Good point Lukasz, Might be a viable option … you do need to program against the SPI in that case however. I just noticed that I tried to reuse the readRequest inside a try-with-connection block and noticed that was BS and as soon as the connection dies, the pre-built request is no longer usable.
Not 100% happy with the fact, that it’s not an API feature. We could possibly move the PlcReader interface and its fellows into the API. Chris Von: Łukasz Dywicki <[email protected]> Datum: Samstag, 8. November 2025 um 14:08 An: [email protected] <[email protected]> Betreff: Re: [DISCUSS] Re-Using PlcRequests with multiple connections? Our API is constructed in such a way that end user does not need to pass connection to read/write calls. All our interactions start from connection. However we have high level interfaces such as PlcReader/PlcWriter which accept read/write requests. So to me it looks like its possible to pass request from one connection to other. You just can't use request.execute() but use reader.read(request). Best, Łukasz On 11/7/25 21:31, Christofer Dutz wrote: > Hi all, > > Today I stumbled over something when I was testing re-connect behavior when > re-using PlcReadRequests over a connection (Like using the connection-cache). > This seems to work fine as long as the initial connection is still available. > I think I wouldn’t even need to get the connection to execute the repeated > read operation. This also could result in concurrent execution of requests. > > I think possibly passing in the connection to the execute method and using > that to execute the operation would be an option. We would need to save the > connection string for which a request was built and then check this on every > execution. > > The more I think of it the more problematic do I think this issue is. > > However it would be quite a big breaking change :-( > > Chris > >
