Add A Custom Component To The NLP Pipeline Using Spacy

import spacy
from spacy.tokens import Doc
from spacy.language import Language

# Step 1: Register a custom Doc-level attribute
Doc.set_extension("has_email", default=False)

# Step 2: Define and register the custom pipeline component
@Language.component("email_detector")
def email_detector(doc):
    """
    Custom pipeline component that checks if a Doc contains an email address.
    If found, sets doc._.has_email = True, otherwise False.
    """
    for token in doc:
        if "@" in token.text:
            doc._.has_email = True
            break
    else:
        doc._.has_email = False
    return doc

# Step 3: Load a spaCy model
nlp = spacy.load("en_core_web_sm")

# Step 4: Add the custom component to the pipeline using its registered name
nlp.add_pipe("email_detector", last=True)

# Step 5: Run the pipeline on sample text
doc1 = nlp("Contact us at support@example.com for more info.")
doc2 = nlp("This text has no email address.")

# Step 6: Print results
print(f"Text: {doc1.text}")
print(f"Contains email? {doc1._.has_email}")

print(f"Text: {doc2.text}")
print(f"Contains email? {doc2._.has_email}")

# Step 7: Inspect pipeline order
print("\nPipeline components:", nlp.pipe_names)

Query/Feedback