Refactor: Implement Arara OARMP architecture and dynamic editors

2026-06-21 18:15:03 -03:00
parent f730c4fb6e
commit 21664a7aeb
16 changed files with 365457 additions and 3570 deletions
--- a/src/fleet_assignment/ingest.py
+++ b/src/fleet_assignment/ingest.py
@@ -0,0 +1,77 @@
+import json
+import pandas as pd
+from vincenty import vincenty
+import sys
+import os
+
+# Ensure src module is visible if run standalone
+sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
+
+from src.fleet_assignment.config import (
+    CSV_FILEPATH, JSON_AEROPORTOS, JSON_EMISSORES, AIRCRAFT_CONFIG
+)
+
+def load_data():
+    """Reads raw CSV and JSON to build the baseline demand and fleet dataframes."""
+    df = pd.read_csv(CSV_FILEPATH, low_memory=False)
+    df_airports = pd.read_json(JSON_AEROPORTOS, orient='index')
+    
+    # Basic Cleaning
+    df.replace('NIL', 0, inplace=True)
+    df = df[(df['Localidade Decolagem'] != 0) & (df['Localidade Pouso'] != 0)]
+    df = df[df['Modelo'].isin(AIRCRAFT_CONFIG.keys())]
+    
+    df['PAX'] = pd.to_numeric(df['PAX']).fillna(0)
+    df['Dep TimeStamp'] = pd.to_datetime(df['Data de Decolagem'].astype(str) + ' ' + df['Hora de Decolagem'].astype(str), format='mixed', dayfirst=True, errors='coerce')
+    df['Data Apenas'] = df['Dep TimeStamp'].dt.date
+    df['Combustível'] = pd.to_numeric(df['Combustível'], errors='coerce').fillna(0)
+    
+    # Merge Coordinates
+    df = df.merge(df_airports[['lat', 'lon']], left_on='Localidade Decolagem', right_index=True, how='left')
+    df.rename(columns={'lat': 'lat_dep', 'lon': 'lon_dep'}, inplace=True)
+    df = df.merge(df_airports[['lat', 'lon']], left_on='Localidade Pouso', right_index=True, how='left')
+    df.rename(columns={'lat': 'lat_arr', 'lon': 'lon_arr'}, inplace=True)
+    
+    # Distance & Time Computations
+    def calc_dist(row):
+        if pd.isna(row['lat_dep']) or pd.isna(row['lat_arr']): return 0.0
+        return vincenty((row['lat_dep'], row['lon_dep']), (row['lat_arr'], row['lon_arr']))
+        
+    df['Distancia'] = df.apply(calc_dist, axis=1)
+    df['Tempo de Voo'] = pd.to_timedelta(df['Tempo de Voo'].astype(str), errors='coerce')
+    df['Tempo de Voo (h)'] = df['Tempo de Voo'].dt.total_seconds() / 3600.0
+
+    # Build Fleet Specifications
+    df_fleet = df.groupby(['Modelo', 'Emissor']).agg({
+        'Combustível': 'sum',
+        'Distancia': 'sum',
+        'Tempo de Voo (h)': 'sum',
+        'Matrícula': 'nunique'
+    })
+    df_fleet.rename(columns={'Matrícula': 'Tamanho da Frota'}, inplace=True)
+    
+    df_fleet['Consumo (km/l)'] = (df_fleet['Distancia'] / df_fleet['Combustível']).replace([float('inf'), -float('inf')], 0).fillna(0)
+    df_fleet['Velocidade Média (km/h)'] = (df_fleet['Distancia'] / df_fleet['Tempo de Voo (h)']).replace([float('inf'), -float('inf')], 0).fillna(0)
+    
+    models = df_fleet.index.get_level_values('Modelo')
+    df_fleet['Capacidade'] = models.map(lambda m: AIRCRAFT_CONFIG[m]['cap'])
+    df_fleet['Alcance'] = models.map(lambda m: AIRCRAFT_CONFIG[m]['range'])
+    df_fleet['Diaria_Litros'] = models.map(lambda m: AIRCRAFT_CONFIG[m]['daily_fuel'])
+
+    # Build Demands
+    df_valid = df.dropna(subset=['Data Apenas']).copy()
+    df_valid = df_valid[(df_valid['Localidade Decolagem'] != '0') & (df_valid['Localidade Pouso'] != '0')]
+    demands_grouped = df_valid.groupby(['Data Apenas', 'Localidade Decolagem', 'Localidade Pouso'])['PAX'].sum().reset_index()
+    demands = demands_grouped[demands_grouped['PAX'] > 0].copy()
+
+    # Load External Lookups
+    with open(JSON_EMISSORES, 'r') as f:
+        emissores_data = json.load(f)[0]
+    
+    with open(JSON_AEROPORTOS, 'r') as f:
+        airports_geo = json.load(f)
+        
+    # Validation Lists
+    valid_icaos = df_airports.index.intersection(list(set(demands['Localidade Decolagem'].unique().tolist() + demands['Localidade Pouso'].unique().tolist() + list(emissores_data.values())))).tolist()
+    
+    return demands, df_fleet.reset_index(), valid_icaos, airports_geo