Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,284 @@
-import altair as alt
-import numpy as np
-import pandas as pd
 import streamlit as st
+import syllables
+import re
+import random
+from typing import List, Tuple
 
-"""
-# Welcome to Streamlit!
-
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
-
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
-
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))
+# Configure page
+st.set_page_config(
+    page_title="Text Syllabification Tool",
+    page_icon="🔤",
+    layout="wide"
+)
+
+def generate_color_palette(num_colors: int) -> List[str]:
+    """Generate distinct colors for syllables"""
+    colors = [
+        "#FF6B6B", "#4ECDC4", "#45B7D1", "#96CEB4", "#FFEAA7",
+        "#DDA0DD", "#98D8C8", "#F7DC6F", "#BB8FCE", "#85C1E9",
+        "#F8C471", "#82E0AA", "#F1948A", "#85C1E9", "#D7BDE2",
+        "#A9DFBF", "#F9E79F", "#D5A6BD", "#AED6F1", "#ABEBC6"
+    ]
+    
+    # If we need more colors than predefined, generate random ones
+    while len(colors) < num_colors:
+        color = f"#{random.randint(0, 255):02x}{random.randint(0, 255):02x}{random.randint(0, 255):02x}"
+        colors.append(color)
+    
+    return colors[:num_colors]
+
+def syllabify_word(word: str) -> List[str]:
+    """Syllabify a single word using the syllables library"""
+    # Clean the word of punctuation for syllabification
+    clean_word = re.sub(r'[^\w]', '', word)
+    
+    if not clean_word:
+        return [word]
+    
+    try:
+        # Get syllable count
+        syl_count = syllables.estimate(clean_word)
+        
+        # For basic syllabification, we'll use a simple approach
+        # The syllables library mainly provides counts, so we'll implement basic rules
+        syllable_list = simple_syllabify(clean_word)
+        
+        # If our simple method doesn't match the count, adjust
+        if len(syllable_list) != syl_count and syl_count > 0:
+            syllable_list = adjust_syllables(clean_word, syl_count)
+            
+        return syllable_list
+    except:
+        return [word]
+
+def simple_syllabify(word: str) -> List[str]:
+    """Simple syllabification based on vowel patterns"""
+    word = word.lower()
+    vowels = "aeiouy"
+    syllables_list = []
+    current_syllable = ""
+    
+    i = 0
+    while i < len(word):
+        current_syllable += word[i]
+        
+        # If current character is a vowel
+        if word[i] in vowels:
+            # Look ahead for consonant clusters
+            consonant_cluster = ""
+            j = i + 1
+            while j < len(word) and word[j] not in vowels:
+                consonant_cluster += word[j]
+                j += 1
+            
+            # If we're at the end of the word, add everything
+            if j >= len(word):
+                current_syllable += consonant_cluster
+                syllables_list.append(current_syllable)
+                break
+            
+            # Split consonant cluster
+            if len(consonant_cluster) <= 1:
+                # Single consonant goes with next syllable
+                syllables_list.append(current_syllable)
+                current_syllable = consonant_cluster
+            else:
+                # Multiple consonants: split them
+                split_point = len(consonant_cluster) // 2
+                current_syllable += consonant_cluster[:split_point]
+                syllables_list.append(current_syllable)
+                current_syllable = consonant_cluster[split_point:]
+            
+            i = j - 1
+        
+        i += 1
+    
+    if current_syllable:
+        if syllables_list:
+            syllables_list[-1] += current_syllable
+        else:
+            syllables_list.append(current_syllable)
+    
+    return syllables_list if syllables_list else [word]
+
+def adjust_syllables(word: str, target_count: int) -> List[str]:
+    """Adjust syllable split to match target count"""
+    simple_split = simple_syllabify(word)
+    
+    if len(simple_split) == target_count:
+        return simple_split
+    
+    # If we have fewer syllables than target, try to split more
+    if len(simple_split) < target_count:
+        # Split longest syllable
+        longest_idx = max(range(len(simple_split)), key=lambda i: len(simple_split[i]))
+        longest = simple_split[longest_idx]
+        
+        if len(longest) > 2:
+            mid = len(longest) // 2
+            simple_split[longest_idx:longest_idx+1] = [longest[:mid], longest[mid:]]
+    
+    # If we have more syllables than target, try to merge
+    elif len(simple_split) > target_count and len(simple_split) > 1:
+        # Merge shortest adjacent syllables
+        min_combined_len = float('inf')
+        merge_idx = 0
+        
+        for i in range(len(simple_split) - 1):
+            combined_len = len(simple_split[i]) + len(simple_split[i + 1])
+            if combined_len < min_combined_len:
+                min_combined_len = combined_len
+                merge_idx = i
+        
+        merged = simple_split[merge_idx] + simple_split[merge_idx + 1]
+        simple_split[merge_idx:merge_idx+2] = [merged]
+    
+    return simple_split
+
+def process_text(text: str) -> Tuple[List[Tuple[str, str]], List[Tuple[str, List[Tuple[str, str]]]]]:
+    """Process text and return colored words and syllable breakdown"""
+    words = re.findall(r'\S+|\s+', text)  # Keep spaces
+    
+    colored_words = []
+    syllable_breakdown = []
+    
+    for word in words:
+        if word.strip():  # If it's a word (not just whitespace)
+            syllables_list = syllabify_word(word.strip())
+            
+            # Generate colors for this word's syllables
