Create app.py

app.py

@@ -0,0 +1,152 @@
+import gradio as gr
+import pandas as pd
+import requests
+from datetime import datetime
+import json
+import os
+
+API_URL = "https://api-inference.huggingface.co/models/mistralai/Mistral-Small-24B-Instruct-2501"
+
+EXAMPLE_CSV = """Citation
+Máñez, I., Lipnevich, A.A., Lopera-Oquendo, C. et al. Examining pre-service teachers' feedback on low- and high-quality written assignments. Educ Asse Eval Acc 36, 225–256 (2024). https://doi.org/10.1007/s11092-024-09432-x
+Roda, A., & Menken, K. (2024). The Conflation of Dual Language Bilingual Education With Gifted Programs in New York City Schools. Educational Policy, 0(0). https://doi.org/10.1177/08959048241237724"""
+
+def get_journal_name(citation):
+    headers = {"Content-Type": "application/json"}
+    payload = {
+        "inputs": f"""<s>[INST]What is the journal name in this academic citation? Return only the journal name, nothing else. If the journal name is abbreviated, return the full journal name.
+
+Citation: {citation}[/INST]</s>"""
+    }
+    
+    response = requests.post(API_URL, headers=headers, json=payload)
+    result = response.json()
+    
+    # The model returns a list of responses, we take the first one
+    if isinstance(result, list) and len(result) > 0:
+        return result[0]["generated_text"].strip()
+    return "Journal name not found"
+
+def get_oa_policies(journal):
+    base_url = 'https://v2.sherpa.ac.uk/cgi/retrieve_by_id'
+    api_key = os.environ.get('SHERPA_ROMEO_API_KEY')
+    full_url = f'{base_url}?item-type=publication&api-key={api_key}&format=Json&identifier={journal}'
+    try:
+        response = requests.get(full_url)
+        response.raise_for_status()  # Raise an exception for bad status codes
+        data = response.json()
+
+        policies = data['items'][0]['publisher_policy']
+
+        policies_dict = {
+            'journal': journal,
+            'policies': []
+        }
+
+        for policy in policies:
+            list_of_policies = []
+
+            for article_version_policy in policy.get('permitted_oa', []):
+                article_version = ''.join(article_version_policy['article_version'])
+                oa_fee = article_version_policy.get('additional_oa_fee', 'no')
+
+                embargo = article_version_policy.get('embargo', {})
+                formatted_embargo = f"{embargo.get('amount', 'no')} {embargo.get('units', 'no')}"
+
+                locations = article_version_policy.get('location', {})
+                location_list = ', '.join(locations.get('location', ['none']))
+
+                conditions = article_version_policy.get('conditions', 'none')
+                formatted_conditions = ', '.join(map(str, conditions)) if isinstance(conditions, list) else str(conditions)
+
+                compiled_article_version_policy = {
+                    'article_version': article_version,
+                    'oa_fee': oa_fee,
+                    'embargo': formatted_embargo,
+                    'locations': location_list,
+                    'conditions': formatted_conditions
+                }
+
+                list_of_policies.append(compiled_article_version_policy)
+
+            policies_dict['policies'].append(list_of_policies)
+
+        return policies_dict
+
+    except (IndexError, requests.exceptions.RequestException) as e:
+        print(f"Error retrieving policies for {journal}: {e}")  # Added error logging
+        return {'journal': journal, 'policies': 'No information found or error occurred.'}
+
+# --- Main Processing Function ---
+def process_citations(csv_file):
+    try:
+        df = pd.read_csv(csv_file.name)
+        if 'Citation' not in df.columns:
+            return "Error: CSV file must have a 'Citation' column."
+
+        df['Journal'] = df['Citation'].apply(get_journal_name)
+
+        output_message = f"""
+Queens College Library Research Services
+= = = = = = = = = = = = = = = = = = = = =
+({datetime.now().strftime('%Y-%m-%d %H:%M:%S')})
+
+"""
+        for i, row in df.iterrows():
+            citation = row['Citation']
+            journal = row['Journal']
+            oa_policies_data = get_oa_policies(journal)
+
+            oa_policies = oa_policies_data['policies']
+
+            paragraphs = []
+            if oa_policies == 'No information found or error occurred.':
+                paragraphs.append(oa_policies)
+            else:
+                for policy_list in oa_policies:
+                  for policy in policy_list:
+                    paragraphs.append(f"""
+Version: {policy['article_version'].title()} Manuscript
+OA Fee: {policy['oa_fee'].title()}
+Embargo: {policy['embargo'].title()}
+Locations: {policy['locations']}
+Conditions: {policy['conditions']}
+""")
+
+            per_journal_oa_policies = f"""
++ + + + + + + + + + Citation {i+1} + + + + + + + + + +
+
+{citation}
+
+Journal: {journal}
+"""
+            per_journal_oa_policies += '\n' + '\n'.join(paragraphs) if len(paragraphs) > 0 else '\nNo policies found.\n'
+
+            output_message += per_journal_oa_policies
+
+        return output_message
+
+    except Exception as e:
+        return f"An error occurred: {e}"
+
+# --- Gradio Interface ---
+with gr.Blocks() as interface:
+    gr.Markdown("# Open Access Policy Checker")
+    gr.Markdown("Upload a CSV file with a 'Citation' column to get Open Access policies.")
+    
+    with gr.Accordion("Example CSV Format", open=False):
+        gr.Code(
+            EXAMPLE_CSV,
+            language="csv",
+            label="Example CSV content"
+        )
+    
+    with gr.Row():
+        input_file = gr.File(label="Upload CSV")
+        output_text = gr.Textbox(label="OA Policies Report", show_copy_button=True)
+    process_button = gr.Button("Get OA Policies")
+    process_button.click(process_citations, inputs=input_file, outputs=output_text)
+
+
+# Launch the interface
+interface.launch()