Spaces:
Running
Running
| import gradio as gr | |
| import json | |
| from huggingface_hub import hf_hub_download | |
| import spaces | |
| import regex as re | |
| FOLDER = "Selma" | |
| master_idx = {} | |
| print("Loading index...") | |
| try: | |
| file_path = hf_hub_download( | |
| repo_id="pnugues/selma_idx", | |
| filename="master.json", | |
| repo_type="dataset" | |
| ) | |
| with open(file_path, "r", encoding="utf-8") as f: | |
| master_idx = json.load(f) | |
| print("Loading complete!") | |
| except Exception as e: | |
| print(f"Error loading index: {e}") | |
| def concordance(word, master_index, window): | |
| # Check if word exists in index | |
| if word not in master_index: | |
| print(f"The word '{word}' was not found.") | |
| return | |
| # Iterate files and positions | |
| for file_name, positions in master_index[word].items(): | |
| print(file_name) | |
| # Open file and read | |
| with open('Selma/' + file_name, 'r', encoding='utf-8') as f: | |
| text = f.read().lower() | |
| # extract the surrounding text | |
| for pos in positions: | |
| # Calculate start and end indices, no negative indices | |
| start_idx = max(0, pos - window) | |
| end_idx = min(len(text), pos + len(word) + window) | |
| # Extract snippet, replace newlines with spaces | |
| snippet = text[start_idx:end_idx].replace('\n', ' ') | |
| print(f"\t{snippet}") | |
| with gr.Blocks(title="Selma explorer") as demo: | |
| gr.Markdown("# Selma explorer") | |
| with gr.Row(): | |
| with gr.Column(): | |
| in_text = gr.Textbox( | |
| label="Input text", | |
| placeholder="Write your word..." | |
| ) | |
| with gr.Column(): | |
| out_novel = gr.Textbox( | |
| label="Selma's novels", | |
| placeholder="Selma's concordances will show here...", | |
| lines=20 | |
| ) | |
| btn = gr.Button("Show me the concordances!") | |
| btn.click(fn=concordance, inputs=in_text, outputs=out_novel) | |
| if __name__ == "__main__": | |
| demo.launch(ssr_mode=False) | |