+            colors = generate_color_palette(len(syllables_list))
+            
+            # For original text coloring, we need to map syllables back to original word
+            word_colored = []
+            syllable_colored = []
+            
+            # Handle punctuation
+            clean_word = re.sub(r'[^\w]', '', word.strip())
+            prefix_punct = re.match(r'^[^\w]*', word.strip()).group()
+            suffix_punct = re.search(r'[^\w]*$', word.strip()).group()
+            
+            if clean_word:
+                # Color each syllable
+                for i, syl in enumerate(syllables_list):
+                    color = colors[i % len(colors)]
+                    syllable_colored.append((syl, color))
+                
+                # For the original word, we'll color it uniformly with the first syllable color
+                # or create a gradient effect
+                word_with_punct = prefix_punct + clean_word + suffix_punct
+                colored_words.append((word_with_punct, colors[0] if colors else "#000000"))
+                
+                syllable_breakdown.append((word.strip(), syllable_colored))
+            else:
+                colored_words.append((word, "#000000"))
+        else:
+            colored_words.append((word, "#000000"))  # Spaces remain uncolored
+    
+    return colored_words, syllable_breakdown
+
+def display_colored_text(colored_words: List[Tuple[str, str]]):
+    """Display text with colored words"""
+    html_parts = []
+    for word, color in colored_words:
+        if word.strip():
+            html_parts.append(f'<span style="color: {color}; font-weight: bold;">{word}</span>')
+        else:
+            html_parts.append(word)
+    
+    html_content = ''.join(html_parts)
+    st.markdown(f'<div style="font-size: 18px; line-height: 1.6; padding: 15px; border: 2px solid #ddd; border-radius: 10px; background-color: #f9f9f9;">{html_content}</div>', unsafe_allow_html=True)
+
+def display_syllable_breakdown(syllable_breakdown: List[Tuple[str, List[Tuple[str, str]]]]):
+    """Display syllable breakdown"""
+    html_parts = []
+    
+    for word, syllables in syllable_breakdown:
+        if syllables:
+            word_html = []
+            for syl, color in syllables:
+                word_html.append(f'<span style="color: {color}; font-weight: bold; margin-right: 2px;">{syl}</span>')
+            
+            # Add separator between syllables
+            syllable_display = '<span style="color: #666;">|</span>'.join([
+                f'<span style="color: {color}; font-weight: bold;">{syl}</span>' 
+                for syl, color in syllables
+            ])
+            
+            html_parts.append(f'<div style="margin-bottom: 10px;"><strong>{word}:</strong> {syllable_display}</div>')
+    
+    if html_parts:
+        html_content = ''.join(html_parts)
+        st.markdown(f'<div style="font-size: 16px; line-height: 1.8; padding: 15px; border: 2px solid #ddd; border-radius: 10px; background-color: #f0f8ff;">{html_content}</div>', unsafe_allow_html=True)
+
+# Main app
+def main():
+    st.title("🔤 Text Syllabification Tool")
+    st.markdown("Enter text below to see syllable breakdown with color coding!")
+    
+    # Input text area
+    input_text = st.text_area(
+        "Enter your text:",
+        placeholder="Type or paste your text here...",
+        height=150,
+        help="Enter any text to see how words are broken down into syllables"
+    )
+    
+    if input_text.strip():
+        # Process the text
+        colored_words, syllable_breakdown = process_text(input_text)
+        
+        # Create two columns
+        col1, col2 = st.columns(2)
+        
+        with col1:
+            st.subheader("Original Text (Colored by First Syllable)")
+            display_colored_text(colored_words)
+        
+        with col2:
+            st.subheader("Syllable Breakdown")
+            display_syllable_breakdown(syllable_breakdown)
+        
+        # Statistics
+        total_words = len([word for word, _ in syllable_breakdown])
+        total_syllables = sum(len(syls) for _, syls in syllable_breakdown)
+        
+        st.markdown("---")
+        st.markdown(f"**Statistics:** {total_words} words, {total_syllables} syllables")
+        
+        # Show syllable counts
+        if syllable_breakdown:
+            st.subheader("Syllable Counts")
+            counts_html = []
+            for word, syllables in syllable_breakdown:
+                count = len(syllables)
+                counts_html.append(f"<span style='margin-right: 15px;'><strong>{word}</strong>: {count}</span>")
+            
+            counts_display = " ".join(counts_html)
+            st.markdown(f'<div style="padding: 10px; background-color: #f0f0f0; border-radius: 5px;">{counts_display}</div>', unsafe_allow_html=True)
+    
+    else:
+        st.info("👆 Enter some text above to see the syllabification in action!")
+        
+        # Example
+        st.subheader("Example")
+        example_text = "Beautiful mountains and valleys create wonderful landscapes."
+        st.code(example_text)
+        
+        if st.button("Try Example"):
+            st.session_state.example_clicked = True
+            # Process example
+            colored_words, syllable_breakdown = process_text(example_text)
+            
+            col1, col2 = st.columns(2)
+            
+            with col1:
+                st.subheader("Original Text (Colored)")
+                display_colored_text(colored_words)
+            
+            with col2:
+                st.subheader("Syllable Breakdown")
+                display_syllable_breakdown(syllable_breakdown)
+
+if __name__ == "__main__":
+    main()