Image numbers should be a unique sequence for each basename

[cfremling]

imnum should update when basename is changed. currently not happening until next observation starts

[astronomerdave]

@cfremling Can you provide more details? I'm not sure what you mean by imnum is not updating. Are you saying that you changed the imnum and it didn't change?

[cfremling]

lets say we have 3 images ngps_20250101_001-3.fits if you check what imnum will be after doing a focus sequence taking 12 images on a different base name, then changing the base name back to ngps, the next image will be reported to be 13 and not 4 like it should be. the system believes its taking more focus images until you actually go and take an image. this has the potential to confuse anything else relying on imnum

[astronomerdave]

In that case I would say the issue is not that imnum is not updating. It is updating. But what you would like is to have a unique imnum series for each basename, correct?

[cfremling]

I believe that would work fine. I add Matt to this email since he was the one to point this out in the first place Christoffer On Jan 26, 2026, at 11:13 AM, David Hale ***@***.***> wrote: [https://avatars.githubusercontent.com/u/3028392?s=20&v=4]astronomerdave left a comment (CaltechOpticalObservatories/NGPS#379)<#379 (comment)> In that case I would say the issue is not that imnum is not updating. It is updating. But what you would like is to have a unique imnum series for each basename, correct? — Reply to this email directly, view it on GitHub<#379 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ABEKDGLIIIV5IBX2QGSTD3D4IZRL3AVCNFSM6AAAAACS27EYPWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTQMBRGI4DGOBVGI>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[astronomerdave]

Renamed issue.
New feature request to let each unique basename have a unique image number sequence.
For example,

ngps_0001.fits
ngps_0002.fits
ngps_0003.fits
test_0001.fits
test_0002.fits
ngps_0004.fits
:     :     :

instead of currently, where imnum is the image number for all images on a given day:

ngps_0001.fit
ngps_0002.fits
ngps_0003.fits
test_0004.fits
test_0005.fits
ngps_0006.fits
:     :     :

[MateuszKM]

I am not sure this is all clear to me.
let's say we start in a new directory

This is the current behavior, I think.

camera basename ngps
... take 3 images (0,1,2)
camera imnum
3
camera basename newngps
camera imnum (** should return 0)
3
.. take two images (writes newngps...0, newngps...1)
camera imnum (## should return 2)
2
camera basename ngps
camera imnum ($$ should return 3)
2
... take an image (writes ngps...3)
camera imnum (%% should return 4)
4

---

In other words, when you change the basename, imnum still returns the number it would have written next for the old basename, rather than the number it will write for the new basename.

Dave's proposed solution of just incrementing the imnum regardless of the basename will be OK. Making it so the next imnum to be written gets updated when the basename is changed would be OK too.