Create app.py

app.py

@@ -0,0 +1,100 @@
+import gradio as gr
+import requests
+import xml.etree.ElementTree as ET
+import pandas as pd
+import csv
+from io import StringIO
+#from google.colab import userdata
+#userdata.get('secretName')
+
+# Correctly defined functions for fetching articles and converting them to CSV format
+def fetch_articles(keyword, max_results=10):
+    search_url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi"
+    search_params = {"db": "pubmed", "term": keyword, "retmax": max_results, "usehistory": "y"}
+    search_response = requests.get(search_url, params=search_params)
+    search_tree = ET.fromstring(search_response.content)
+    id_list = [id_.text for id_ in search_tree.findall('.//Id')]
+
+    fetch_url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi"
+    fetch_params = {"db": "pubmed", "id": ",".join(id_list), "retmode": "xml"}
+    fetch_response = requests.get(fetch_url, params=fetch_params)
+    fetch_tree = ET.fromstring(fetch_response.content)
+
+    articles = []
+    for article in fetch_tree.findall('.//PubmedArticle'):
+        title = article.find('.//ArticleTitle').text
+        authors_list = [author.find('.//LastName').text + " " + author.find('.//ForeName').text for author in article.findall('.//Author') if author.find('.//LastName') is not None and author.find('.//ForeName') is not None]
+        authors_formatted = ", ".join(authors_list) if len(authors_list) <= 2 else authors_list[0] + " et al."
+
+        pub_date_element = article.find('.//PubDate')
+        year = "Unknown"
+        if pub_date_element is not None:
+            year_element = pub_date_element.find('Year')
+            if year_element is not None:
+                year = year_element.text
+
+        journal = article.find('.//Journal/Title').text if article.find('.//Journal/Title') is not None else 'No Journal'
+        abstract = article.find('.//Abstract/AbstractText').text if article.find('.//Abstract/AbstractText') is not None else 'No Abstract'
+       # mesh_terms = [mesh.text for mesh in article.findall('.//MeshHeading/DescriptorName')]
+        article_doi = article.find(".//ArticleId[@IdType='doi']")
+        doi = article_doi.text if article_doi is not None else "No DOI available"
+        doi_link = f"https://doi.org/{doi}" if doi != "No DOI available" else ""
+
+        articles.append({
+            "Title": title,
+            "Authors": authors_formatted,
+            "Year": year,
+            "DOI": doi_link,
+            "Abstract": abstract,
+            "Journal": journal,
+           # "MeSH Terms": "; ".join(mesh_terms)
+        })
+
+    return articles
+
+def write_articles_to_csv(articles, filename="articles.csv"):
+    with open(filename, mode='w', newline='', encoding='utf-8') as file:
+        writer = csv.writer(file)
+        writer.writerow(['Authors', 'Year', 'Title', 'DOI', 'Abstract', 'Journal'])
+        for article in articles:
+            writer.writerow([article['Authors'], article['Year'], article['Title'], article['DOI'], article['Abstract'], article['Journal']])
+    print(f"Articles written to {filename}: {len(articles)}")
+
+def articles_to_csv_string(articles):
+    output = StringIO()
+    writer = csv.writer(output)
+    writer.writerow(['Authors', 'Year', 'Title', 'DOI', 'Abstract', 'Journal'])
+    for article in articles:
+        writer.writerow([article['Authors'], article['Year'], article['Title'], article['DOI'], article['Abstract'], article['Journal']])
+    output.seek(0)
+    return output.getvalue()
+
+def process_inputs(first_name, help_message, keyword1, keyword2, keyword3):
+    if len(help_message) > 50:
+        return pd.DataFrame([{"Error": "Message exceeds 50 characters. Please shorten your message."}])
+    
+    # Combine keywords with "AND" for a broader search, not strictly as MeSH terms
+    keywords = f"{keyword1} AND {keyword2} AND {keyword3}"
+    articles = fetch_articles(keywords, max_results=10)
+    if articles:
+        csv_string = articles_to_csv_string(articles)  # Convert articles to CSV string
+        df = pd.read_csv(StringIO(csv_string))  # Convert CSV string to DataFrame
+        return df
+    else:
+        return pd.DataFrame([{"Error": "No articles found for the given keywords."}])
+iface = gr.Interface(
+    fn=process_inputs,
+    inputs=[
+       # gr.Textbox(label="First Name", placeholder="Enter your first name here..."),
+       # gr.Textbox(label="How can I help you today?", placeholder="Your answer (under 50 characters)"),
+        gr.Textbox(label="Word 1", placeholder="First emotional state or concern..."),
+        gr.Textbox(label="Word 2", placeholder="Second emotional state or concern..."),
+        gr.Textbox(label="Word 3", placeholder="Third emotional state or concern...")
+    ],
+    outputs=gr.Dataframe(label="Related Research Articles"),
+    title="Mental Health Research Assistant",
+    description="This tool helps you find research articles related to your mental health concerns. Enter your emotional states or concerns as keywords."
+)
+
+iface.launch(debug=True)
+