Rudder

I don't succeed in getting what you describe. Or at least, perhaps I did, but I need more precision.

I didn't get the log message one time: I don't know how for now, but for one case, I had some kind of synchronization error between the button in the form (saying "disabled"), and the one in the pop-up (saying: "enable").

So, in that case, I "enabled" an already enabled directive, and there wasn't any real modification in the back-end storage, and so nothing was logged - not even the message. Note that it would have been the same behavior if someone else have edited the status before my click.

That raise three questions in plus of the need to understand why, in my case, the button was not bound to the correct action:

• 1) was it your case, to ? OK, I'm quite sure you didn't remember that, so do you successed in reproducing that bug, and if yes, was it in the described case ?
• 2) when a user save without modifying anything, or if an identical modification was made by someone else in the meantime, should we advertise the user ?
• 3) should we log event without modification. The last time (for 2.3), we choose to not, but perhaps our minds changed since that time ? In all case, no Git log would be done since Git only track actual modifications.

François ARMAND wrote:

I don't succeed in getting what you describe. Or at least, perhaps I did, but I need more precision.

I didn't get the log message one time: I don't know how for now, but for one case, I had some kind of synchronization error between the button in the form (saying "disabled"), and the one in the pop-up (saying: "enable").

So, in that case, I "enabled" an already enabled directive, and there wasn't any real modification in the back-end storage, and so nothing was logged - not even the message. Note that it would have been the same behavior if someone else have edited the status before my click.

That raise three questions in plus of the need to understand why, in my case, the button was not bound to the correct action:

• 1) was it your case, to ? OK, I'm quite sure you didn't remember that, so do you successed in reproducing that bug, and if yes, was it in the described case ?

You are right. I reproduced this, but in fact, the Directive was not disabled! So nothing happened, so there was no log. This seems actually to be a duplicate of #1676. If you agree, would you please mark one as rejected?

This bug should be changed to reflect the real problem: creating a directive and then disabling it immediately appears to fail. This is less important, so degrading the priority a bit.

• 2) when a user save without modifying anything, or if an identical modification was made by someone else in the meantime, should we advertise the user ?

Ideally, yes.

• 3) should we log event without modification. The last time (for 2.3), we choose to not, but perhaps our minds changed since that time ? In all case, no Git log would be done since Git only track actual modifications.

No, I don't see any reason to record events that didn't change anything.

Jonathan CLARKE wrote:
[...]

You are right. I reproduced this, but in fact, the Directive was not disabled! So nothing happened, so there was no log. This seems actually to be a duplicate of #1676. If you agree, would you please mark one as rejected?

This bug should be changed to reflect the real problem: creating a directive and then disabling it immediately appears to fail. This is less important, so degrading the priority a bit.

OK, I reject that one, marking it as duplicated.

• 2) when a user save without modifying anything, or if an identical modification was made by someone else in the meantime, should we advertise the user ?

Ideally, yes.

See #2567

• 3) should we log event without modification. The last time (for 2.3), we choose to not, but perhaps our minds changed since that time ? In all case, no Git log would be done since Git only track actual modifications.

No, I don't see any reason to record events that didn't change anything.

OK.