system.perspective.getSessionInfo / page path

General Discussion
1

I use system.perspective.getSessionInfo to create this nifty page to see who's in the system.

image
image1010×563 30.2 KB

What I's really like to add is to know what page path the users have loaded in the pageids. Essentially the contents of below:
image

Any ideas? I have put a suggestion in the suggestion box to add this to getSessionInfo, just thought I'd put this out there to get any other ideas I may be missing...

Thanks!

2

I would guess implementing this would be difficult as a session could be using multiple pages at any one time in different tabs/windows and maybe even different browsers.

I've previosuly wanted to do something similiar so that a user would have their previous pages automatically open when they logged back in, but we struggled to get any information of this sort, and it might not be possible. I wonder if being browser based would limit what kind of data can be returned.

3

Yes.

You would have to track such things yourself, in the various session/page/view life cycle events, while the session is active. But it will be a bit tricky, as you will likely miss page/view closure events when the user closes tabs (browsers don't necessarily honor JS requests to intervene in those operations).

4

I started working that and saving to a tag but then decided I didn't need it that bad. Figured this would be something IA could add as they do know the pages from the server in the session details / Pages, you know, if they wanted to...

Just making sure I wasn't missing something.

image
image1407×458 35.5 KB

image

5

The session info payload is deliberately flattened so that it's easier to represent in a '2d' format like a dataset, and so it's not (even more complicated) for new users.

In the meantime, you could get access to the same information the status page gives you with a modification of the code here:

1 Like
6

Thank you, very kindly sir! I'll look into this.

7

In case folks are interested, I did get this to work. I noticed most of my main pages would load at address 'C$0:0$0:2', a couple base pages at 'C'. This helped me target the path that I wanted to see and not all the embedded views. Thanks again everyone for your help!

inputPageIds = []	# my fuunction passes in pageids from getSessionInfo() call

from com.inductiveautomation.ignition.gateway import IgnitionGateway

project_name = system.util.getProjectName()
# project_name = 'myProject'

context = IgnitionGateway.get()
mod_mgr = context.getModuleManager()
# need to get "com.inductiveautomation.perspective.common.api.PropertyType" from module manager to
# get the "props" and "custom" enums for fetching the session props and the custom props
prop_type_class = mod_mgr.resolveClass('com.inductiveautomation.perspective.common.api.PropertyType')
# the below uses the gateway context module manager to function essentially as an import statement
persp_ctx_class = mod_mgr.resolveClass('com.inductiveautomation.perspective.gateway.api.PerspectiveContext')
# instantiate the PerspectiveContext using the gateway context
persp_ctx = persp_ctx_class.get(context)
# get the PerspectiveSessionMonitor object running in gateway context
psm = persp_ctx.getSessionMonitor()
# grab <InternalSession> object list for the target project
int_sessions = psm.getClientSessionsForProject(project_name)
#	return int_sessions
active_sessions = {}
#	listResults = []
dictResults = {}

# loop over internal sessions to build a map of active sessions with the target view open
for i, sess in enumerate(int_sessions):
	# loop over the open pages for each session and get the id for each
	for page in sess.getPages():
		# this returns the page id that matches the gateway Status -> Connections -> Perspective Sessions -> Session detail page
		page_id = page.getId()
		if page_id not in inputPageIDs : continue
		views = page.getViews()
		for view in views :
			system.perspective.print(str(view.getId()))
			result = str(view.getId()).split('@') 	# split at @.  left (0) is page path, right (1) is address
			# With the way I have things configured, all my pages appear to be at address C$0:0$0:2,
			#	a few home pages come in at "C".  This takes 'C$0:0$0:2' if there, 'C' if not
			if result[1] in ['C', 'C$0:0$0:2' ] :	
				pageKey = "_" + page_id 
				if pageKey in dictResults.keys() :
					if dictResults[pageKey]['key'] != 'C' : continue
					
#					listResults.append( [page_id, result] )
				# write to dict structure formatted string of page path.  (some string parts removed for easier reading.)
				dictResults.update( { "_" + page_id : { "path" : str(result[0]).split("/",1)[1].replace("/Large",""), "key" : str(result[1]) } } )
				if str(result[1]) == 'C$0:0$0:2' : break  # no need to keep looking on views if I found a 'C$0:0$0:2'
				
return dictResults
3 Likes