Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Schema based configuration #105

Draft
wants to merge 140 commits into
base: master
Choose a base branch
from
Draft

Conversation

brettviren
Copy link
Member

This has the start of a more formal handling of configuration using schema and code generation to address some long-standing design omissions.

See the util/schema/README.org for more motivations and details.

In principle, we can merge this any time and then make adiabatic progress by converting each component one-by-one. But, before doing that, I think the rest of the team needs to be okay with it. So, we can keep this as a "draft PR" for a while.

As a start, @HaiwangYu please take a look and we can discuss what next steps are best.

@HaiwangYu
Copy link
Member

Thanks @brettviren for the nice documentation and example. Looks really great! I will try it out and get some experience for discussion tomorrow.

brettviren and others added 13 commits June 24, 2021 16:20
Refactor AnodePlane, only 44 more to go...
May use C-bindings to Go jsonnet via './wcb --with-jsonnet-libs=gojsonnet [...]'
Drop C++ jsonnet API and use plain C (or Go + C-bindings)
Fix #109.  Adds back multithreading.
This has been with us for a while but would be rare to hit.

- job must run TbbFlow instead of Pgrapher
- just have more than one AnodePlane
- two threads must both request spectrum() for the same wire
- at the same time
- and both be the first time requested

There are still other races of this type though maybe all in "service"
type components that are only used from one "node" component.
brettviren and others added 14 commits October 18, 2021 13:22
…on HDF5 to remove illegal cross dependency on hio
… the anode plane

Note, this is a broken concept in general and will only work accidentally with some detectors.
There is a stand-alone PlaneSelector to do this and DNNROIFinding
changes from cbeg/cend to using the same pattern internally.

This is based on a channel selection from OmnibusSigProc that is moved
into a new Aux/PlaneTools.h.  But, OmnibusSigProc still is left as-is.
It should be factored to use this moved code.
Prepare for hybrid full chain with wc/ls stage separate from follow-on pure-WCT stage
@brettviren brettviren mentioned this pull request Dec 24, 2023
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.

4 participants