Highlights from the new AF 2018 SP2 release

Blog Post created by bbregenzer on Apr 8, 2019

As you may or may not have heard, PI Server 2018 SP2 was released today (HOORAY!). I realize not everyone is an AF nerd like me who enjoys reading through release notes and playing with betas so I did this a while back and put together a preview presentation for an internal audience. Now that the official release is out, I figured why not get some extra mileage out of that material and post a blog here for others who might be interested? While this isn't an exhaustive list of all the new stuff, I still hope you find it useful. I've included the item number from the release notes if you want to cross reference and find more details.

1. 217767: Set the the default data archive name per AF Server or AF database. As most of you know, the %Server% substitution parameter will resolve the selected default server specified by the client. This can sometimes lead to confusion or unintentionally mapping to the wrong server when different users have different settings for their default data archive. Now we can specify this at the AF Server or AF db level!

Specify at the server level:

The first choice is to use the previous behavior or I can select a specific server (e.g., PISRV1 or PISRV2).

Specify at the DB level:


2. 90893: Event-weighted tag summaries. All of the tag summary functions (e.g., TagTot, TagAvg, etc.) now have an event-weighted option. That's right! We now have a native event-weighted TagTot so no more of this!


3. 215828: Time-weighted array summaries. Similar to above but the opposite for the data summary functions (e.g., Total, Avg, etc.), which were formerly only event-weighted. Useful for operations on arrays:


4. 59618: Cancel bulk backfill/recalc jobs from the Management Plugin. How many times have you started a huge backfill/recalc job in the Management Plugin of PSE and immediately regretted it (raises hand). Now you can cancel those jobs directly in the plugin!

Previously (uh oh, now I'm stuck):



5. Base template only setting. It has been often requested that a template used only as a base for other derived templates have a setting that keeps it from being used to create an element. Now that exists:

215688: Set it in PSE:

215692: Set it in PI Builder:


6. Better integration with digital tags and state sets. How many times have you manually created an enumeration set to match a digital set used by the tag you just mapped into AF? Or have you used that less than perfect workaround of data type = <Anything>? Now your life just got easier!

36885: Import digital state sets in bulk from a selected data archive using PSE:

216418: Be prompted to import the needed set when mapping a digital tag to a PI Point attribute in PSE:


7. 34888: New right-click delete options in PSE, including permanently delete items without separate check-in for increased performance:

1st option is the same as the old Delete, 2nd is the same as Delete Reference, 3rd is the new option.


8. 127550: Import analyses/notifications in disabled state. This is useful when moving new analyses to another database and/or server where you want to check the import before enabling analyses, which is easier than previous options to disable before export, find/replace in XML, or use PI Builder with the AnalysisStatus column:


9. A few other nuggets, I found interesting but haven't really played with yet, but I'll try to post some edits as I do:

59850: Exit() function to end analysis execution.

68297, 221627: “Or” searches.

144299: Split() function to make string arrays in analysis; specify string and delimitter.

140365: NewLine() function added to the StringBuilder to enable multi-line strings.


There are obviously many, many more items in the release and I encourage you to check them out. Thanks for your time and please post any questions or comments below.