import requests import streamlit as st from bs4 import BeautifulSoup import pandas as pd from transformers import pipeline import plotly.express as px import time import io import os import zipfile import re import json from cryptography.fernet import Fernet from streamlit_extras.stylable_container import stylable_container from comet_ml import Experiment st.set_page_config( layout="wide", page_title="English Keyphrase TXT & URL Entity Finder" ) # --- Configuration for Comet ML --- COMET_API_KEY = os.environ.get("COMET_API_KEY") COMET_WORKSPACE = os.environ.get("COMET_WORKSPACE") COMET_PROJECT_NAME = os.environ.get("COMET_PROJECT_NAME") comet_initialized = False if COMET_API_KEY and COMET_WORKSPACE and COMET_PROJECT_NAME: comet_initialized = True # --- Persistent Counter & History Configuration --- PERSISTENCE_FILE = "app_data.json" max_attempts = 10 def load_persistent_data(): """ Loads the attempts count and file upload history from a persistent JSON file. Returns default values if the file doesn't exist or is invalid. """ if os.path.exists(PERSISTENCE_FILE): try: with open(PERSISTENCE_FILE, "r") as f: data = json.load(f) return data.get('source_type_attempts', 0), data.get('file_upload_history', []) except (json.JSONDecodeError, KeyError): st.warning("Warning: Could not read persistent data file. Starting with a fresh state.") return 0, [] return 0, [] def save_persistent_data(attempts, history): """ Saves the current attempts count and file upload history to the persistent JSON file. """ with open(PERSISTENCE_FILE, "w") as f: json.dump({'source_type_attempts': attempts, 'file_upload_history': history}, f, indent=4) def clear_input_history_and_rerun(): """Callback function for the "Clear Input History" button.""" st.session_state['file_upload_history'] = [] save_persistent_data(st.session_state['source_type_attempts'], []) st.experimental_rerun() # --- Initialize session state for attempts and encrypted text --- if 'source_type_attempts' not in st.session_state: attempts, history = load_persistent_data() st.session_state['source_type_attempts'] = attempts st.session_state['file_upload_history'] = history if 'encrypted_text_to_process' not in st.session_state: st.session_state['encrypted_text_to_process'] = None if 'uploaded_file_content' not in st.session_state: st.session_state['uploaded_file_content'] = None if 'file_uploader_key' not in st.session_state: st.session_state['file_uploader_key'] = 0 # --- Fernet Encryption Setup --- @st.cache_resource def load_encryption_key(): try: key_str = os.environ.get("FERNET_KEY") if not key_str: raise ValueError("FERNET_KEY environment variable not set. Cannot perform encryption/decryption.") key_bytes = key_str.encode('utf-8') return Fernet(key_bytes) except ValueError as ve: st.error(f"Configuration Error: {ve}. Please ensure the 'FERNET_KEY' environment variable is set securely.") st.stop() except Exception as e: st.error(f"An unexpected error occurred while loading encryption key: {e}. Please check your key format and environment settings.") st.stop() # Initialize the Fernet cipher instance globally (cached) fernet = load_encryption_key() def encrypt_text(text_content: str) -> bytes: """Encrypts a string using the loaded Fernet cipher.""" return fernet.encrypt(text_content.encode('utf-8')) def decrypt_text(encrypted_bytes: bytes) -> str | None: """ Decrypts bytes using the loaded Fernet cipher. Returns the decrypted string, or None if decryption fails. """ try: return fernet.decrypt(encrypted_bytes).decode('utf-8') except Exception as e: st.error(f"Decryption failed. This might indicate data tampering or an incorrect encryption key. Error: {e}") return None # --- UI Header and Notes --- st.subheader("English Keyphrase TXT & URL Entity Finder", divider="rainbow") st.link_button("by nlpblogs", "https://nlpblogs.com", type="tertiary") expander = st.expander("**Important notes on the English Keyphrase TXT & URL Entity Finder**") expander.write(''' **Named Entities:** This English Keyphrase TXT & URL Entity Finder extracts keyphrases from English academic and scientific papers. Results are presented in an easy-to-read table, visualized in an interactive bar chart and tree map, and are available for download along with a Glossary of tags. **How to Use:** 1. Paste a URL and press Enter. 2. Alternatively, type or paste text directly into the text area and press Ctrl + Enter. 