Forums PDG/TOPs how to get unique ID from a workitem index?

how to get unique ID from a workitem index?

   2601   4   1
User Avatar
Andr
Member
900 posts
Joined: Feb. 2016
Offline
 
dict = {} ## pdg_index and unique id     
for w in pdg_node.workItems:
   dict[w.index] = w.id

I'm using this code to reconstruct an ordered dictionary of workitems of a pdg_node and their corresponding unique ID.
Problem: I just need to know a single specific unique ID, from a workitem index of a pdg.node. I don't want to iterate over all the workitems to reconstruct a dictionary.

Naively, I thought I could just access a workitem like this: 
pdg_node.workItems[3]
and then asking for its unique workItem.id, but by doing so I'm not getting the workitem with index 3.

Indeed the following code produce a unordered list of workitems, and an ordered dictionary of workitems. Might be a python thing? I didn't know that dictionaries get auto-sorted. My question remain, anyway..
wlist = [] 
dict = {}
for w in pdg_node.workItems:
    wlist.append(w.index)
 
for w in pdg_node.workItems:  
    dict[w.index] = w.id
Edited by Andr - Oct. 1, 2019 04:57:17
 Report
User Avatar
Ostap
Member
209 posts
Joined: Nov. 2010
Offline
 
Hi,

But what is wrong with 'pdg_node.workItems.index'? if you set index 3 for workitem 3 then it returns 3.


If I understand it properly, ID is just your scene workitem unique id.
 Report
User Avatar
Andr
Member
900 posts
Joined: Feb. 2016
Offline
 
Hello,
my aim is to know the unique ID, from a workitem index of a top node, without doing any loop.
I just found out that a Python Script node in your topnet (even blank) mixes things up.
And you can't use a simple 
pdg_node.workItems[52]
to get the workitem index 52 of that Top Node.
The pdg_node.workItems list you get from the Python Script node is probably sorted raw based on the order of the processed workitems by the node. I guess. (on each new cook it will change indeed)
You have have to create a new list of workitems and resort it based on their index, and that's what the dictionary approach seems to do (even if I don't understand why it does this autosorting)

Try the attached example:
topnode = hou.pwd().inputs()[0].node('./OUT')
pdgnode = topnode.getPDGNode()
check_index = hou.pwd().parm('checkworkitem').eval()

wlist = [] 
for w in pdgnode.workItems:
    wlist.append(w)
    #print w.index

wdict = {}
for w in pdgnode.workItems:  
    wdict[w.index] = w.id
    
print wdict[check_index], wlist[check_index].id
## the two numbers should be equal, but they won't if you have a python script node in your topnet
## bypass it to see matching pairs of numbers displayed
Edited by Andr - Oct. 1, 2019 05:34:48

Attachments:
uniqueidfromindex.hiplc (128.2 KB)
workitemsindexJPG.JPG (160.4 KB)

 Report
User Avatar
chrisgreb
Member
603 posts
Joined: Sept. 2016
Offline
 
The work item index is just an arbitrary number, it doesn't have to be unique although nodes usually follow the convention of assigning indexes so that they form a continuous range.

Also, dictionaries don't get auto-sorted, it's because it's implemented using a hash table so it just happens that the .values() are in increasing order of their keys because hash(i) = i for integers.
Edited by chrisgreb - Oct. 3, 2019 21:11:53
 Report
User Avatar
Ostap
Member
209 posts
Joined: Nov. 2010
Offline
 
Task ID and Index are different things (They should not be equal).
As Chris mentioned that index is just an arbitrary number (you can assign whatever you want).
 Report
  • Quick Links