Polarized opcards

[giacomomagni]

This PR is to allow the computation of polarized Ekos.

As is it now the polarisation is triggered by the name of the DIS observable.
As this does not work for Hadronic datasets, we should refine it.

[Radonirinaunimi]

This looks good as long as we stick to the idea that any term that multiplies the polarised structure functions are accounted as $k$-factors (hence one single observable) $-$ which will always be the case unless we do a simultaneous fit. In addition to this, no polarised observable are given as a linear combination of (polarised) structure functions. So, this LGTM!

[giacomomagni]

This looks good as long as we stick to the idea that any term that multiplies the polarised structure functions are accounted as k-factors (hence one single observable) − which will always be the case unless we do a simultaneous fit.

This should not be a problem even in that case as the dataset will contain 2 Fk-tables so 2 Operator cards will be needed.

In addition to this, no polarised observable are given as a linear combination of (polarised) structure functions. So, this LGTM!

Any new polarised DIS observable not starting with g (or viceversa) could be a problem.

For hadronic datasets we can add and elif once such grids will be available.

[Radonirinaunimi]

This should not be a problem even in that case as the dataset will contain a 2 FKtables so 2 Operator cards will be needed.

Right! That makes sense.

Any new polarised DIS observable not starting with g (or viceversa) could be a problem.

Indeed! However, for now and for the far future I don't foresee this being the case.

For hadronic datasets we can add and elif once such grids will be available.

Agree, we should address this once these hadronic pineappl grids are available.

[Radonirinaunimi]

If this could be merged soon (followed by a release) that'd be great 🙃

[felixhekhorn]

That's the minimal change I was expecting 🙃

[felixhekhorn]

@Radonirinaunimi feel free to make immediately an release (else #99 will for sure trigger a release, but maybe you want to disentangle 😇 )

[Radonirinaunimi]

@Radonirinaunimi feel free to make immediately an release (else #99 will for sure trigger a release, but maybe you want to disentangle 😇 )

Thanks @felixhekhorn ! We surely want a release in which this is first disentangle from #99.