Core: expose ID.session_uid in RNA #116888

Jacques Lucke · 2024-01-08T12:55:57+01:00

Jacques Lucke commented

2024-01-08 12:55:57 +01:00

This exposes the internal ID.session_uuid in the Python API as ID.session_id. The exposed name is not session_uuid, because from what I gathered people would prefer to rather rename the internal name because it is universally unique.

The reason for exposing this is to allow Python scripts to call operators that take the session id as input. A fair number of operators do this as you can see when searching for WM_operator_properties_id_lookup.

I don't see a strong reason against exposing this as read-only property, but maybe I'm missing something.

This exposes the internal `ID.session_uuid` in the Python API as `ID.session_id`. The exposed name is not `session_uuid`, because from what I gathered people would prefer to rather rename the internal name because it is universally unique. The reason for exposing this is to allow Python scripts to call operators that take the session id as input. A fair number of operators do this as you can see when searching for `WM_operator_properties_id_lookup`. I don't see a strong reason against exposing this as read-only property, but maybe I'm missing something.

Jacques Lucke added 1 commit 2024-01-08 12:56:08 +01:00

de0b2081a3 expose session id

Jacques Lucke requested review from Brecht Van Lommel 2024-01-08 12:56:25 +01:00

Jacques Lucke requested review from Bastien Montagne 2024-01-08 12:56:25 +01:00

Bastien Montagne requested changes 2024-01-08 13:28:56 +01:00

Bastien Montagne left a comment

I could have sworn that this was already exposed to RNA...

Indeed see no reasons to not do it, as long as it remain read-only, it can also help in some debugging cases.

Would rather use the name session_uid though, for two reasons:

id is already used in our codebase and API in for something completely different (the data-blocks).
This session id is assumed to be unique during that session (at least within 'reasonable' situations), so using uid makes sense to me?

Think it's also time to rename our C++ code to use the matching name too, unless someone strongly object would do that in coming days?

I could have sworn that this was already exposed to RNA... Indeed see no reasons to not do it, as long as it remain read-only, it can also help in some debugging cases. Would rather use the name `session_uid` though, for two reasons: * `id` is already used in our codebase and API in for something completely different (the data-blocks). * This session id is assumed to be unique during that session (at least within 'reasonable' situations), so using `uid` makes sense to me? Think it's also time to rename our C++ code to use the matching name too, unless someone strongly object would do that in coming days?

Jacques Lucke added 1 commit 2024-01-08 13:43:48 +01:00

eb38197b73 rename to `session_uid`

Jacques Lucke commented

2024-01-08 13:46:07 +01:00

Using session_uid is fine with me.
Renaming the internal name soon is a good idea too, especially since the term "session_uuid" is already exposed as parameter name of some operators. Now, it's likely still possible to change this without breaking any scripts, because the session uid was not exposed.

Using `session_uid` is fine with me. Renaming the internal name soon is a good idea too, especially since the term "session_uuid" is already exposed as parameter name of some operators. Now, it's likely still possible to change this without breaking any scripts, because the session uid was not exposed.

👍 1

Bastien Montagne approved these changes 2024-01-08 14:31:02 +01:00

Jacques Lucke changed title from ~~Core: expose ID.session_id in RNA~~ to Core: expose ID.session_uid in RNA

2024-01-08 15:31:04 +01:00

Jacques Lucke referenced this issue from a commit

2024-01-08 15:34:20 +01:00

Core: expose ID.session_uid in RNA

Jacques Lucke merged commit c68b22cfdf into main

2024-01-08 15:34:20 +01:00

Jacques Lucke deleted branch expose-session-id

2024-01-08 15:34:22 +01:00

Kazashi Yoshioka referenced this issue from a commit

2024-01-12 19:10:52 +01:00

Core: expose ID.session_uid in RNA

Jacques Lucke referenced this pull request

2024-01-19 21:11:22 +01:00

Core: rename Session UUID to Session UID #117350

Arye Ramaty referenced this issue from a commit

2024-01-21 15:37:38 +01:00

Core: expose ID.session_uid in RNA

Jacques Lucke referenced this issue from a commit

2024-01-22 13:47:24 +01:00

Core: rename Session UUID to Session UID

Tom Edwards commented

2024-01-27 15:22:09 +01:00

First-time contributor

Thanks for making this addition! Is there a way to retrieve a datablock Python object if I have its session_uid, other than iterating through all of them?

Thanks for making this addition! Is there a way to retrieve a datablock Python object if I have its `session_uid`, other than iterating through all of them?

Jacques Lucke commented

2024-01-28 11:28:59 +01:00

@artfunkel no that's currently not possible. Even internally we have to iterate through all of them currently. If you need to look up many data blocks by session uid, you can create temporary dict of course.

@artfunkel no that's currently not possible. Even internally we have to iterate through all of them currently. If you need to look up many data blocks by session uid, you can create *temporary* `dict` of course.

Sign in to join this conversation.

No reviewers

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

Core: expose ID.session_uid in RNA #116888