Hi, hmpf ... would have been a lot less upset if I had gotten these emails last week :-/
Well ... after being called out for mixing my open-source and my $job work last week, I decided to strictly separate the both. I will therefore not be taking the plan I already had. Doing it manually would be a bit too much work as I want to finalize the release within the next 2 days. I guess this will be something for later. Chris On 2026/08/20 13:26:03 QuanticPony wrote: > +1 to all. > I don't think there will be a better time to clean them than now. > > Unai > > On Thursday, August 20th, 2026 at 3:10 PM, Christofer Dutz > <[email protected]> wrote: > > > And ideally …. Especially in order to help after updating: > > > > I’d love to have the config parser warn about using unknown config > > parameters. > > > > I think in the past we had several people reporting issues which turned out > > that they were using unknown or old versions of renamed config parameters.. > > > > Explicitly warning about this would make our life a lot simpler. > > > > Chris > > > > > > Von: Christofer Dutz <[email protected]> > > Datum: Donnerstag, 20. August 2026 um 15:04 > > An: [email protected] <[email protected]> > > Betreff: Cleaning up on the naming of config parameters? > > > > Hi all, > > > > I recently cleaned up the naming of all config parameters in my own drivers > > as I noticed that the same thing was configured using different names in > > PLC4X. > > I would propose to do this here too … this is the 1.0.0 release, I would > > like to stop constantly renaming things with this and I guess our users are > > expecting some breaking changes anyway. I’d love them to do this now and > > once and not with every minor version update. > > > > Also did I start adding the measure to the names … as some were ms and some > > microseconds (my Profinet driver) … adding mostly „-ms“ makes it clear. > > > > ┌────────────────────────────────────────────────────────────┬───────────────────┬────────────────────────────────────────────────────────────────────────────────────────────────────────┐ > > │ Current │ > > Proposed │ Why > > │ > > > > ├────────────────────────────────────────────────────────────┼───────────────────┼────────────────────────────────────────────────────────────────────────────────────────────────────────┤ > > │ keystore, keystore-password, keystore-type (TLS transport) │ > > key-store-* │ OPC UA already uses key-store-*, and trust-store-* is > > hyphenated everywhere — TLS is the outlier │ > > > > ├────────────────────────────────────────────────────────────┼───────────────────┼────────────────────────────────────────────────────────────────────────────────────────────────────────┤ > > │ timeout-request (ADS) │ > > request-timeout │ 15 other drivers already use that order; ADS stands > > alone │ > > > > ├────────────────────────────────────────────────────────────┼───────────────────┼────────────────────────────────────────────────────────────────────────────────────────────────────────┤ > > │ tcp-no-delay, tls-version │ no-delay, > > version │ Transport params are addressed <code>.<name>, so today these read > > tcp.tcp-no-delay and tls.tls-version │ > > > > ├────────────────────────────────────────────────────────────┼───────────────────┼────────────────────────────────────────────────────────────────────────────────────────────────────────┤ > > │ 10 duration params │ …-ms > > suffix │ So nobody opens the source to learn whether a number is > > seconds or millis │ > > > > └────────────────────────────────────────────────────────────┴───────────────────┴────────────────────────────────────────────────────────────────────────────────────────────────────────┘ > > > > What do you think? > > > > Chris > > > > >
