Behavior of time based operators

[Harshvardhan Agrawal]

Hello,

I have recently started reading Stream Processing with Apache Flink by Fabian and Vasiliki. In Chapter 3 of the book there is a statement that says: None of the functions expose an API to set time stamps of emitted records, manipulate the event-time clock of a task, or emit watermarks. Instead, time-based DataStream operator tasks internally set the time stamps of emitted records to ensure that they are properly aligned with the emitted watermarks. For instance, a time-window operator task attached the end time of a window as time stamp to all records emitted by the window computation before it emits the watermark with the time stamp that triggered the computation of the window.

Does this mean that time stamps in the records are overwritten by these time-based operators when using Event Time?

--

[Elias Levy]

See the section on Operators here https://docs.google.com/document/d/1b5d-hTdJQsPH3YD0zTB4ZqodinZVHFomKvt41FfUPMc/edit?usp=sharing

On Thu, Aug 2, 2018 at 3:42 PM Harshvardhan Agrawal <[hidden email]> wrote:

Hello,

I have recently started reading Stream Processing with Apache Flink by Fabian and Vasiliki. In Chapter 3 of the book there is a statement that says: None of the functions expose an API to set time stamps of emitted records, manipulate the event-time clock of a task, or emit watermarks. Instead, time-based DataStream operator tasks internally set the time stamps of emitted records to ensure that they are properly aligned with the emitted watermarks. For instance, a time-window operator task attached the end time of a window as time stamp to all records emitted by the window computation before it emits the watermark with the time stamp that triggered the computation of the window.

Does this mean that time stamps in the records are overwritten by these time-based operators when using Event Time?

--