WBPP 2.0 released

sbenz

Member
Jul 8, 2017
16
0
Hello Robyx,

sorry to bother you again, but probably you overlooked my message from Sunday.

I wonder why there are no MasterDark files created, I found only the MasterFlat and of course the MasterLight in the output folder (20 darks were provided and used by the script). Is this something I did wrong or something that I overlooked?

Thanks,
Stefan
 

rppass

Member
Feb 8, 2015
7
0
I am a bit confused by the new wpbb 2.0. Prior to the release I had master files for bias, dark, and flats and the prior scripts recognized and used them. Now, 2.0 does not recognize them. They have FITS header values for imagetyp (like MASTER BIAS, MASTER DARK, and MASTER FLAT). I thought I read the announcement that this was sufficient to get them recognized as masters. The associated file names are masterbais..... etc, What am I missing?
 

ngc1535

PixInsight Ambassador
Feb 1, 2014
751
133
AdamBlockStudios.com
Hi Rppass,

Some screenshots to illustrate the issue would help. What you describe is true if you create the masters in WBPP it does insert "Master Xxx" into the IMGTYP keyword as a value. Can you show the WBPP panel were it is identifying any files and a screenshot of the FITs header showing the what is written as an entry for IMGTYP ? Including "master" in the file or folder name also does the trick.

Note the VALUEs are case sensitive. You literally wrote "MASTER XXX" with all capitalization..but if the master files are being discovered in the FITs header..I think it needs to read "Master Xxx" as shown in the screen capture.

-adam
 

Attachments

robyx

PixInsight Staff
Jul 17, 2019
159
40
44
Zurich
I am a bit confused by the new wpbb 2.0. Prior to the release I had master files for bias, dark, and flats and the prior scripts recognized and used them. Now, 2.0 does not recognize them. They have FITS header values for imagetyp (like MASTER BIAS, MASTER DARK, and MASTER FLAT). I thought I read the announcement that this was sufficient to get them recognized as masters. The associated file names are masterbais..... etc, What am I missing?
It should absolutely be recognized as master, considering that even if in FITS header the information is not recognized the term "master" is in the file name. So I think the issue shouldn't regard the fits header content, neither regards the file name which contains "master" as expected for a master....

Please follow @ngc1535 suggestion and post some screenshots :)

Note the VALUEs are case sensitive.
@ngc1535 just one note, the value is NOT case sensitive, to detect a dark frame the value it could be one of the following variations:

"dark frame"
"dark"
"master dark"
"flatdark"
"flat dark"
"darkflat"
"dark flat"

To detect if the file is a master the word "master" NON case-sensitive should be in the IMGAETYP.
 

ngc1535

PixInsight Ambassador
Feb 1, 2014
751
133
AdamBlockStudios.com
It should absolutely be recognized as master, considering that even if in FITS header the information is not recognized the term "master" is in the file name. So I think the issue shouldn't regard the fits header content, neither regards the file name which contains "master" as expected for a master....

Please follow @ngc1535 suggestion and post some screenshots :)


@ngc1535 just one note, the value is NOT case sensitive, to detect a dark frame the value it could be one of the following variations:

"dark frame"
"dark"
"master dark"
"flatdark"
"flat dark"
"darkflat"
"dark flat"

To detect if the file is a master the word "master" NON case-sensitive should be in the IMGAETYP.
AH! OK. We should have a list of "reserved" (special) words probably listed somewhere.
-adam
 

rppass

Member
Feb 8, 2015
7
0
I solved my problem (not recognizing master bias, darks, and flats). The problem was not case sensitivity, but rather there were two two FITS header cards with the same key word: IMAGETYP. The first had a value such as 'Master Flat' and the second had a value such as "Flat Field'. Apparently PixInsight uses the last card with a particular key word.

