add pipe loader

textdescriptives/load_components.py

@@ -1,4 +1,4 @@
-"""Adds components to a spaCy pipeline"""
+"""Adds all components to a spaCy pipeline"""
 from components import (
     create_readability_component,
     create_dependency_distance_component,
@@ -9,9 +9,8 @@
 from spacy.tokens import Doc
 
 
-@Language.component("textdescriptives")
-def textdescriptives_component(nlp: Language, name: str):
-    """Add all components to a spaCy pipeline"""
+@Language.factory("textdescriptives")
+def create_textdescriptives_component(nlp: Language, name: str):
     for component in [
         "descriptive_stats",
         "readability",
@@ -22,33 +21,26 @@ def textdescriptives_component(nlp: Language, name: str):
 
 
 class TextDescriptives:
+    """
+    Utility spaCy v3.0 component to add all functionality from the TextDescriptives
+    package to a `Doc` objects. See `DescriptiveStatistics`, `Readability`, and `DependencyDistance`
+    for more details.
+
+    USAGE:
+        >>> import spacy
+        >>> from textdescriptives import TextDescriptives
+        >>> nlp = spacy.load("en_core_web_sm")
+        >>> nlp.add_pipe("textdescriptives")
+        >>> doc = nlp("This is a the first sentence. Here's a second, slightly longer one.")
+        >>> doc._.dependency_distance
+        >>> doc._.readability
+        >>> doc[0:4]._.token_length
+    """
+
     def __init__(self, nlp: Language):
-        """Don't do anything except initialise the other components"""
+        """Don't do anything, just return the Doc"""
 
     def __call__(self, doc: Doc):
         """Run the pipeline component"""
         return doc
 
-
-# def add_components(nlp: Language):
-#     """Add all components to a spaCy pipeline"""
-#     for component in [
-#         "utilities",
-#         "descriptive_stats",
-#         "readability",
-#         "dependency_distance",
-#     ]:
-#         nlp.add_pipe(component, last=True)
-#     return nlp
-
-
-import spacy
-
-nlp = spacy.load("da_core_news_sm")
-nlp = nlp.add_pipe("textdescriptives")
-nlp.add_pipe("descriptive_stats")
-doc = nlp("Det her er en fin lille sætning. Ja, den er bare rigtig sød.")
-print(doc._.counts)
-print(doc[0:4]._.token_length)
-print(doc._.readability)
-print(doc._.dependency_distance)