3. Or, upload your TXT file. **Usage Limits:** You can request results up to 10 times. **Customization:** To change the app's background color to white or black, click the three-dot menu on the right-hand side of your app, go to Settings and then Choose app theme, colors and fonts. **Technical issues:** If your connection times out, please refresh the page or reopen the app's URL. For any errors or inquiries, please contact us at info@nlpblogs.com ''') # --- Sidebar Content --- with st.sidebar: container = st.container(border=True) container.write("**Named Entity Recognition (NER)** is the task of extracting and tagging entities in text data. Entities can be persons, organizations, locations, countries, products, events etc.") st.subheader("Persistent Data", divider="rainbow") st.info(f"Requests remaining today: **{max_attempts - st.session_state['source_type_attempts']}**") if st.session_state['file_upload_history']: st.subheader("File & URL History", divider="rainbow") history_df = pd.DataFrame(st.session_state['file_upload_history']) st.dataframe(history_df, use_container_width=True, hide_index=True) st.button("Clear Input History", on_click=clear_input_history_and_rerun, type="secondary") st.subheader("Related NER Web Apps", divider="rainbow") st.link_button("Scandinavian JSON Entity Finder", "https://nlpblogs.com/shop/named-entity-recognition-ner/scandinavian-json-entity-finder/", type="primary") # --- Input Fields --- def clear_inputs(): st.session_state.url = "" st.session_state.my_text_area = "" st.session_state['uploaded_file_content'] = None st.session_state['encrypted_text_to_process'] = None st.session_state['file_uploader_key'] += 1 st.experimental_rerun() url = st.text_input("Enter URL from the internet, and then press Enter:", key="url") text = st.text_area("Type or paste your text below, and then press Ctrl + Enter", key='my_text_area') uploaded_file = st.file_uploader("Or upload a .txt file", type=["txt"], key=f"file_uploader_{st.session_state['file_uploader_key']}") st.button("Clear All Inputs", on_click=clear_inputs) source_type = None current_run_text = None if uploaded_file is not None and st.session_state.get('uploaded_file_content') is None: source_type = 'file' try: string_data = io.StringIO(uploaded_file.getvalue().decode("utf-8")).read() current_run_text = string_data st.session_state['uploaded_file_content'] = current_run_text st.session_state['file_upload_history'].append({ 'source_type': 'file', 'filename': uploaded_file.name, 'timestamp': time.strftime('%Y-%m-%d %H:%M:%S') }) save_persistent_data(st.session_state['source_type_attempts'], st.session_state['file_upload_history']) st.success("TXT file uploaded successfully. File content encrypted and secured. Due to security protocols, the file content is hidden.") st.divider() st.write("**Input text content (from uploaded file)**") st.write(current_run_text[:500] + "..." if len(current_run_text) > 500 else current_run_text) except Exception as e: st.error(f"Error processing uploaded file: {e}") current_run_text = None elif url: source_type = 'url' if not url.startswith(("http://", "https://")): st.error("Please enter a valid URL starting with 'http://' or 'https://'.") current_run_text = None else: try: with st.spinner(f"Fetching and parsing content from **{url}**...", show_time=True): f = requests.get(url, timeout=10) f.raise_for_status() soup = BeautifulSoup(f.text, 'html.parser') current_run_text = soup.get_text(separator=' ', strip=True) st.session_state['file_upload_history'].append({ 'source_type': 'url', 'filename': url, 'timestamp': time.strftime('%Y-%m-%d %H:%M:%S') }) save_persistent_data(st.session_state['source_type_attempts'], st.session_state['file_upload_history']) st.divider() st.write("**Input text content (from URL)**") st.write(current_run_text[:500] + "..." if len(current_run_text) > 500 else current_run_text) except Exception as e: st.error(f"Error fetching or parsing URL: {e}") current_run_text = None elif text: source_type = 'text' current_run_text = text st.divider() st.write("**Input text content (from text area)**") st.write(current_run_text[:500] + "..." if len(current_run_text) > 500 else current_run_text) if current_run_text and current_run_text.strip(): if st.session_state.get('encrypted_text_to_process') is None: st.session_state['encrypted_text_to_process'] = encrypt_text(current_run_text) else: st.session_state['encrypted_text_to_process'] = None if uploaded_file is None: st.session_state['uploaded_file_content'] = None st.session_state['file_uploader_key'] += 1 # --- Main Processing Logic (corrected placement) --- # The button must be outside the conditional logic that populates the session state # so that it is always rendered and can be clicked to trigger the analysis. if st.button("Analyze Text", type="primary"): if st.session_state['encrypted_text_to_process']: try: start_time_overall = time.time() if st.session_state['source_type_attempts'] >= max_attempts: st.error(f"You have requested results {max_attempts} times. You have reached your request limit.") st.stop() st.session_state['source_type_attempts'] += 1 save_persistent_data(st.session_state['source_type_attempts'], st.session_state['file_upload_history']) @st.cache_resource def load_ner_model(): return pipeline("token-classification", model="ml6team/keyphrase-extraction-kbir-inspec", aggregation_strategy="max", stride=128, ignore_labels=["O"]) model = load_ner_model() text_for_ner = decrypt_text(st.session_state['encrypted_text_to_process']) if text_for_ner and len(text_for_ner.strip()) > 0: with st.spinner("Analyzing text...", show_time=True): entities = model(text_for_ner) data = [] if entities: for entity in entities: if all(k in entity for k in ['word', 'entity_group', 'score', 'start', 'end']): data.append({ 'word': entity['word'], 'entity_group': entity['entity_group'], 'score': entity['score'], 'start': entity['start'], 'end': entity['end'] }) else: st.warning(f"Skipping malformed entity encountered: {entity}. Missing expected keys.") df = pd.DataFrame(data) else: df = pd.DataFrame(columns=['word', 'entity_group', 'score', 'start', 'end']) if not df.empty: pattern = r'[^\w\s]' df['word'] = df['word'].replace(pattern, '', regex=True) df = df.replace('', 'Unknown') st.subheader("All Extracted Keyphrases", divider="rainbow") st.dataframe(df, use_container_width=True) with st.expander("See Glossary of tags"): st.write(''' **word**: ['entity extracted from your text data'] **score**: ['accuracy score; how accurately a tag has been assigned to a given entity'] **entity_group**: ['label (tag) assigned to a given extracted entity'] **start**: ['index of the start of the corresponding entity'] **end**: ['index of the end of the corresponding entity'] ''') st.divider() st.subheader("Most Frequent Keyphrases", divider="rainbow") word_counts = df['word'].value_counts().reset_index() word_counts.columns = ['word', 'count'] df_frequent = word_counts.sort_values(by='count', ascending=False).head(15) if not df_frequent.empty: tab1, tab2 = st.tabs(["Table", "Chart"]) with tab1: st.dataframe(df_frequent, use_container_width=True) with tab2: fig_frequent_bar = px.bar( df_frequent, x='count', y='word', orientation='h', title='Top Frequent Keyphrases by Count', color='count', color_continuous_scale=px.colors.sequential.Viridis ) fig_frequent_bar.update_layout(yaxis={'categoryorder':'total ascending'}) st.plotly_chart(fig_frequent_bar, use_container_width=True) if comet_initialized and 'experiment' in locals(): experiment.log_figure(figure=fig_frequent_bar, figure_name="frequent_keyphrases_bar_chart") else: st.info("No keyphrases found with more than one occurrence to display in tabs.") st.divider() experiment = None if comet_initialized: experiment = Experiment( api_key=COMET_API_KEY, workspace=COMET_WORKSPACE, project_name=COMET_PROJECT_NAME, ) experiment.log_parameter("input_source_type", source_type) experiment.log_parameter("input_content_length", len(text_for_ner)) experiment.log_table("predicted_entities", df) st.subheader("Treemap of All Keyphrases", divider="rainbow") fig_treemap = px.treemap( df, path=[px.Constant("all"), 'entity_group', 'word'], values='score', color='word', color_continuous_scale=px.colors.sequential.Plasma ) fig_treemap.update_layout(margin=dict(t=50, l=25, r=25, b=25)) st.plotly_chart(fig_treemap, use_container_width=True) if comet_initialized and experiment: experiment.log_figure(figure=fig_treemap, figure_name="entity_treemap") # --- Download Section --- dfa = pd.DataFrame( data={ 'Column Name': ['word', 'entity_group', 'score', 'start', 'end'], 'Description': [ 'entity extracted from your text data', 'label (tag) assigned to a given extracted entity', 'accuracy score; how accurately a tag has been assigned to a given entity', 'index of the start of the corresponding entity', 'index of the end of the corresponding entity' ] } ) buf = io.BytesIO() with zipfile.ZipFile(buf, "w") as myzip: if not df.empty: myzip.writestr("Summary_of_results.csv", df.to_csv(index=False)) myzip.writestr("Most_frequent_keyphrases.csv", df_frequent.to_csv(index=False)) myzip.writestr("Glossary_of_tags.csv", dfa.to_csv(index=False)) with stylable_container( key="download_button", css_styles="""button { background-color: yellow; border: 1px solid black; padding: 5px; color: black; }""", ): st.download_button( label="Download zip file", data=buf.getvalue(), file_name="nlpblogs_ner_results.zip", mime="application/zip", ) st.divider() else: st.warning("No entities found to generate visualizations.") else: st.warning("No meaningful text found to process. Please enter a URL, upload a text file, or type/paste text.") except Exception as e: st.error(f"An unexpected error occurred during processing: {e}") finally: if comet_initialized and experiment is not None: try: experiment.end() except Exception as comet_e: st.warning(f"Comet ML experiment.end() failed: {comet_e}") if start_time_overall is not None: end_time_overall = time.time() elapsed_time_overall = end_time_overall - start_time_overall st.info(f"Results processed in **{elapsed_time_overall:.2f} seconds**.") st.write(f"Number of times you requested results: **{st.session_state['source_type_attempts']}/{max_attempts}**") else: st.warning("Please enter some text, a URL, or upload a file to analyze.")