Tayyab

59 minutes ago, Mark Derbecker said:

Your colleague could also specify the workbook ID directly, something like this:

spy.push(metadata=build,workbook='09A863DF-53F0-4D2A-BD60-A9DEDF07BC2E')

 

We're not trying to push into the same workbook. We're trying to publish new workbooks so it would not yet have an ID.

1 hour ago, Patrick said:

Hi Tayyab - One alternative would be to disable inheritance on a folder in the Corporate Drive and then removing the "Everyone" group.  To do so, create a folder in the Corporate Drive, click the "three dots" on the right, and select "Manage Permissions".  You will see that "Everyone" has access.  Click on "Advanced" and "Disable permission inheritance".  You can now remove the "Everyone" group from your folder access and add your collaborators and have a folder all your collaborators share (and not "Everyone").

 

This is helpful. Are we able to disable Everyone alltogether from the corporate folder? Or create similar tabs out like the corporate folder that we could pre-split based on our data segregation needs?

Will try this route. Still curious if we can do it for shared locations.

The workbook is already shared. The issue is as the owner, it shows in My Folder and I'm able to specify paths with Folder >> Folder1 etc, and it pushes it correctly to that folder.

When a colleague uses the same path for a push, it would create a copy of that path in his own "My Folder", rather than pushing to the same shared folder. From my perspective it's my folder that I've shared, but for him it shows up in his shared folders.

i.e. when pushing using the file path, i would specify 

workbook_name = 'Dev >> Folder_xyz >> File'
spy.push(metadata=build,workbook=workbook_name)

For someone who's seeing it in their shared folder this does not work.

I see that we can push to corporate folder in the below post, but can this be done for shared folders? If two people wanted to collaborate and push to the same folder, how would they do this? Using corporate folder is not possible for us since Everyone is forced on it and we need to segregate data from Everyone.

I would also like to avoid putting in dummy values like -999 for the other signals that don't have a pressure tag associated. Is there a way to assign a property of null/NaN? I'm thinking something like .setProperty("Max Pressure", "") but it would not be recognized as a string.

A couple of issues when it comes to capsule properties:

1. When trying to display a custom calculated property as a part of a table with assets, I get this error:
   property for reference: .setProperty("Max Pressure",$p,maxValue()), where $p is Pressure tag in kPa
   
   It works fine, when i take the add column approach and select pressure as a signal and hit max. Sometimes, the capsule property ends up correctly displaying but not after throwing a whole bunch of these errors.
   
   
2. This capsule property does not exist for all conditions under all assets, even for the same named condition. I would expect the table to leave blanks for those cells, but instead it filters them out of the table completely. So if I want to display all conditions, I would have to make two tables: one that shows the properties that have them and one for conditions that don't have the property.

When using capsule properties to do calculated values, the outputs of the signals have a lot of decimal places. I'm looking to trim that down to just 2. Example code:

($signal > $mop)
.removeLongerThan(1d)
.setProperty("MOP Percent",$mp,maxValue())

Where, $mp is $p/$l*100 (pressure signal/limit signal)

I've tried adding .round(2) to the $signal, to the output of ($signal > $mop) and to $mp within the .setProperty() and all of them have the same issue of the .round() function converting the signal to a step function first, then applying the rounding criteria. This causes problems because I want the interpolated value between two raw data points. Resampling the data to a more frequent amount is not desired for risk of losing short (<2s) spikes that need to be caught. Resampling every second would be a lot of data points to calculate when doing historical analysis.

Example condition, desired value is these to 2 decimal places. Samples are hours apart.

Using .round() I get

What's the best way to achieve this outcome?