当Spark流在view.py中完成时，我将自动更新模板变量而不刷新网页
视图.py
spark = SparkSession.builder.appName("Python Spark SQL basic example").config("spark.some.config.option", "some-value").getOrCreate()
schema = StructType().add("_c0", "integer").add("InvoiceNo", "string").add("Quantity","integer").add("InvoiceDate","date").add("UnitPrice","integer").add("CustomerID","double").add("TotalSales","integer")
INPUT_DIRECTORY = os.path.join(os.path.dirname(__file__), "../jsonFileInput")
OUTPUT_DIRECTORY = os.path.join(os.path.dirname(__file__), "../jsonFileOutput")
CHECKPOINT_DIRECTORY = os.path.join(os.path.dirname(__file__), "../writestream_checkpoints")
dataStream = spark.readStream.format("json").schema(schema).load(INPUT_DIRECTORY)
query = dataStream \
.writeStream \
.format("json") \
.option("checkpointLocation",CHECKPOINT_DIRECTORY) \
.option("path", OUTPUT_DIRECTORY) \
.start()
def dashboardV2 (request):
df = spark.read.format('json').schema(schema).option('header',True).load(OUTPUT_DIRECTORY)
totalSales = codeData.totalSales(df)
ncustomer = codeData.allCustomer(df)
return render(request,'dashboardV2.html',{
'totalSales': totalSales,
'ncustomer': ncustomer
})我需要自动更新totalsales和ncustomer变量的模板时，流数据(df变量将读取触发流的json文件，并使用它在函数中计算totalsales和ncustomer变量）
模板.html
<!DOCTYPE html>
<html>
<body>
<label class="valueTotalSales2017">$ {{totalSales}} M</label>
<label class="valueNumCus2017">{{ncustomer}}</label>
</body>
</html>代码数据.py
spark = SparkSession.builder.appName("Python Spark SQL basic example").config("spark.some.config.option", "some-value").getOrCreate()
def totalSales(df):
total = df.select('TotalSales').groupby().sum().toPandas()
total = total['sum(TotalSales)'][0]/1000000 # convert to millions
total = "%.2f" % total # trim 2 decimal
return total
def allCustomer(df):
ncus = df.select('CustomerID').distinct().count()
return ncus谢谢你的帮助。
暂无答案！
目前还没有任何答案，快来回答吧！