web_app/app.py

"""
Streamlit web application for Linguistic Data Analysis I tutorials.
Provides lexical sophistication analysis and POS/dependency parsing.

Refactored version with modular architecture for better maintainability.
"""

import sys
import os
from pathlib import Path

# Add parent directory to path for imports
sys.path.append(os.path.dirname(os.path.dirname(__file__)))

# CRITICAL: Initialize GPU BEFORE any SpaCy/model imports
from web_app.gpu_init import GPU_AVAILABLE

import streamlit as st

# Import custom modules
from web_app.session_manager import SessionManager
from web_app.components.ui_components import UIComponents
from web_app.handlers.analysis_handlers import AnalysisHandlers
from web_app.reference_manager import ReferenceManager
from web_app.handlers.pos_handlers import POSHandlers
from web_app.handlers.frequency_handlers import FrequencyHandlers
from web_app.handlers.corpus_viz_handlers import CorpusVizHandlers

# Import logging for GPU verification
import logging
logger = logging.getLogger(__name__)

# Configure Streamlit page
st.set_page_config(
    page_title="Linguistic Data Analysis I - Text Analysis Tools",
    page_icon="📊",
    layout="wide",
    initial_sidebar_state="expanded"
)


def main():
    """Main application entry point."""
    st.title("📊 Linguistic Data Analysis I - Text Analysis Tools")
    st.markdown("*Educational tools for lexical sophistication analysis, POS/dependency parsing, and word frequency visualization*")
    
    # GPU status is already initialized in gpu_init module
    if GPU_AVAILABLE:
        logger.info("GPU initialization successful - models will use GPU")
    else:
        logger.info("GPU not available - models will use CPU")
    
    # Initialize session state
    SessionManager.initialize_session_state()
    
    # Render sidebar and get tool choice
    tool_choice = render_sidebar()
    
    # Handle language changes
    SessionManager.handle_language_change()
    
    # Route to appropriate interface
    if tool_choice == 'Lexical Sophistication':
        render_lexical_sophistication_interface()
    elif tool_choice == 'POS & Dependency Parser':
        render_pos_parser_interface()
    elif tool_choice == 'Corpus Data Visualizer':
        render_corpus_visualization_interface()
    else:  # Frequency Analysis
        render_frequency_analysis_interface()


def render_sidebar():
    """Render sidebar configuration options."""
    with st.sidebar:
        st.title("Configuration")
        
        # Language selection
        UIComponents.render_language_selector()
        
        # Model size selection
        UIComponents.render_model_selector()
        
        # Tool selection
        tool_choice = UIComponents.render_tool_selector()
        
        # Debug mode toggle
        st.write("---")
        debug_mode = st.checkbox("🐛 Debug Mode", key="debug_mode", help="Enable debug information for troubleshooting")
        
        if debug_mode:
            from web_app.debug_utils import show_environment_info, test_file_operations, debug_file_upload, show_gpu_status
            
            with st.expander("Environment Info", expanded=False):
                show_environment_info()
            
            with st.expander("File Operations Test", expanded=False):
                test_file_operations()
            
            with st.expander("File Upload Test", expanded=False):
                debug_file_upload()
            
            with st.expander("GPU Status", expanded=False):
                show_gpu_status()
        
        return tool_choice


def render_lexical_sophistication_interface():
    """Render lexical sophistication analysis interface."""
    st.header("🔍 Emulation of the Tool for Automatic Analysis of Lexical Sophistication (emuTAALES)")
    
    # Get analyzer
    analyzer = AnalysisHandlers.get_analyzer()
    if analyzer is None:
        st.error("Failed to load analyzer. Please check your SpaCy model installation.")
        return
    
    # Analysis mode selection
    analysis_mode = st.radio(
        "Analysis Mode",
        options=['Single Text', 'Two-Text Comparison', 'Batch Analysis'],
        horizontal=True
    )
    
    
    # Route to appropriate analysis handler
    if analysis_mode == 'Single Text':
        AnalysisHandlers.handle_single_text_analysis(analyzer)
    elif analysis_mode == 'Batch Analysis':
        AnalysisHandlers.handle_batch_analysis(analyzer)
    else:
        AnalysisHandlers.handle_two_text_comparison(analyzer)


def render_pos_parser_interface():
    """Simplified POS parser interface without separate rule testing."""
    st.header("🏷️ POS and Dependency Parser")
    
    # Get parser
    parser = AnalysisHandlers.get_pos_parser()
    if parser is None:
        st.error("Failed to load POS parser. Please check your SpaCy model installation.")
        return
    
    # Simplified analysis mode selection (removed Rule Testing)
    analysis_mode = st.radio(
        "Analysis Mode",
        options=['POS Analysis', 'Construction Extraction'],
        horizontal=True,
        key='pos_analysis_mode'
    )
    
    if analysis_mode == 'POS Analysis':
        # Sub-mode for POS analysis
        pos_sub_mode = st.radio(
            "POS Analysis Type",
            options=['Single Text', 'Batch Analysis'],
            horizontal=True,
            key='pos_sub_mode'
        )
        
        if pos_sub_mode == 'Single Text':
            POSHandlers.handle_single_text_pos_analysis(parser)
        else:
            POSHandlers.handle_batch_pos_analysis(parser)
            
    else:  # Construction Extraction (now includes integrated rule testing)
        construction_sub_mode = st.radio(
            "Construction Analysis Type",
            options=['Single Text', 'Batch Analysis'],
            horizontal=True,
            key='construction_sub_mode'
        )
        
        if construction_sub_mode == 'Single Text':
            POSHandlers.handle_construction_analysis(parser)  # Now includes visual parsing
        else:
            POSHandlers.handle_batch_construction_analysis(parser)


def render_frequency_analysis_interface():
    """Render frequency analysis interface."""
    st.header("📊 Word Frequency Analysis")
    st.markdown("Analyze and visualize word frequency distributions from TSV data files.")
    
    # Handle frequency analysis
    FrequencyHandlers.handle_frequency_analysis()


def render_corpus_visualization_interface():
    """Render corpus data visualization interface."""
    st.header("🗂️ Corpus Data Visualizer")
    st.markdown("Merge and visualize corpus metadata with analysis results to create insightful visualizations.")
    
    # Handle corpus visualization
    CorpusVizHandlers.handle_corpus_visualization()


if __name__ == "__main__":
    main()