKQdPvSwXeaG2sA8v28iADN1yzbmc,4325
+contentintelpy/nodes/summarization_node.py,sha256=XHdT8lMN-1_fiGBGkqbblPfJgkvQTf7XIq99z7w_6Bw,2803
+contentintelpy/nodes/translation_node.py,sha256=zebwFqp22bo4Oi5gKyNGZKduAl_grfZTC53PKPsKVko,4511
 contentintelpy/pipeline/base_node.py,sha256=hYLx2yAURpbmTr9x4kG8qVIlNI1Q0UJckBltW5LJl-o,1394
 contentintelpy/pipeline/context.py,sha256=u_YsEe4oi-A6MM9igtQ0cOeX88fRd_Uj9umU4040W0E,1257
 contentintelpy/pipeline/pipeline.py,sha256=gTgRcF34KxAJMxtac7wHdesD33q3CIP9hncvILHQ-3c,888
@@ -14,9 +14,10 @@ contentintelpy/services/ner_service.py,sha256=7-sEAqxYRpVksd-sZ5CPgAq3HfVeeb0OaR
 contentintelpy/services/sentiment_service.py,sha256=Yc6u0l8m_uN5ZxgUMr9DQziwi50cMlTZuaAOS8A7pJc,1130
 contentintelpy/services/summarization_service.py,sha256=XK3vAGGoQS1dXxaO4nKjyrFlWwN_wZKY2qFNcDJ9IIM,748
 contentintelpy/services/translation_service.py,sha256=6yNLLJ7mAE7ptHvprX1JUoUN-65Ot7ZdTszqqxMY1TA,1191
-contentintelpy/utils/model_registry.py,sha256=OyixstAVsvQ-nkoICeZykdwcPDtuBZRtPHWmoIjfi2o,5344
-contentintelpy-0.1.0.dist-info/licenses/LICENSE,sha256=lZ8hT4isGfdFVxdD7gDRnt3RJqyrkO1L5GseyN3A9hM,1092
-contentintelpy-0.1.0.dist-info/METADATA,sha256=uqVJA361j3FW1xVekYg5dvbXXaOr2zOWZ073ks82lXM,4271
-contentintelpy-0.1.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-contentintelpy-0.1.0.dist-info/top_level.txt,sha256=sxoE-r2-frUi3qwADEiYcFFxZW5hMI1Mjw87hcGMulQ,15
-contentintelpy-0.1.0.dist-info/RECORD,,
+contentintelpy/utils/lazy_import.py,sha256=km1_WmFYKBGJwNbB1lEJSiix3Rnr1RwMbW7MbrMtqf4,661
+contentintelpy/utils/model_registry.py,sha256=e8KK1AiZ1jJvsxNwapQjNLKiCeETZd8gWJKaoO-Bl_E,5823
+contentintelpy-0.1.1.dist-info/licenses/LICENSE,sha256=lZ8hT4isGfdFVxdD7gDRnt3RJqyrkO1L5GseyN3A9hM,1092
+contentintelpy-0.1.1.dist-info/METADATA,sha256=SbRCE3Q7PYnIfFk1uNuJeE0Vv0TEtkeQLD8_bk97h5M,4324
+contentintelpy-0.1.1.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+contentintelpy-0.1.1.dist-info/top_level.txt,sha256=sxoE-r2-frUi3qwADEiYcFFxZW5hMI1Mjw87hcGMulQ,15
+contentintelpy-0.1.1.dist-info/RECORD,,