cloomis should this issue have the MCS label?

In the v7.0 NAOJ FITS bible, Sections 9.1 and 10.1 address OBS-MOD values and thus (as I understand it) possible sk directory names. Do we need to change the directory names to "IMAG" for some scripts and "SPEC" for others? Not sure I understand. Hisanori Furusawa?

I'm sorry I didn't noticed The book claim that you can put anything by putting "_" after allowed words.

Following it, I'd propose "SPEC_ENG". (PFS has only spectrograph mode, so I think we should keep SPEC to every images.)

cloomis Basically, yes. The abstract commands should be grouped separately for imaging and spectrograph modes so that the corresponding skeleton files are located under relevant directories that are named one of IMAGE(_*), SPEC(*), IPOL(*), and SPOL(_*). There are some exceptions with the existing instruments due to some historical reason (in some cases, the rule was simply overlooked unfortunately), and now Gen2 & FITS teams at the observatory are figuring out how to mitigate those violations on the Gen2 code side. This is an undesired exception handling, so we want to avoid this as much as possible with the active and future instruments.

Hisanori Furusawa , thank you for telling the background. Then, as long as we start with SPEC_  , ** is there no other convention to follow? (I thing yes, reading FITS bible ver 7.0.). I will change PFS kansoku scripts accordingly after our run this week.

To my understanding, there is no other mandatory convention, as long as a registered value is used for consistently the same meaning. There might be some offline discussion of a practical convention recently. I hope monodera could comment on this.

You can also put a special assignment in the header of an observation script to override the directory name so that the FITS.PFS.OBS-MOD will be set correctly.  We added that for some of the older instruments who had used "non-standard" directory names (e.g. "LAUNCHER", "ENG", etc.)

As long as the OBS-MOD starts with one of "IMAGE", "SPEC", "IPOL", and "SPOL" followed by "_", there is no particular convention. "SPEC_ENG" meets the regulation but it looks to me too generic as one cannot distinguish whether the data are for engineering for MCS, SM, other modules, or even combination of more than one. On the other hand, making different OBS-MOD for different types of engineering (e.g, SPEC_ENG_MCS, SPEC_ENG_SM1, etc) could be even more complicated.

For example, SMOKA lists some details of engineering observations (https://smoka.nao.ac.jp/ENG/ENGschedule.jsp). It would be helpful if PFS can do the same, while just using "SPEC_ENG" for OBS-MOD.

(For recording)
I have implemented a line "OBE_MODE=SPEC_ENG" to overwrite OBS-MOD status. 
We'll check the header when we can run Gen2 command on Gen2 simulator (or summit system) next time.

I tested the new commands during the engineering run03, but r was found the correct parameter is OBS_MODE, not OBE_MODE. Eric-san pointed out that if you run the command via launcher, overwriting function sometimes may not work.
So I'll modify the directory name to avoid any confusion..

> I tested the new commands during the engineering run03, but r was found the correct parameter is OBS_MODE, not OBE_MODE.

Just to prevent any confusion: the correct keyword is "OBS_MOD" not "OBS_MODE"