a few question and comments

[zhaozhiwen]

I am testing the new code since 9/22 commit

One thing needs to change everywhere is NOT assuming "/w/hallb-scshelf2102/clas12/rg-b/offline_monitoring/clas12-timeline/outfiles/" as default dir so that files from different run group won't be mixed together. maybe the default should be just "./"

There are too many options now to run things.
It would be really helpful to have a short and concise instruction.
for example, here is what I figure out
******* my current quickstart for chef ****************************
module load clas12/pro
../../bin/run-monitoring.sh -d v1 --submit --focus-detectors -o ./ --rundir ../pass0/v29.38.1/mon/recon/
../../bin/run-detectors-timelines.sh -i timeline_detectors -d v1
mv /w/hallb-scshelf2102/clas12/rg-b/offline_monitoring/clas12-timeline/outfiles ./
../../bin/deploy-timelines.sh -d v1 -i outfiles/v1/timeline_web -t rgb/pass0
++++++++++++++++++++++++++++++++++++++++++++++++++++++

I see a bug as follows
clas12-1@ifarm1901 v1> ../bin/run-detectors-timelines.sh -i timeline_detectors -d v1 -o ./
../bin/run-detectors-timelines.sh: illegal option -- o

I see many difference for timeline made before and after 9/22 code. why is it?
https://clas12mon.jlab.org/rgb/pass0/v29.38.1/tlsummary/ (before) (without run 11252 on purpose)
https://clas12mon.jlab.org/rgb/pass0/v29.38.1_new/tlsummary/ (after) (with run 11252 on purpose)

[c-dilks]

I see many difference for timeline made before and after 9/22 code. why is it?
https://clas12mon.jlab.org/rgb/pass0/v29.38.1/tlsummary/ (before) (without run 11252 on purpose)
https://clas12mon.jlab.org/rgb/pass0/v29.38.1_new/tlsummary/ (after) (with run 11252 on purpose)

Can you give more details on what differences you see? I checked 3 timelines and see no difference, other than the missing run 11252.

These are RGB data, which is unfortunately the one run group whose detector timelines deviate from the rest, so you need to use the -r b option until resolution of #74.

[c-dilks]

Here's one difference: #116

[c-dilks]

One thing needs to change everywhere is NOT assuming "/w/hallb-scshelf2102/clas12/rg-b/offline_monitoring/clas12-timeline/outfiles/" as default dir so that files from different run group won't be mixed together. maybe the default should be just "./"

This is a serious issue and I'll take care of it in a linked PR. Thanks for pointing it out.

[c-dilks]

All the extra options are primarily for developers and flexibility. Chefs will soon be re-trained to produce timelines in swif workflows, automatically producing them during usual cooking, so they should not need to worry about the extra options.

We're in the middle of this transition and will be testing this workflow on RGD data.

If you need to prioritize timeline production now, I recommend staying on v0.3 until things are more stable. The first version with swif integration will be v1.0, but likely will still have bugs until we've done more thorough testing.

[zhaozhiwen]

Chris. I missed the option for rgb.
I will keep using the old version for now.
It's great if timeline would be part of workflow.

[c-dilks]

Chris. I missed the option for rgb.

I think the old version required this too.

Anyway, thanks a lot for the feedback and testing the new version!

[c-dilks]

One thing needs to change everywhere is NOT assuming "/w/hallb-scshelf2102/clas12/rg-b/offline_monitoring/clas12-timeline/outfiles/" as default dir so that files from different run group won't be mixed together. maybe the default should be just "./"

Fixed by #122

I see a bug as follows
clas12-1@ifarm1901 v1> ../bin/run-detectors-timelines.sh -i timeline_detectors -d v1 -o ./
../bin/run-detectors-timelines.sh: illegal option -- o

Fixed by #120