Closed bosonie closed 3 years ago
I have put my comments and suggestions for some changes in the manual in the branch 'ion-ag'.
Dear @albgar, I implemented your suggestions including the management of lj
projectors.
Please note that in the old .ion.xml format, there was no way to distinguish which j
was associated to the corresponding data. I assumed that the first listed j
is always the smaller j
. Meaning that the projector (l=1 j=0.5)
comes before the (l=1 j=1.5)
, the projector (l=2 j=1.5)
comes before the (l=2 j=2.5)
and so on. I hope it is correct.
If you could approve I then merge.
Please note that in the old .ion.xml format, there was no way to distinguish which
j
was associated to the corresponding data. I assumed that the first listedj
is always the smallerj
. Meaning that the projector(l=1 j=0.5)
comes before the(l=1 j=1.5)
, the projector(l=2 j=1.5)
comes before the(l=2 j=2.5)
and so on. I hope it is correct.
I have first checked the case in which Siesta reads a psf file and generates the projectors itself. In this case, "j-", that is, j=l-1/2, comes first.
When Siesta reads the projectors directly from a PSML file, regardless of the ordering in the file, Siesta always puts the "j-" projector first in the tables.
So in both cases j- comes first.
How do I approve????
Thanks a million! If you go in "Files changed", in the top right you have the "review changes" button!
Introduce the use of
IonData
in input. The implementation naturally leads to start to use validators. They will be heavily used in the coming refactoring. The use ofIonData
for theSiestaBaseWorkChain
has not been allowed yet. It is an unnecessary effort to implement it when it will come natural in the coming refactoring (we will expose the inputs ofSiestaCalculation
.