So I have a carefully crafted ordered Output list via the python node.
Now when I later down the pipe want to pick up the 1st file in that list I should be able to use
`@pdg_input.0`
also
`pdginput(0,'',1)`
results in the same outcome
But the result is not correct, I get the result as if the list was alphabetically ordered. Am I doing something wrong, is this as expected and can't we depend on the order of the output files? Using production build .499.
Do you have an example file you can attach? Most TOP nodes don't do any reordering of files, unless they have options to do so. Do the files appear in the correct order in the MMB?
If you look at the screenshot above you can see I've manually reordered the output files in a python node, they appear in correct order in the correct order with MMB debug window, but when I use @pdg_input.0, it gets the file that is alphabetically the first instead of the order as shown.
Here's a test file. Looking at it now I think somehow the python node that sort output items causes the real issue. But after that node issues occur.
By default, the code in the Python Script node runs when the work item cooks. It's the “job” of that work item. The attribute create node is static in that file, though – it generates work items before the graph begins to cook. That means that when it generates, there won't be any reordered files on the input work item. That's also why it has warnings on the node.
You need to change your Python Script so the “Evaluate Script During” to “Generate” instead of “Cook (In-Process)”.
While you're browsing the forum, could you perhaps have a quick glance at this post : https://www.sidefx.com/forum/topic/74851/ [www.sidefx.com] Looking for a reliable way to run a houdini instance where I'm constantly listening to input data. Thanks!
