Spectra wavelength array

[dr-rodriguez]

Some spectra do not get properly processed by astrodbkit, namely some reductions of the FIRE data. The wavelength is logarithmic, but does not get constructed that way. I've added a temporary fix (dr-rodriguez@3131f76) but additional testing needs to be done to see the magnitude of the problem and confirm this fix does not affect other spectra.

[hover2pi]

David,

I added a fix for this a while back (I think). If you prepend 'log' to
either the flux or wavelength units, it converts it to linear scale in the
Spectrum() class.

I.e. in the database, make the wavelength_units = 'log A' and/or the
flux_units='log ergs-1cm-2A-1' and then the database query will return the
linear spectrum.

In the problematic spectra, do they have 'log' units or units at all?

• Joe

On Mon, Nov 14, 2016 at 3:57 PM, David Rodriguez notifications@github.com
wrote:

Some spectra do not get properly processed by astrodbkit, namely some
reductions of the FIRE data. The wavelength is logarithmic, but does not
get constructed that way. I've added a temporary fix (dr-rodriguez/astrodbkit@3131f76
dr-rodriguez@3131f76)
but additional testing needs to be done to see the magnitude of the problem
and confirm this fix does not affect other spectra.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#80, or mute the thread
https://github.com/notifications/unsubscribe-auth/AACHf515a2cNHxkVA6PwU7koLmnuAFqxks5q-Ms6gaJpZM4KxzSK
.

[dr-rodriguez]

The convert_spectrum() function does not check the database for the units, it only takes the file path. I see that it checks the header to see if CTYPE1=='LINEAR' but it doesn't do anything with this information.