Hi Friedrich and Branislav,

 

there is currently an initial version of the REST API and design of the 
Java-based client. I’m currently more working on the basic design and will add 
more supported API features afterwards.

 

The current feature set includes:

*       Generating API tokens from the UI
*       Getting all registered data streams and also live data (only from the 
client)
*       Getting all running adapters and creating new adapters through the API
*       Getting all pipelines and triggering pipeline operations (starting, 
stopping, deleting)
*       Creating and receiving pipeline element templates (pre-configured 
pipeline elements) that are then displayed in the UI

 

On my roadmap for the following days is receiving historical data for each data 
stream.

 

There are two things that will be a little bit more complicated:

*       Creating pipelines through the API: As pipeline elements can be 
installed on the fly and the set of pipeline elements is not known before 
runtime, there cannot be a very specific API for handling pipelines and a 
generic way is not-so user-friendly (e.g., you need to know all configurations 
of pipelines and we have some pipeline elements modify their configuration at 
runtime). I’m thinking of some code generation feature that depends on all 
currently registered pipeline elements and will create helper classes to ease 
the creation of pipelines. In a first version, you would probably need to know 
which pipeline elements are available and how they are configured and can then 
post a new pipeline to the API.
*       For accessing live data from the REST API, I think of replacing the 
current sinks for the Dashboard and Data Explorer by a new concept “Virtual 
Stream”. All streams (created from adapters and virtual streams) are 
automatically persisted and users would be able to configure a retention time 
for stream persistence. The dashboard and data explorer would only rely on 
streams instead of a pipeline (so you just select a stream and get live or 
historic data back). That would ease our conceptual model a bit and also 
improve usability, I guess. In this case, there could also be an HTTP endpoint 
that exposes the latest value of each stream directly from the timeseries 
database in the background.

 

What do you think? Is there anything else you think should be supported for the 
initial version of the API?

 

I’m also adding the dev list to cc, so that everyone is up-to-date and can give 
feedback.

 

Dominik

 

 

From: Philipp Zehnder 
Sent: Thursday, January 7, 2021 12:07 PM
To: Volz, Friedrich <[email protected]>
Cc: Branislav Jovicic <[email protected]>; Dominik Riemer 
<[email protected]>
Subject: Re: Streampipes API

 

Hi Friedrich,

 

thanks for your mail. Yes, the goal is to also provide access to pipelines and 
data.

Currently there is no documentation for the API, but there is an open Issue for 
the feature in Jira [1].

Dominik (see in cc) is working on it, if you have specific questions you can 
also ask him or write to the mailing list.

 

Kind regards,

Philipp

 

[1] 
https://issues.apache.org/jira/projects/STREAMPIPES/issues/STREAMPIPES-272?filter=allopenissues





On 7. Jan 2021, at 09:24, Volz, Friedrich <[email protected] 
<mailto:[email protected]> > wrote:

 

Hi Philipp,

 

Branislav and me had some questions regarding the upcoming Streampipes API. Is 
there documentation or plans on how this will work?

I would like to access pipelines and data from pipelines over this API. Is this 
what the API is made for?

 

Best regards

Friedrich

 

 

.........................................................

M. Sc. Philipp Zehnder

Wissenschaftlicher Mitarbeiter | Research Scientist

Information Process Engineering (IPE)

 

FZI Forschungszentrum Informatik

Haid-und-Neu-Str. 10–14 

76131 Karlsruhe, Germany

Tel.: +49 721 9654-805

Fax: +49 721 9654-806

 

 <mailto:[email protected]> [email protected]

https://www.fzi.de/mitarbeiter/philipp-zehnder

 

.........................................................

FZI Forschungszentrum Informatik

Stiftung des bürgerlichen Rechts

Stiftung Az: 14-0563.1 Regierungspräsidium Karlsruhe

Vorstand: Prof. Dr. Andreas Oberweis, Jan Wiesenberger, Prof. Dr.-Ing. J. 
Marius Zöllner

Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus

.........................................................

 

Reply via email to