Modify table type (date and time) + plots

history.py

 import matplotlib.pyplot as plt
 from cassandra.cluster import Cluster
 from datetime import datetime
+import numpy as np
+
+import loading
+
+table_name_space = loading.table_name_space
+table_name_date = loading.table_name_date
+
+MIN_DATE = 2001
+MAX_DATE = 2010
 
-table_name_station = "station"
 numeric_columns = ["lon","lat","tmpf","dwpf","relh","drct","sknt","p01i","alti","mslp","vsby","gust","skyl1","skyl2","skyl3","skyl4","feel","ice_accretion_1hr","ice_accretion_3hr","ice_accretion_6hr","peak_wind_gust","peak_wind_drct","peak_wind_time"]
 
 cluster = Cluster()
@@ -10,39 +18,103 @@ session = cluster.connect()
 session.set_keyspace("bazinsim_roisinos_metar")
 
 def getHistory(station, indicator):
-	datas = session.execute(f"SELECT year, month, day, hour, minute, {indicator} FROM {table_name} where station = '{station}'")
-	# for t in datas:
-	# 	print(t[0])
+	datas = session.execute(f"SELECT datee, {indicator} FROM {table_name_space} where station = '{station}'")
 	return datas
 
-def plotHistory(station, indicator):
+def getMeanByDay(table, dateMin, dateMax):
+	table_date = {}
+	for r in table:
+		year = int(str(r[0])[0:4])
+		if year >= dateMin and year < dateMax and  r[len(r) - 1] != None:
+			date = str(r[0])
+			if date not in table_date.keys():
+				table_date[date] = 0,0
+			table_date[date] = (table_date[date][0] + r[len(r) - 1], table_date[date][1] + 1)
+
+	for d in table_date.keys():
+		table_date[d] = table_date[d][0] / table_date[d][1]
+	return table_date
+
+def getMeanByMonth(table):
+	table_month = {}
+	for r in table:
+		if r[len(r) - 1] != None:
+			month = str(r[0])[5:7]
+			if month not in table_month.keys():
+				table_month[month] = 0,0
+			table_month[month] = (table_month[month][0] + r[len(r) - 1], table_month[month][1] + 1)
+
+	for d in table_month.keys():
+		table_month[d] = table_month[d][0] / table_month[d][1]
+
+	return table_month
+
+def plotHistory(station, indicator, dateMin, dateMax):
 	if indicator in numeric_columns:
+		if dateMin == dateMax:
+			print(f"Les dates ne doivent pas être égales")
+			return
+
+		if dateMin < MIN_DATE or dateMin > (MAX_DATE + 1) or dateMax < MIN_DATE or dateMax > (MAX_DATE + 1):
+			print(f"Les dates doivent être comprises entre {MIN_DATE} et {MAX_DATE}")
+			return
+
 		table = getHistory(station, indicator)
+		table = list(table)
 
 		if not table:
 			print(f"Aucune donnée pour la station {station} et pour l'indicateur {indicator}")
 			return
 
-		table_date = {}
-		for r in table:
-			if (r[len(r) - 1] != None):
-				date = datetime.strptime(str(r[0]) + "-" + str(r[1]) + "-" + str(r[2]), '%Y-%m-%d').date()
-				if date not in table_date.keys():
-					table_date[date] = 0,0
-				table_date[date] = (table_date[date][0] + r[len(r) - 1], table_date[date][1] + 1)
+		table_mean = getMeanByDay(table, dateMin, dateMax)
 
-		for d in table_date.keys():
-			table_date[d] = table_date[d][0] / table_date[d][1]
+		table_mean_by_month = getMeanByMonth(table)
 
+		# Duplicate list for each year
+		liste = []
+		for i in range(dateMax - dateMin):
+			for key,value in table_mean_by_month.items():
+				liste.append([key, value])
+
+		i = dateMin
+		j = 1
+		
+		for k in range(len(liste)):
+			j += 1
+			liste[k][0] = str(i) + '-' + str(liste[k][0])
+			if j > 12:
+				i += 1
+				j = 1
+
+		# Name for file
 		currentDateTime = datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
 		file_name = str(currentDateTime) + "_" + station + "_" + indicator + ".png"
 
