REF: db.json can be relative since happi v1.14.0 (Jul 2022)

[klauer]

db.json can be relative to happi.cfg since happi v1.14.0 (Jul 2022)

• This PR makes it relative
• This PR makes it easier for non-PCDS machines to use the happi database

[tangkong]

While I sometimes used the happi cfg to remember what the /reg/g/pcds... path is, that's a me problem. I like this 👍

[ZLLentz]

This was deployed in pcds-5.4.2
Here are the currently deployed hutch python versions (* = too old for this to work)

cxi/cxienv:export CONDA_ENVNAME="pcds-5.2.1" (*)
det/detenv:export CONDA_ENVNAME="pcds-3.3.0" (*)
icl/iclenv:export CONDA_ENVNAME="pcds-4.2.0" (*)
las/lasenv:export CONDA_ENVNAME="pcds-5.3.1" (*)
mec/mecenv:export CONDA_ENVNAME="pcds-5.2.0" (*)
mfx/mfxenv:export CONDA_ENVNAME="pcds-5.5.1"
rix/rixenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
tmo/tmoenv:export CONDA_ENVNAME="pcds-5.5.1"
tst/tstenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
ued/uedenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
xcs/xcsenv:export CONDA_ENVNAME="pcds-5.4.1" (*)
xpp/xppenv:export CONDA_ENVNAME="pcds-5.5.0"

I think this is good to go for master branch but we need to wait to drop it into the deploy branch

[klauer]

I didn't realize those were so far behind. Back to draft this goes.

[ZLLentz]

It should be fine as long as we wait on merging it into the deploy branch

[klauer]

I'd rather not have something in master that we have to be super careful about merging back in. It's better to leave it here as a visible indicator of "we want to do this but we're waiting for this reason..." in my opinion

[klauer]

As of today, who is still below pcds-5.4.2?

cxi/cxienv:export CONDA_ENVNAME="pcds-5.7.2"
det/detenv:export CONDA_ENVNAME="pcds-3.3.0" (*)
icl/iclenv:export CONDA_ENVNAME="pcds-4.2.0" (*)
las/lasenv:export CONDA_ENVNAME="pcds-5.3.1" (*)
mec/mecenv:export CONDA_ENVNAME="pcds-5.2.0" (*)
mfx/mfxenv:export CONDA_ENVNAME="pcds-5.7.2"
rix/rixenv:export CONDA_ENVNAME="pcds-5.7.2"
tmo/tmoenv:export CONDA_ENVNAME="pcds-5.7.1"
tst/tstenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
ued/uedenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
xcs/xcsenv:export CONDA_ENVNAME="pcds-5.7.3"
xpp/xppenv:export CONDA_ENVNAME="pcds-5.7.2"

That is: det, icl, las, mec, tst, ued
We're better off than we were!
We should:

• Ping the laser controls team about upgrading
• Ping MEC about upgrading
• UED has no happi devices that I can see - I have a feeling it can be ignored
• TST, ICL, DET - can we just upgrade these to latest?

stale

[klauer]

$ /cds/group/pcds/pyps/apps/hutch-python
$ grep '^export CONDA_ENVNAME' */*env |sort
cxi/cxienv:export CONDA_ENVNAME="pcds-5.8.0"
det/detenv:export CONDA_ENVNAME="pcds-3.3.0" (*)
icl/iclenv:export CONDA_ENVNAME="pcds-4.2.0" (*)
las/lasenv:export CONDA_ENVNAME="pcds-5.3.1" (*)
mec/mecenv:export CONDA_ENVNAME="pcds-5.2.0" (*)
mfx/mfxenv:export CONDA_ENVNAME="pcds-5.7.2"
rix/rixenv:export CONDA_ENVNAME="pcds-5.7.2"
tmo/tmoenv:export CONDA_ENVNAME="pcds-5.7.1"
tst/tstenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
ued/uedenv:export CONDA_ENVNAME="pcds-5.3.0" (*)
xcs/xcsenv:export CONDA_ENVNAME="pcds-5.7.3"
xpp/xppenv:export CONDA_ENVNAME="pcds-5.7.2"

As of today, the story has not changed. We should follow up with the actions from my last comment, I think.

[tangkong]

This entirely slipped my mind. I'll shoot out some slack messages