Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi Bo,

Thanks for your feedback.

The first task presented to the analyst is Select Action.  Before completing the task, he must edit the field  multiselect field called "Action" to specify which actions need to be performed.  For example, if he selects Action1 and Action3, the corresponding scripts (Action1 and Action3) will be executed but the other two (Action2 and Action4) will not. This will make the Wait point wait forever.

The only solution I could come up with is the one below.  The Noop script does nothing: it is just there to make the paths converge before entering the Wait point.

But the playbook looks less simple.  If you have any other idea, I would be glad to hear them.  Thanks

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi Pierre - 

This is the right idea. I think you could simply send those "Else" (Si-non in the French translation) directly to the wait point, rather than going through the noop script. 

I do agree that having those four condition points is clunky -- but I think that it is necessary to achieve the goal that you have.

Hi Bo,

Thanks for your feedback.

The first task presented to the analyst is Select Action.  Before completing the task, he must edit the field  multiselect field called "Action" to specify which actions need to be performed.  For example, if he selects Action1 and Action3, the corresponding scripts (Action1 and Action3) will be executed but the other two (Action2 and Action4) will not. This will make the Wait point wait forever.

The only solution I could come up with is the one below.  The Noop script does nothing: it is just there to make the paths converge before entering the Wait point.

But the playbook looks less simple.  If you have any other idea, I would be glad to hear them.  Thanks

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Also - you can define multiple conditions for one condition point. So that would allow you to simplify it to one condition point with four conditions, each pointing to a different script, and the "Else" going to the wait point. Then each script would also send to the wait point.

I've achieved this by clicking "Create condition" as many times as needed, and using the setting: "First true condition". Then when attaching the script node from the condition node, I select the condition that is appropriate for my script

Hi Pierre - 

This is the right idea. I think you could simply send those "Else" (Si-non in the French translation) directly to the wait point, rather than going through the noop script. 

I do agree that having those four condition points is clunky -- but I think that it is necessary to achieve the goal that you have.

Hi Bo,

Thanks for your feedback.

The first task presented to the analyst is Select Action.  Before completing the task, he must edit the field  multiselect field called "Action" to specify which actions need to be performed.  For example, if he selects Action1 and Action3, the corresponding scripts (Action1 and Action3) will be executed but the other two (Action2 and Action4) will not. This will make the Wait point wait forever.

The only solution I could come up with is the one below.  The Noop script does nothing: it is just there to make the paths converge before entering the Wait point.

But the playbook looks less simple.  If you have any other idea, I would be glad to hear them.  Thanks

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Apologies for the multiple replies. You should actually use "Any true condition" for your case so that all the applicable paths are followed, based on their conditions. Read more on this here: https://www.ibm.com/docs/en/sqsp/51?topic=points-condition

Also - you can define multiple conditions for one condition point. So that would allow you to simplify it to one condition point with four conditions, each pointing to a different script, and the "Else" going to the wait point. Then each script would also send to the wait point.

I've achieved this by clicking "Create condition" as many times as needed, and using the setting: "First true condition". Then when attaching the script node from the condition node, I select the condition that is appropriate for my script

Hi Pierre - 

This is the right idea. I think you could simply send those "Else" (Si-non in the French translation) directly to the wait point, rather than going through the noop script. 

I do agree that having those four condition points is clunky -- but I think that it is necessary to achieve the goal that you have.

Hi Bo,

Thanks for your feedback.

The first task presented to the analyst is Select Action.  Before completing the task, he must edit the field  multiselect field called "Action" to specify which actions need to be performed.  For example, if he selects Action1 and Action3, the corresponding scripts (Action1 and Action3) will be executed but the other two (Action2 and Action4) will not. This will make the Wait point wait forever.

The only solution I could come up with is the one below.  The Noop script does nothing: it is just there to make the paths converge before entering the Wait point.

But the playbook looks less simple.  If you have any other idea, I would be glad to hear them.  Thanks

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi Pierre - 

This is the right idea. I think you could simply send those "Else" (Si-non in the French translation) directly to the wait point, rather than going through the noop script. 

I do agree that having those four condition points is clunky -- but I think that it is necessary to achieve the goal that you have.

Hi Bo,

Thanks for your feedback.

The first task presented to the analyst is Select Action.  Before completing the task, he must edit the field  multiselect field called "Action" to specify which actions need to be performed.  For example, if he selects Action1 and Action3, the corresponding scripts (Action1 and Action3) will be executed but the other two (Action2 and Action4) will not. This will make the Wait point wait forever.

The only solution I could come up with is the one below.  The Noop script does nothing: it is just there to make the paths converge before entering the Wait point.

But the playbook looks less simple.  If you have any other idea, I would be glad to hear them.  Thanks

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

This is the solution I will use.

I think the Noop scripts are necessary though.  When leaving the condition point, you cannot directly go into a wait point: The condition is either true or not. If both branches enter a wait point, only one of them will have been taken, either the true branch or the else branch and the wait point will wait indefinitely.  At least this is my comprehension of the behaviour of a wait point.

