utils.py

import io
from cairosvg import svg2png
from PIL import Image
# import base64

def get_entity_explanation(label):
    """Return explanation for named entity labels"""
    explanations = {
        'PERSON': 'People, including fictional',
        'NORP': 'Nationalities, religious or political groups',
        'FAC': 'Buildings, airports, highways, bridges, etc.',
        'ORG': 'Companies, agencies, institutions, etc.',
        'GPE': 'Countries, cities, states',
        'LOC': 'Non-GPE locations, mountain ranges, water bodies',
        'PRODUCT': 'Objects, vehicles, foods, etc.',
        'EVENT': 'Named hurricanes, battles, wars, sports events, etc.',
        'WORK_OF_ART': 'Titles of books, songs, etc.',
        'DATE': 'Absolute or relative dates or periods',
        'TIME': 'Times smaller than a day',
        'MONEY': 'Monetary values, including unit',
        'QUANTITY': 'Measurements, as of weight or distance'
    }
    return explanations.get(label, 'Other type of entity')

def analyze_text(nlp, text):
    doc = nlp(text)
    
    # Basic tokenization and POS analysis
    tokens = [{
        'Text': token.text,
        'Lemma': token.lemma_,
        'POS': token.pos_,
        'Tag': token.tag_,
        'Dependency': token.dep_,
        'Shape': token.shape_,
        'Is Alpha': token.is_alpha,
        'Is Stop': token.is_stop
    } for token in doc]
    
    # Named Entity Recognition
    entities = [{
        'Text': ent.text,
        'Label': ent.label_,
        'Explanation': get_entity_explanation(ent.label_),
        'Start': ent.start_char,
        'End': ent.end_char
    } for ent in doc.ents]
    
    # Noun Chunks (phrases)
    noun_chunks = [{
        'Text': chunk.text,
        'Root Text': chunk.root.text,
        'Root Dep': chunk.root.dep_,
        'Root Head Text': chunk.root.head.text
    } for chunk in doc.noun_chunks]
    
    # Text Statistics
    stats = {
        'Word Count': len([token for token in doc if not token.is_punct]),
        'Sentence Count': len(list(doc.sents)),
        'Average Words per Sentence': round(len([token for token in doc if not token.is_punct]) / len(list(doc.sents)), 2),
        'Unique Words': len(set([token.text.lower() for token in doc if token.is_alpha])),
        'Stop Words %': round(len([token for token in doc if token.is_stop]) / len(doc) * 100, 2)
    }
    
    return tokens, entities, noun_chunks, stats, doc

def svg_to_png(svg_content, background_color='white'):
    """Convert SVG to PNG with specified background color"""
    # Split multiple SVGs if present
    svg_parts = svg_content.split('<br><br>')
    images = []
    
    for svg in svg_parts:
        # Add SVG namespace if missing
        if not 'xmlns="http://www.w3.org/2000/svg"' in svg:
            svg = svg.replace('<svg', '<svg xmlns="http://www.w3.org/2000/svg"')
        
        try:
            # Convert SVG to PNG bytes
            png_bytes = svg2png(bytestring=svg.encode('utf-8'),
                              background_color=background_color,
                              scale=1)
            
            # Create PIL Image from PNG bytes
            img = Image.open(io.BytesIO(png_bytes))
            
            # Convert RGBA to RGB with white background
            if img.mode == 'RGBA':
                background = Image.new('RGB', img.size, background_color)
                background.paste(img, mask=img.split()[3])  # Use alpha channel as mask
                img = background
            
            # Add some padding
            padding = 20  # pixels
            img_with_padding = Image.new('RGB', 
                                       (img.width, img.height + padding * 2),
                                       background_color)
            img_with_padding.paste(img, (0, padding))
            images.append(img_with_padding)
            
        except Exception as e:
            st.error(f"Error converting SVG to PNG: {str(e)}")
            continue
    
    if not images:
        return None
    
    # Combine images vertically if there are multiple
    if len(images) > 1:
        # Calculate total height and max width
        total_height = sum(img.height for img in images)
        max_width = max(img.width for img in images)
        
        # Create new image to hold all sentences
        combined = Image.new('RGB', (max_width, total_height), background_color)
        
        # Paste each image
        y_offset = 0
        for img in images:
            # Center image horizontally
            x_offset = (max_width - img.width) // 2
            combined.paste(img, (x_offset, y_offset))
            y_offset += img.height
    else:
        combined = images[0]
    
    # Convert to bytes for Streamlit
    img_byte_arr = io.BytesIO()
    combined.save(img_byte_arr, format='PNG')
    img_byte_arr.seek(0)
    
    return img_byte_arr.getvalue()