In PI SDK Utility → KST Cleanup, there are 3 check boxes and 1 text box. When a check box is checked, the check box's label and description are added to the text box.
Since there are only 3 check boxes, just eliminate the text box and have each check box description show underneath its corresponding check box. This would look a lot neater.1 vote
As a PI user, I want the OPC DA and HDA Servers to honor the Windows identity of any connected OPC client.3 votes
We understand the requirement. It makes sense. We have been looking into it but it turns out to be difficult to do. We used a third-party framework to build the PI OPC DA Server which separates interactions with the OPC client from interactions with the PI Data Archive. It solved a number of problems but unfortunately there is no easy way to have OPC client user credentials flow to the PI Data Archive. We are evaluating our options.
Currently the GetEventFrames method returns only the properties of the event frame (start time, end time, extended properties, severity, ect.). In order to get a table including the event frame attributes, you need to get the event frame list first and then go through each event frame to get the related attribute values.
It would be helpful if the GetEventFrames function returned a snapshot of the event frame's attribute values to improve the speed of custom applications.0 votes
Add the ability to retrieve Attributes and Elements by category for Powewhell command-lets.
$cat = Get-AFCategory -Name "Product A Manufacturing" -Element -AFDatabase $afDB
get-AFElement -category $cat -AFDatabase $afDB
Get a list of all the elements in category "Product A Manufacturing"
$cat = Get-AFCategory -Name "KPIs" -Attribute -AFDatabase $afDB
Get-AFAttribute -category $cat
Get a list of all attributes in category "KPIs"0 votes
The stream endpoints should allow passing input attributes/values if the attribute's Data Reference supports it (refer to the AFDataReference.GetValues method in the AF SDK).
For example this could look something like:
One strong use-case for this would be an attribute configured as a Table Lookup Data Reference that returns a time series but allows a parameter to be defined.0 votes
Customer would like to write a query to return a time column with a negative offset in the PI OLEDB Provider such as:
SELECT time, (time - 1d) as datekey
This query works
SELECT time, (time + 1d) as datekey
A similar query can be written using a linked server but requires the additional overhead of a sql server or ssms.
SELECT time, DATEADD(day, -1, time) as datekey
WHERE tag='sinusoid'1 vote
Recalculate an Event Frame after a manual entry through PI Vision.
So after inserting data in a manual manner, I want to be able to automatically recalculate the EF. In new AF version this works for analysis but not for EF. There is a way to do program it by using PI AF SDK? I would be happy if this can be done already from PI Vision with PIweb API.0 votes
Currently, a method like PIServer.CreatePIPoints returns AFListResults<String, PIPoint>. Very similarly, PIPoint.FindPIPointsByPath will return AFKeyedResults<String, PIPoint>. The functionality of each result is essentially the same; they even have, for most practical purposes, the same associated properties and methods and thus; they should not be different object types.
Sharing an object type across all of these PIPoint-related functionalities would allow developers to more easily share methods for these related functions (Such as error handling methods).0 votes
When requesting time series data for multiple attributes using a streamset request and the PlotValues endpoint, it would be helpful if the response had the timestamps all lined up. This makes it easier to render the data, particularly if the data is being visualized in a stacked area series.0 votes
This is something we would like to do but the nature of PlotValues makes this difficult to do if not impossible. The PlotValues endpoint returns a time series with the most visually significant values for each pixel on a display. Even though the data values are actual recorded values, the values from different data streams have nothing to do with each other. The timestamps for different streams probably won’t line up.
- Don't see your idea?