Skip to content

Add support for config-driven channel overrides :gz and :null#1069

Open
m3d wants to merge 3 commits into
masterfrom
feature/config-driven-modifiers-gz-null
Open

Add support for config-driven channel overrides :gz and :null#1069
m3d wants to merge 3 commits into
masterfrom
feature/config-driven-modifiers-gz-null

Conversation

@m3d

@m3d m3d commented Jun 22, 2026

Copy link
Copy Markdown
Member

Accepted is also ':' which would cancel hard-coded modifiers in the code.

(as described in the email, but maybe code is more clear)

Accepted is also ':' which would cancel hard-coded modifiers in the code.
@m3d m3d requested review from tajgr and zbynekwinkler June 22, 2026 17:50
@m3d

m3d commented Jul 2, 2026

Copy link
Copy Markdown
Member Author

@tajgr can se set some milestone/deadline for go or not to go? My motivation is to release OSGAR 1.1 and this could be new feature (while in case of problems users still can fallback to 1.0.0)
thanks m.

@tajgr

tajgr commented Jul 2, 2026

Copy link
Copy Markdown
Collaborator

Ok, that takes sense. This week?
How about standardizing that drivers always register streams without the :suffix? :gz and :nul should always be set in the configuration. Among other, oak_camera has compress_depth_stream parametr.
While compression saves disk space, it also consumes CPU time.

This feature should be also added to the documentations.

@m3d

m3d commented Jul 6, 2026

Copy link
Copy Markdown
Member Author

Ok, that takes sense. This week? How about standardizing that drivers always register streams without the :suffix? :gz and :nul should always be set in the configuration. Among other, oak_camera has compress_depth_stream parametr. While compression saves disk space, it also consumes CPU time.

This feature should be also added to the documentations.

Take your time - this or next week is fine, but I would like to close it before "Dobyvani hradu 2026".

Regarding parameters - in my opinion the driver itself should provide the best reasonable parameters. For example to publish depth without gzip is total non-sense, and I would like to avoid cases when "I forgot". So hard-coded would be recommended parameters, which you can override in JSON. You are right, that I should remove also the duplicity in some drivers where the gzip is hacked with extra (non-standard) way, which I am trying to remove by this PR.

Regarding documentation, Elena added some notes - let me know if you would like to see more.
thanks m.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants