calcul moyenne par mois et trace les courbes + insertion dans la table par location

insertLocation.py

+import csv
+import re
+
+def getData(container, value, type):
+    result = container[value]
+    if (result == "M" or result == "T") :
+        result = None
+
+    if (type == "string"):
+        return result
+    elif (type == "float") :
+        if (result != None):
+            result = float(result)
+        return result
+      
+    elif (type == "date") :
+        if (result != None) :
+            dateparser = re.compile("(?P<year>\d+)-(?P<month>\d+)-(?P<day>\d+) (?P<hour>\d+):(?P<minute>\d+)")
+            match = dateparser.match(container[value])
+            if not match :
+                return "continue"
+            data = match.groupdict()
+            result = (
+                int(data["year"]),
+                int(data["month"]),
+                int(data["day"]),
+                int(data["hour"]),
+                int(data["minute"]),
+            )
+            return result
+    else :
+        return None
+
+def loadataLocation(filename): 
+    with open(filename) as f:
+        for r in csv.DictReader(f):
+            timestamp = getData(r, "valid", "date")
+            if (timestamp == "continue" or timestamp == None):
+                continue
+            data = {}
+            data["valid"] = timestamp
+            data["station"] = getData(r, "station", "string")
+
+            data["lon"] = getData(r, "lon", "float")
+            data["lat"] = getData(r, "lat", "float")
+            data["tmpc"] = getData(r, "tmpc", "float")
+            data["dwpc"] = getData(r, "dwpc", "float")
+            data["relh"] = getData(r, "relh", "float")
+            data["sknt"] = getData(r, "sknt", "float")
+            data["p01m"] = getData(r, "p01m", "float")
+            data["vsby"] = getData(r, "vsby", "float")
+            data["feel"] = getData(r, "feel", "float")
+
+            yield data
+
+
+import json
+
+from cassandra.cluster import Cluster
+#Connexion au cluster
+cluster = Cluster(['localhost'])   
+session = cluster.connect('dbermond_projet')
+
+#Création de la table 
+query = '''
+    DROP TABLE IF EXISTS weatherByLocation ;
+    '''
+
+session.execute(query)
+
+query = '''
+    CREATE TABLE weatherByLocation(
+    station text,
+
+    valid_year varint,
+    valid_month varint,
+    valid_day varint,
+    valid_hour varint,
+    valid_minute varint,
+
+    lon float,
+    lat float,
+    tmpc float,
+    dwpc float,
+    relh float,
+    sknt float,
+    p01m float,
+    vsby float,
+
+    feel float,
+
+    PRIMARY KEY ((station), valid_year, valid_month, valid_day,  valid_hour, valid_minute)
+    );
+'''
+
+session.execute(query)
+
+#Fonction de chargement des données dans la table
+def getWeatherByLocation(csvfilename, session):
+    data = loadataLocation(csvfilename)
+    for r in data:
+        t = (
+            r["station"],
+
+            r["valid"][0],
+            r["valid"][1],
+            r["valid"][2],
+            r["valid"][3],
+            r["valid"][4],
+
+            r["lon"],
+            r["lat"],
+            r["tmpc"],
+            r["dwpc"],
+            r["relh"],
+            r["sknt"],
+            r["p01m"],
+            r["vsby"],
+
+            r["feel"],
+            )
+        query = """
+        INSERT INTO weatherByLocation(
+
+                station,
+
+                valid_year,
+                valid_month,
+                valid_day,
+                valid_hour,
+                valid_minute,
+
+                lon,
+                lat,
+                tmpc,
+                dwpc,
+                relh,
+                sknt,
+                p01m,
+                vsby,
+                
+                feel)
+        VALUES (%s, %s, %s ,%s ,%s, %s, %s, %s ,%s ,%s, %s, %s, %s, %s, %s)
+        """
+        session.execute(query, t)
+
+
+#Chargement des données
+
+getWeatherByLocation("/home/dbermond/Data/asos-2004>2013.txt", session)
\ No newline at end of file

meanQ1.py

+import matplotlib.pyplot as plt
+from cassandra.cluster import Cluster
+
+#Connexion au cluster
+
+cluster = Cluster(['localhost'])
+session = cluster.connect('dbermond_projet')
+
+# Equivalence entre les mots-clé et leur place dans la table
+dictParameter = {
+    "temperature": 13,
+    "dew point": 6,
+    "humidity": 11,
+    "wind speed": 12,
+    "precipitation": 10,
+    "visibility": 14,
+
+    "feel": 7,
+}
+
+#Récupération des information pour une station à une date donnée
+
+def getDataByStationAndDate(station, year, month):
+    query = '''
+        SELECT * FROM weatherbylocation WHERE station = '%s' AND valid_year = %d AND valid_month = %d ;
+        '''
+    resultat = session.execute(query % (station, year, month))
+    return resultat
+
+#Calcul de la moyenne pour un paramétre donné
+def getMean(result, parameter):
+    n = 0
+    mean = 0
+    c = dictParameter[parameter]
+    for row in result:
+        if (row[c] != None):
+            n = n + 1
+            mean += (row[c] - mean)/n
+    
+    return mean
+
+#Dessin des courbes
+def getHistoryByStation(station, parameter):
+    months = [1,2,3,4,5,6,7,8,9,10,11,12]
+    year = 2010
+    meanSeason = [0,0,0,0,0,0,0,0,0,0,0,0,]
+
+    for i in range(0,3):
+        year += 1
+        month = 0
+        listMean = []
+        for j in range(0,12):
+            month += 1
+            try :
+                result = getDataByStationAndDate(station, year, month)
+            except:
+                print("Erreur lors de la récupération des données")
+            if result != None:
+                mean = getMean(result, parameter)
+                listMean.append(mean)                       #Stock la moyenne de chaque mois dans une liste
+                meanSeason[j] += mean                       #Variable intermédiraire pour calculer la moyenne par mois sur plusieurs années
+        p = plt.plot(months, listMean, label=str(year))     #Traçage de la courbe pour une année
+
+    for k in range(0,12):
+        meanSeason[k] = meanSeason[k]/3
+
+    plt.ylabel(parameter)
+    plt.xlabel("Mois")
+    plt.plot(months, meanSeason, label='Moyenne')           #Traçage de la courbe moyenne
+    plt.legend()
+    plt.savefig("courbeMean.png")                           #Sauvegarde des courbes
+    plt.clf()
+    return meanSeason
\ No newline at end of file