Romeo and Juliet Text Extraction with LangExtract¶

This notebook demonstrates extracting characters, emotions, and relationships from Shakespeare's Romeo and Juliet using LangExtract.

Setup¶

In [8]:

# Install LangExtract
%pip install -q langextract

Note: you may need to restart the kernel to use updated packages.

In [9]:

# Set up your Gemini API key
# Get your key from: https://aistudio.google.com/app/apikey
import os
from getpass import getpass

if 'GEMINI_API_KEY' not in os.environ:
    os.environ['GEMINI_API_KEY'] = getpass('Enter your Gemini API key: ')

Define Extraction Task¶

In [10]:

import langextract as lx
import textwrap

# Define the extraction task
prompt = textwrap.dedent("""\
    Extract characters, emotions, and relationships in order of appearance.
    Use exact text for extractions. Do not paraphrase or overlap entities.
    Provide meaningful attributes for each entity to add context.""")

# Provide a high-quality example
examples = [
    lx.data.ExampleData(
        text="ROMEO. But soft! What light through yonder window breaks? It is the east, and Juliet is the sun.",
        extractions=[
            lx.data.Extraction(
                extraction_class="character",
                extraction_text="ROMEO",
                attributes={"emotional_state": "wonder"}
            ),
            lx.data.Extraction(
                extraction_class="emotion",
                extraction_text="But soft!",
                attributes={"feeling": "gentle awe"}
            ),
            lx.data.Extraction(
                extraction_class="relationship",
                extraction_text="Juliet is the sun",
                attributes={"type": "metaphor"}
            ),
        ]
    )
]

Extract from Sample Text¶

In [11]:

# Simple extraction from a short text
input_text = "Lady Juliet gazed longingly at the stars, her heart aching for Romeo"

result = lx.extract(
    text_or_documents=input_text,
    prompt_description=prompt,
    examples=examples,
    model_id="gemini-2.5-flash",
)

# Display results
print(f"Extracted {len(result.extractions)} entities:\n")
for extraction in result.extractions:
    print(f"• {extraction.extraction_class}: '{extraction.extraction_text}'")
    if extraction.attributes:
        for key, value in extraction.attributes.items():
            print(f"  - {key}: {value}")

LangExtract: model=gemini-2.5-flash, current=68 chars, processed=68 chars:  [00:01]

✓ Extraction processing complete
✓ Extracted 3 entities (3 unique types)
  • Time: 1.96s
  • Speed: 35 chars/sec
  • Chunks: 1
Extracted 3 entities:

• character: 'Lady Juliet'
  - emotional_state: longing
• emotion: 'gazed longingly at the stars, her heart aching'
  - feeling: melancholy longing
• relationship: 'her heart aching for Romeo'
  - type: romantic

Interactive Visualization¶

In [12]:

# Save results to JSONL
lx.io.save_annotated_documents([result], output_name="romeo_juliet.jsonl", output_dir=".")

# Generate interactive visualization
html_content = lx.visualize("romeo_juliet.jsonl")

# Display in notebook
print("Interactive visualization (hover over highlights to see attributes):")
html_content

LangExtract: Saving to romeo_juliet.jsonl: 1 docs [00:00, 995.33 docs/s]

✓ Saved 1 documents to romeo_juliet.jsonl

LangExtract: Loading romeo_juliet.jsonl: 100%|██████████| 961/961 [00:00<00:00, 2.49MB/s]

✓ Loaded 1 documents from romeo_juliet.jsonl
Interactive visualization (hover over highlights to see attributes):

Out[12]:

Highlights Legend: character emotion relationship

Lady Juliet gazed longingly at the stars, her heart aching for Romeo

Entity 1/3 | Pos [0-11]

In [13]:

# Save visualization to file (for downloading)
with open("romeo_juliet_visualization.html", "w") as f:
    # Handle both Jupyter (HTML object) and non-Jupyter (string) environments
    if hasattr(html_content, 'data'):
        f.write(html_content.data)
    else:
        f.write(html_content)

print("✓ Visualization saved to romeo_juliet_visualization.html")
print("You can download this file from the Files panel on the left.")

✓ Visualization saved to romeo_juliet_visualization.html
You can download this file from the Files panel on the left.

Try Your Own Text¶

Experiment with your own Shakespeare quotes or any literary text!

In [14]:

# Try your own text
your_text = """
JULIET: O Romeo, Romeo! wherefore art thou Romeo?
Deny thy father and refuse thy name;
Or, if thou wilt not, be but sworn my love,
And I'll no longer be a Capulet.
"""

custom_result = lx.extract(
    text_or_documents=your_text,
    prompt_description=prompt,
    examples=examples,
    model_id="gemini-2.5-flash",
)

print("Extractions from your text:\n")
for e in custom_result.extractions:
    print(f"• {e.extraction_class}: '{e.extraction_text}'")
    if e.attributes:
        for key, value in e.attributes.items():
            print(f"  - {key}: {value}")

LangExtract: model=gemini-2.5-flash, current=163 chars, processed=163 chars:  [00:05]

✓ Extraction processing complete
✓ Extracted 6 entities (3 unique types)
  • Time: 5.84s
  • Speed: 28 chars/sec
  • Chunks: 1
Extractions from your text:

• character: 'JULIET'
  - emotional_state: longing
• emotion: 'O Romeo, Romeo! wherefore art thou Romeo?'
  - feeling: desperate questioning
• relationship: 'thy father'
  - type: familial
• relationship: 'thy name'
  - type: lineage
• relationship: 'my love'
  - type: romantic bond
• relationship: 'Capulet'
  - type: family affiliation