Hi Pierre - 

This is the right idea. I think you could simply send those "Else" (Si-non in the French translation) directly to the wait point, rather than going through the noop script. 

I do agree that having those four condition points is clunky -- but I think that it is necessary to achieve the goal that you have.

Hi Bo,

Thanks for your feedback.

The first task presented to the analyst is Select Action.  Before completing the task, he must edit the field  multiselect field called "Action" to specify which actions need to be performed.  For example, if he selects Action1 and Action3, the corresponding scripts (Action1 and Action3) will be executed but the other two (Action2 and Action4) will not. This will make the Wait point wait forever.

The only solution I could come up with is the one below.  The Noop script does nothing: it is just there to make the paths converge before entering the Wait point.

But the playbook looks less simple.  If you have any other idea, I would be glad to hear them.  Thanks

Hi Pierre -

Can you elaborate on when "not all the scripts may need to be performed" and what would determine that? It sounds to me like you need a more involved condition point with more branches.

In general, the wait point is designed to do exactly what you're observing -- gather all incoming branches of the playbook and wait until those branches have complete.

You could consider making multiple endpoints in the playbook, one for each path through the conditions and scripts. Or else, like I suggested above, expand your conditional node to be more specific.

If you share with me more details of the condition and when you want what to happen, I'd be happy to give feedback on that.

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

What I would do in your place is change the conditional block into a function. The output would be a dict, action names are the keys and the values would be true or false. If you need to run action1 and action2 the dict would be {"action1": True, "action2": True, "action3": False, "action4":False} then in each action script check if it should run the whole code or pass. This way all the actions would technically run but only the ones that need to do something would do it and the wait point won't wait forever. 

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

I like this option. I will have to try this out!

The only place where I see this failing and having to go back to the second post the OP did was if you are performing functions and not just internal scripts. This would be because the internal scripts could have the logic built in, do this if this is true and like you said script always completes. If you have a function in there instead you would then need to fall back to the option of the decision point and needing that useless script on the end to bring it back together before moving forward to the wait point. Do you potentially have a solution I am not seeing for that?

What I would do in your place is change the conditional block into a function. The output would be a dict, action names are the keys and the values would be true or false. If you need to run action1 and action2 the dict would be {"action1": True, "action2": True, "action3": False, "action4":False} then in each action script check if it should run the whole code or pass. This way all the actions would technically run but only the ones that need to do something would do it and the wait point won't wait forever. 

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

If you're using functions you would still need to use a script in the function to set all the variables so the if would be there too.

I like this option. I will have to try this out!

The only place where I see this failing and having to go back to the second post the OP did was if you are performing functions and not just internal scripts. This would be because the internal scripts could have the logic built in, do this if this is true and like you said script always completes. If you have a function in there instead you would then need to fall back to the option of the decision point and needing that useless script on the end to bring it back together before moving forward to the wait point. Do you potentially have a solution I am not seeing for that?

What I would do in your place is change the conditional block into a function. The output would be a dict, action names are the keys and the values would be true or false. If you need to run action1 and action2 the dict would be {"action1": True, "action2": True, "action3": False, "action4":False} then in each action script check if it should run the whole code or pass. This way all the actions would technically run but only the ones that need to do something would do it and the wait point won't wait forever. 

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

What I would do in your place is change the conditional block into a function. The output would be a dict, action names are the keys and the values would be true or false. If you need to run action1 and action2 the dict would be {"action1": True, "action2": True, "action3": False, "action4":False} then in each action script check if it should run the whole code or pass. This way all the actions would technically run but only the ones that need to do something would do it and the wait point won't wait forever. 

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?

Hi Maria,

Your post had me thinking.  Maybe I could dispense with the the condition, the function and the Noop script with this new version.

As you explained, each script could begin with an If statement like:

If "action1" in playbook.properties.action:

  do what the script has to do

Else:

  do nothing and end the script

It's even simpler than the original version!  Using a condition point seemed so natural but it just complicated things.

Thnaks

What I would do in your place is change the conditional block into a function. The output would be a dict, action names are the keys and the values would be true or false. If you need to run action1 and action2 the dict would be {"action1": True, "action2": True, "action3": False, "action4":False} then in each action script check if it should run the whole code or pass. This way all the actions would technically run but only the ones that need to do something would do it and the wait point won't wait forever. 

Hi everybody,

I am writing a playbook like the one below.  I created a multiselect field called "Action" which can take one or more values: Action1, Action2, Action3, Action4.

The condition is configured with "Any true condition".

For each value selected, there is an associated script that needs to be executed.  When all the selected srcipt(s) have been executed, there is a Wait point to make sure each script has finished.  The problem is: because not all the scripts may need to be performed, the wait point will wait forever for those not run  and the playbook will never terminate.

Does anyone can suggest a way around this situation?