Closed Jo-Be-Co closed 3 years ago
Hi,
I've already thought about merging the implementation. But in particular there are many small differences. This could also lead to downward compatibility problems (changing the properties in the outgoing message, etc.), so that such a massive change is only possible in a mayor version jump.
It's on my todo, but it's not a high priority.
To your findings:
Hi,
I've already thought about merging the implementation. But in particular there are many small differences. This could also lead to downward compatibility problems (changing the properties in the outgoing message, etc.), so that such a massive change is only possible in a mayor version jump.
It's on my todo, but it's not a high priority.
To your findings:
- I don't understand what you mean with "the 'trigger' option does not work correct"? Can you please explain a little bit more the problem here?
In the code of the blind-control there is a check for "trigger/noOverwrite/triggerOnly". If this is found there won't be any changes on expiring
, importance
, overwriteReason
or the level:
if (!onlyTrigger && node.nodeData.overwrite.active && isNaN(newPos)) {
...
} else if (!onlyTrigger && !isNaN(newPos)) {
...
}
clock-timer
on the other hand will skip the search for overrideData
in case of onlyTrigger
. Otherwise overrideData
will still be undefined
so that the following is executed:
if (node.nodeData.overwrite.active) {
node.debug(`overwrite active, check of nImportance=${nImportance} or nExpire=${nExpire}`);
if (Number.isFinite(nExpire)) {
node.debug(`set to new expiring time nExpire="${nExpire}"`);
// set to new expiring time
setExpiringOverwrite(node, dNow, nExpire, 'set new expiring time by message');
}
if (nImportance > 0) {
// set to new importance
node.nodeData.overwrite.importance = nImportance;
}
// node.debug(`overwrite exit true node.nodeData.overwrite.active=${node.nodeData.overwrite.active}, expire=${nExpire}`);
return setOverwriteReason(node);
}
Grüße ;-)
Hi, the clock-timer as clone of the blind-control does not catch up with the latest updates on blind-control.
Maybe the similar parts should be more equal or even based on a central implementation.
Actually I have spotted some faulty behaviour where I would like to contribute some fixes:
cheers, Jogi