iraf.net

Hi,
I have a problem with SPLOT task. When I tried to do some work using splot I realized
something has changed. I used splot's continuum normalization
option for an echelle spectrum. When I wanted to overwrite the changes that I made in each order
onto the same file, it asked me for a new file name for each order every time. "overwrite" option
is set to "yes" in splot, so there is no problem there but it seems like this option is not doing its
job. Since splot asks me for a new file name every time for each order, and allows me to give the same
file name each time, at the end it actually creates a multi-extension file from which was a single-extension
file in the beginning.
I use IRAF V2.16 on Mac OSX (Mavericks) but the same problem occurred also on someone else's
machine that runs UBUNTU.
The weird thing is that there was no such problem couple weeks ago, everything was normal
with splot. I think IRAF does a self-update every time is activated (right?) and some library or task
update must have affected the SPLOT some how! Could you please help us to solve this problem?
How could we go back to previous SPLOT settings? Thanks.

Hi,

First, IRAF checks for updates but a) there hasn't been any updates in a couple of years and b) all it would tell you is that an update is available. It does not change anything in your system.

So the apparent behavior change is a clue about something different in the echelle files. SPLOT is a very old, though heavily used, tool which predates multiextension files. The overwrite feature was a challenge to implement and I consider it possible fragile in that it can fail to work as expected in some circumstances. I don't know what is happening. It might have something to do with IRAF environment variables, in particular use_new_imt and fkinit. You could show me the values with

cl\$this->_split2($m[0]) show use_new_imt fkinit

My advice is to try and not use SPLOT to write out multiple times in an echelle or multispec format. Instead you can internally do the normalize or other changes on all the orders and then write out once, possibly to a different file. You could also do the normalize with the separate task (I think it is CONTINUUM).

Yours,
Frank Valdes

Forgot to add a watch to this topic. Ignore this reply.