Hi, guys
When upgrading to Flink 1.10 rc0, I am confused by the new DataTypes schema defination.
I am reading and consuming records from kafka with json schema like {"user_id":1866071598998,"recv_time":1547011832729}. and the code snippet is :
.withSchema(
new Schema()
// eventTime
.field("rowtime", DataTypes.TIMESTAMP(3).bridgedTo(java.sql.Timestamp.class)).rowtime(
new Rowtime()
.timestampsFromField("recv_time")
.watermarksPeriodicBounded(1000)
)
.field("user_id", DataTypes.STRING())
But, I am running an issue and got exception like the following:
Caused by: java.time.format.DateTimeParseException: Text '1549705104542' could not be parsed at index 0
at java.time.format.DateTimeFormatter.parseResolved0(DateTimeFormatter.java:1949)
at java.time.format.DateTimeFormatter.parse(DateTimeFormatter.java:1777)
at org.apache.flink.formats.json.JsonRowDeserializationSchema.convertToLocalDateTime(JsonRowDeserializationSchema.java:366)
at org.apache.flink.formats.json.JsonRowDeserializationSchema.lambda$wrapIntoNullableConverter$d586c97$1(JsonRowDeserializationSchema.java:236)
at org.apache.flink.formats.json.JsonRowDeserializationSchema.convertField(JsonRowDeserializationSchema.java:439)
at org.apache.flink.formats.json.JsonRowDeserializationSchema.lambda$assembleRowConverter$77f7700$1(JsonRowDeserializationSchema.java:418)
at org.apache.flink.formats.json.JsonRowDeserializationSchema.lambda$wrapIntoNullableConverter$d586c97$1(JsonRowDeserializationSchema.java:236)
at org.apache.flink.formats.json.JsonRowDeserializationSchema.deserialize(JsonRowDeserializationSchema.java:131)
Locked