-		plt.plot_date(table_date.keys(), table_date.values(), '-', xdate = True)
+		step_graduation = max((dateMax - dateMin) // 5, 1)
+		print(step_graduation)
+		graduation = ["20" + "0" * (2 - len(str(i))) + str(i) + "-01-01" for i in range(int(str(dateMin)[2:4]), int(str(dateMax)[2:4]), step_graduation)]
+		graduation.append("20" + "0" * (2 - len(str(dateMax - 1)[2:4])) + str(dateMax - 1)[2:4] + "-12-31")
+		print (graduation)
+
+		fig, ax1 = plt.subplots()
+		ax1.plot_date(table_mean.keys(), table_mean.values(), '-', xdate = True)
+		ax1.xaxis.set_ticks(graduation)
+		ax2 = ax1.twiny()
+		ax2.plot([elt[0] for elt in liste], [elt[1] for elt in liste], '-', color = "r")
+		ax2.xaxis.set_ticks([])
+
 		plt.title(f"Evolution de {indicator} pour la station {station}")
 		plt.xlabel('Date')
 		plt.ylabel(indicator)
+		plt.tick_params(
+		    axis='x',
+		    which='both',
+		    bottom=False,
+		    top=True
+		)
 		plt.savefig(file_name)
+		print(f"Le graphique a été enregistré à {file_name}")
 	else:
 		print("Les données pour cet indicateur ne sont pas numériques, impossible de tracer un graphique")
 
-plotHistory("EFKI", "tmpf")
\ No newline at end of file
+plotHistory("EFKI", "tmpf", 2001, 2005)
\ No newline at end of file

loading.py

@@ -52,11 +52,8 @@ def loadata(filename):
 			data["station"] = r["station"]
 			valid = dateparser.match(r["valid"]).groupdict()
 
-			data["year"] = int(valid["year"])
-			data["month"] = int(valid["month"])
-			data["day"] = int(valid["day"])
-			data["hour"] = int(valid["hour"])
-			data["minute"] = int(valid["minute"])
+			data["date"] = valid["year"] + "-" + valid["month"] + "-" + valid["day"]
+			data["time"] = valid["hour"] + ":" + valid["minute"] + ":00"
 
 			data["lon"] = float(r["lon"])
 			data["lat"] = float(r["lat"])
@@ -179,11 +176,8 @@ def loadata(filename):
 
 NAME_COLUMNS = """
 	station 		  ,
-	year			  ,
-	month			  ,
-	day				  ,
-	hour			  ,
-	minute			  ,
+	datee 			  ,
+	timee 			  ,
 	lon 			  ,
 	lat 			  ,
 	tmpf 			  ,
@@ -224,11 +218,8 @@ def dropTableQuery(table):
 def createTableQuery(table):
 	query = f"""CREATE TABLE {table}(
 			station 			varchar,
-			year			  	int,
-			month			  	int,
-			day				  	int,
-			hour			  	int,
-			minute			  	int,
+			datee 				date,
+			timee				time,
 			lon 				decimal,
 			lat 				decimal,
 			tmpf 				decimal,
@@ -258,18 +249,15 @@ def createTableQuery(table):
 			peak_wind_drct    	decimal,
 			peak_wind_time    	decimal,
 			metar             	varchar,
-			PRIMARY KEY((station), year, month, day, hour, minute)
+			PRIMARY KEY((station), datee, timee)
 		)"""
 	return query
 
 def createTableQueryPartitionningByDate(table):
 	query = f"""CREATE TABLE {table}(
 			station 			varchar,
-			year			  	int,
-			month			  	int,
-			day				  	int,
-			hour			  	int,
-			minute			  	int,
+			datee 				date,
+			timee				time,
 			lon 				decimal,
 			lat 				decimal,
 			tmpf 				decimal,
@@ -299,7 +287,7 @@ def createTableQueryPartitionningByDate(table):
 			peak_wind_drct    	decimal,
 			peak_wind_time    	decimal,
 			metar             	varchar,
-			PRIMARY KEY((year, month, day), station, hour, minute)
+			PRIMARY KEY((datee), station, timee)
 		)"""
 	return query
 
@@ -311,11 +299,8 @@ def insertQueryData(row, table):
 			)
 			VALUES(
 				'{row["station"]}',
-				{row["year"]},
-				{row["month"]},
-				{row["day"]},
-				{row["hour"]},
-				{row["minute"]},
+				'{row["date"]}',
+				'{row["time"]}',
 				{row["lon"]},
 				{row["lat"]},
 				{row["tmpf"]},