exc_02_15.py

import spacy
from spacy.matcher import PhraseMatcher
from spacy.tokens import Span
import json

with open("exercises/en/countries.json", encoding="utf8") as f:
    COUNTRIES = json.loads(f.read())
with open("exercises/en/country_text.txt", encoding="utf8") as f:
    TEXT = f.read()

nlp = spacy.load("en_core_web_sm")
matcher = PhraseMatcher(nlp.vocab)
patterns = list(nlp.pipe(COUNTRIES))
matcher.add("COUNTRY", patterns)

# Create a doc and reset existing entities
doc = nlp(TEXT)
doc.ents = []

# Iterate over the matches
for match_id, start, end in matcher(doc):
    # Create a Span with the label for "GPE"
    span = ____(____, ____, ____, label=____)

    # Overwrite the doc.ents and add the span
    doc.ents = list(doc.ents) + [____]

    # Get the span's root head token
    span_root_head = ____.____.____
    # Print the text of the span root's head token and the span text
    print(span_root_head.____, "-->", span.text)

# Print the entities in the document
print([(ent.text, ent.label_) for ent in doc.ents if ent.label_ == "GPE"])