General Questions

General
#1

Maybe You can Change the title later

I think there is not a particular place for asking regualr questions in implementin Pyblish and everyone has any question have to ask it in new topic, anyway…

in CEVI we have this:

Methodology
A CVEI-compatible stack assumes that publishes are idempotent and
that published content is immutable. What does that mean? To be
idempotent means that publishing once has the exact same effect as
publishing twice. It means that publishing doesn’t have any effect on
subsequent publishes. The one exception to this rule is immutability. To
be immutable means to never change. Which means that once content has
been published, it is never modified in any way, such as overwriting it
with more recent data. Because it can never be altered, it means that new
data must be created each time a publish occurs, which is the exception
to idempotency as publishing twice then means to output onto different
locations.

CVEI consists of four stages of orthogonal responsibilities.
- Collection is about collecting information
- Validation is about validating this information
- Extraction serialises this information
- Integration then integrates the serialised information with the surrounding pipeline

Each stage runs after the other and hinges on the success of Validation; if Validation fails, nothing is written.

Questions:

  1. in extraction area, where does it save data? on temporary folders structure? and we define our temporary folders? and after integration can we delete the temporary folders?
  2. if we want to have a final published folder and save published data on one file with supervisors, it is possible to do that in our plugins? i mean can we ignore immutabity of it based on our needs?

thanks

#2

Think of the topic title as the question, and the body of your post to contain more details of the problem, and any relevant examples.

The idea is to enable others to search for questions that have already been asked. Similar to StackOverflow. As such, try to ask only 1 question per thread.

I’ll modify the title now to make it more discoverable, but try keeping this in mind for the next question.

Edit: I can’t simplify the thread title. There are too many questions in your post.

I’ll close this and ask that you open a new thread with a single question so it can be found by searching later on.

CVEI are guidelines for how you might want to think about your pipeline, but it is by no means mandatory. You could do exactly as you described, or you could come up with your own, perhaps more suitable, approach.

As an example, you might extract to…

c:\tmp\my_publish

And during integration, you might move this to…

\\server\hulk\seq005\5300\publish\v012

And then delete the temporary folder.

Or you can extract into the final directory in one go and see where that takes you.

Of course. Wherever CVEI isn’t a perfect fit, simply adapt it. They are guidelines, not rules.

#3