Flink SQL CLI client CREATE TABLE from Kafka - apache-kafka

Question

I am trying to create a table in Apache Flink SQL client. I want to filter my JSON data in Flink, which arrives continously from a Kafka cluster.
The JSON looks like this:
{"lat":25.77,"lon":-80.19,"timezone":"America\/New_York",
"timezone_offset":-14400,
"current.dt":1592151550,
"current.sunrise":1592130546,
"current.sunset":1592179999,
"current.temp":302.77,
"current.feels_like":306.9,
"current.pressure":1017,
"current.humidity":78,
"current.dew_point":298.52,
"current.uvi":11.97,
"current.clouds":75,
"current.visibility":16093,
"current.wind_speed":3.6,
"current.wind_deg":60,
"current.weather.0.id":803,
"current.weather.0.main":"Clouds",
"current.weather.0.description":"broken clouds",
"current.weather.0.icon":"04d"}
The part I am interested in :
"current.weather.0.description":"broken clouds"
I want to filter my data whenever the current.weather description is "moderate rain". I tried to create two tables in Flink:
the Rain table, where the whole JSON arrives, and
where my filtered data will be stored and sent back to another Kafka cluster.
CREATE TABLE Rain (current.weather.0.description varchar) WITH ('connector.type' = 'kafka',
'connector.version' = 'universal',
'connector.topic' = 'WeatherRawData',
'format.type' = 'json',
'connector.properties.0.key' = 'bootstrap.servers',
'connector.properties.0.value' = 'kafka:9092',
'connector.properties.1.key' = 'group.id',
'connector.properties.1.value' = 'flink-input-group',
'connector.startup-mode' = 'earliest-offset'
);
CREATE TABLE ProcessedRain(
current.weather.0.description varchar
) WITH (
'connector.type' = 'kafka',
'connector.version' = 'universal',
'connector.topic' = 'WeatherProcessedData',
'format.type' = 'json',
'connector.properties.0.key' = 'bootstrap.servers',
'connector.properties.0.value' = 'kafka:9092',
'connector.properties.1.key' = 'group.id',
'connector.properties.1.value' = 'flink-output-group'
);
The error message I get :
[ERROR] Could not execute SQL statement. Reason: org.apache.flink.table.api.SqlParserException: SQL parse failed. Encountered "current" at line 1, column 20. Was expecting one of:
"PRIMARY" ...
"UNIQUE" ...
"WATERMARK" ...
<BRACKET_QUOTED_IDENTIFIER> ...
<QUOTED_IDENTIFIER> ...
<BACK_QUOTED_IDENTIFIER> ...
<IDENTIFIER> ...
<UNICODE_QUOTED_IDENTIFIER> ...
How should my CREATE TABLE be created correctly?

Answer 1

I think it should be
CREATE TABLE ProcessedRain (
`current.weather.0.description` VARCHAR
) WITH (
'connector.type' = 'kafka',
'connector.version' = 'universal',
'connector.topic' = 'WeatherProcessedData',
'format.type' = 'json',
'connector.properties.bootstrap.servers' = 'kafka:9092',
'connector.properties.group.id' = 'flink-output-group'
);