I think this implies that when PixInsight creates a Master (bias, dark, or flat) it puts an FITS card with the key word IMAGETYP in the FITS header at the head of the list. This means that if there is another IMAGETYP card later in the FITS header PixInsight will get the 'wrong' current IMAGETYP.
 

sbenz

Member
Jul 8, 2017
16
0
it's always stored in the "logs" subfolder that you find inside the output folder :)
Ah, my mistake, I had forgotten to enable logging... :sneaky:

Also, the fact the no master darks were created is also my mistake: For some reason the dark frames I used have a different pixel size than the lights, don't ask me why, same camera, same mode...
 

naavis

Well-known member
May 31, 2012
70
0
Helsinki, Finland
I'm getting a weird star alignment error in the case where someof my groups have only one image. The error says:
Code:
[2021-03-03 22:10:11] ** Warning: Star alignment failed to register 0 images out of 1. A minimum of 3 images must be succesfully registered.
The registration reference image is set to Auto. Selecting it manually did not make a difference. I understand stacking less than 3 images makes little sense, but that shouldn't affect registration, right? I have more than 3 images per filter in the whole process, but over several nights where one night has just one or two images.

Removing the small groups lets the script run through without errors.

The full log is attached in the message.
 

Attachments

Last edited:

aatdalt

Member
Feb 24, 2020
6
0
Something myself and a few others are trying to figure through is how to navigate using Date_ as a keyword. ANSI (and regular humans) date standard format uses hyphens (2021-3-6) Which breaks WBPP's keywording. Date_2021-3-6 gets interpreted as Keyword: Date, Value: 2021.

Another situation that was just raised as an issue with a hyphen separator is temperatures. I think it's quite common for people to record temps as: TEMP_-20C or something similar which would break the keywording as well, as far as I know.
 
Last edited:

clabille

Member
Aug 23, 2019
5
0
but have you successfully run the update and restarted PI before trying this?

rob
Rob, Yes, I ran the update and restarted PI and my computer several times. I've now tried to reinstall the update, repair the update, shut off all of my AV software, and then try the regenerate again and still no luck.
 

clabille

Member
Aug 23, 2019
5
0
Rob, Yes, I ran the update and restarted PI and my computer several times. I've now tried to reinstall the update, repair the update, shut off all of my AV software, and then try the regenerate again and still no luck.
Rob. I just figured it out after uninstalling and reinstalling several times. The "darkarchon" mod I tried was blocking updates for some reason. I uninstalled darkarghon, ran an update, and it loaded perfectly.
 

pfile

PTeam Member
Nov 23, 2009
6,339
208
Rob. I just figured it out after uninstalling and reinstalling several times. The "darkarchon" mod I tried was blocking updates for some reason. I uninstalled darkarghon, ran an update, and it loaded perfectly.
interesting, i wonder if there was a syntax error in the repository URL. also darkarchon's site is i think his own computer at home and uses dynamic DNS to establish the name mapping and this seems to frequently be broken.

rob
 

robyx

PixInsight Staff
Jul 17, 2019
159
40
44
Zurich
@robyx p.s. see this thread: https://pixinsight.com/forum/index.php?threads/master.16264/

do you think you could put logic into WBPP to exclude volume names which include the substring "master"?

rob
Hi rob,

yep I saw it. I think a good way to address these unfortunate cases could be to add a global checkbox to optionally exclude the entire file path from playing any rule to detect master files. Addressing the volume name only solves the issue for that case only but I'm quite sure other users could have other specific issues depending on the naming conventions they have for they libraries or projects.

In case of file path conflicts, they can disable the entire file path, at that point the word "master" plays a role only for file name and I think it's reasonable to assume that a single bias/dark/flat file does not contain the word master.

To be addressed in the next release I'm already working on.

Thanks for reporting!
Robyx
 
  • Like
Reactions: pfile

sbenz

Member
Jul 8, 2017
16
0
Hello Robyx,

thanks again for the great update to the script!

One thing that would help me tremendously:
Would it be possible to add to the status box that appears after the script ran an overall success message if everything went well?
Something like "WPP successfully executed" or "There were errors, please check the messages above"

Currently I need to check all lines to makes sure that everything worked OK, and indeed I once overlooked that my darks could not be processed and I wondered why there was no master dark, but the rest of the script went fine, so there was a master light...

Thanks
Stefan
 

robyx

PixInsight Staff
Jul 17, 2019
159
40
44
Zurich
Hi @sbenz, sure this makes sense. I will integrate this info in the next release.

Thanks for the suggestion!

Robyx