collaboration.py

import requests
import json
import time

# Istituzioni pugliesi dal tuo filtro (lineage normalizzata)
PUGLIA_SHORT_IDS = [
    "I68618741", "I142910587", "I106296451", "I5561750",
    "I4210108634", "I4210166808", "I4210144761", "I4210099368",
    "I4210148743", "I4210131270", "I4210117053", "I4210122493"
]

BASE_URL = "https://api.openalex.org/works"
# Filtro originale (rimosso grant per velocità e stabilità)
QUERY_FILTER = (
    "title_and_abstract.search:\"blue economy\" OR \"blue growth\" OR \"ocean economy\" OR \"marine economy\" "
    "OR \"economia blu\" OR \"crescita blu\" OR \"crescita del blu\" OR fisheries OR fishing OR aquaculture "
    "OR mariculture OR \"fish farming\" OR seafood OR \"marine biotechnology\" OR microalgae OR macroalgae "
    "OR seaweed OR algae OR \"offshore wind\" OR \"marine renewable energy\" OR \"wave energy\" OR \"tidal energy\" "
    "OR \"blue energy\" OR desalination OR \"seawater desalination\" OR \"marine spatial planning\" OR \"blue carbon\" "
    "OR seagrass OR posidonia OR benthic OR \"marine ecology\" OR \"marine biodiversity\" OR \"maritime transport\" "
    "OR shipping OR \"port infrastructure\" OR shipbuilding OR \"ship repair\" OR \"naval architecture\" OR boatbuilding "
    "OR yacht* OR \"coastal management\" OR \"coastal erosion\" OR shoreline OR \"coastal zone\" OR \"coastal tourism\" "
    "OR \"marine tourism\" OR \"nautical tourism\" OR \"cruise tourism\" OR \"microplastic*\" OR \"marine littering\" "
    "OR \"marine pollution\" OR \"port\","
    "authorships.countries:IT,publication_year:2014-2024,"
    "authorships.institutions.lineage:i68618741|i142910587|i106296451|i5561750|i4210108634|i4210166808|i4210144761|i4210099368|i4210148743|i4210131270|i4210117053|i4210122493,"
    "authorships.institutions.lineage:!i4210155236"
)

def run_analysis():
    cursor = "*"
    all_works = []
    collab_map = {} # { id_puglia: { 'name': ..., 'partners': { id_partner: data } } }

    print("Interrogazione OpenAlex in corso...")
    
    while cursor:
        params = {
            "filter": QUERY_FILTER,
            "select": "id,title,authorships",
            "per_page": 200,
            "cursor": cursor
        }
        print(params)
        response = requests.get(BASE_URL, params=params).json()
        results = response.get("results", [])
        if not results: break
        
        all_works.extend(results)
        
        for work in results:
            authorships = work.get("authorships", [])
            
            # Identifica attori pugliesi nel lavoro
            puglia_presenti = []
            for auth in authorships:
                pos = auth.get("author_position")
                # Definizione Utente: IN = First/Last, OUT = Middle
                tipo_collab = "IN" if pos in ["first", "last"] else "OUT"
                
                for inst in auth.get("institutions", []):
                    # Estrae ID pulito dalla lineage (es. I123)
                    lineage = [l.split("/")[-1].upper() for l in inst.get("lineage", [])]
                    if any(pid in lineage for pid in PUGLIA_SHORT_IDS):
                        puglia_presenti.append({
                            "id": inst.get("id"),
                            "name": inst.get("display_name"),
                            "tipo": tipo_collab
                        })

            # Analisi incrociata con i partner
            for p_actor in puglia_presenti:
                p_id = p_actor["id"]
                if p_id not in collab_map:
                    collab_map[p_id] = {"name": p_actor["name"], "partners": {}}
                
                for auth in authorships:
                    paese = auth.get("countries", ["N/A"])
                    for inst in auth.get("institutions", []):
                        c_id = inst.get("id")
                        if c_id == p_id: continue # Salta se stesso
                        
                        if c_id not in collab_map[p_id]["partners"]:
                            c_lineage = [l.split("/")[-1].upper() for l in inst.get("lineage", [])]
                            is_puglia = any(pid in c_lineage for pid in PUGLIA_SHORT_IDS)
                            
                            collab_map[p_id]["partners"][c_id] = {
                                "nome": inst.get("display_name"),
                                "ambito": "Inter-regionale" if is_puglia else "Extra-regionale",
                                "paese": paese[0] if paese else "N/A",
                                "count_IN": 0, # Leadership Puglia
                                "count_OUT": 0 # Supporto Puglia
                            }
                        
                        collab_map[p_id]["partners"][c_id][f"count_{p_actor['tipo']}"] += 1

        cursor = response.get("meta", {}).get("next_cursor")
        print(f"Scaricati {len(all_works)} lavori...")
        time.sleep(0.1)

    # Salvataggio Risultati
    with open("dataset_completo.json", "w", encoding="utf-8") as f:
        json.dump(all_works, f, indent=4, ensure_ascii=False)
    
    with open("mappa_collaborazioni.json", "w", encoding="utf-8") as f:
        json.dump(collab_map, f, indent=4, ensure_ascii=False)

    # Stampa Analisi Finale
    print(f"\nAnalisi completata su {len(all_works)} articoli.")
    for p_id, data in collab_map.items():
        sum_in = sum(v['count_IN'] for v in data['partners'].values())
        sum_out = sum(v['count_OUT'] for v in data['partners'].values())
        print(f"\nISTITUZIONE: {data['name']}")
        print(f"Totale Collaborazioni IN (Puglia First/Last): {sum_in}")
        print(f"Totale Collaborazioni OUT (Puglia Middle): {sum_out}")
        print(f"{'Partner':<45} | {'Ambito':<18} | {'Tipo':<6} | {'Paese':<6} | {'Tot'}")
        print("-" * 90)
        
        # Ordina per volume di collaborazione
        sorted_partners = sorted(data['partners'].items(), 
                                 key=lambda x: x[1]['count_IN'] + x[1]['count_OUT'], reverse=True)
        
        for c_id, d in sorted_partners[:15]:
            tot = d['count_IN'] + d['count_OUT']
            pref_tipo = "IN" if d['count_IN'] >= d['count_OUT'] else "OUT"
            print(f"{d['nome'][:43]:<45} | {d['ambito']:<18} | {pref_tipo:<6} | {d['paese']:<6} | {tot}")

run_analysis()