Improve `pinecard` metadata

[cschwan]

See the pending review on https://github.com/NNPDF/pinefarm/pull/23/files/177c81e1ac686f608970ba1cb79cad93e42f8520.

[felixhekhorn]

How you mean?

[cschwan]

Generate RUN_TEST_SH and extract pinecard. Have a look at launch.txt and you'll see this file:

launch @OUTPUT@
fixed_order = ON
set maxjetflavor 5
set gf @GF@
set mh @MH@
set mt @MT@
set mw @MW@
set mz @MZ@
set wh @WH@
set wt @WT@
set ww @WW@
set wz @WZ@
set ebeam1 3500
set ebeam2 3500
set pdlabel lhapdf
set lhaid @LHAPDF_ID@
set fixed_ren_scale True
set fixed_fac_scale True
set mur_ref_fixed @MZ@
set muf_ref_fixed @MZ@
set reweight_scale True
set ptl = 25.0
set etal = 2.5
set mll_sf = 116
#user_defined_cut set mmllmax = 1500.0
#user_defined_tau_min 116.0
#enable_patch change_etaj_to_rapj
set req_acc_FO 0.1
set pineappl True
done
quit

Without the actuall masses, widths and other constants this file isn't very helpful.

[felixhekhorn]

I see and I agree ... mmm this is not really easy to solve ... I guess we need to list and collect the necessary files dynamically ...

(maybe your review never got published?)

[cschwan]

I see and I agree ... mmm this is not really easy to solve ... I guess we need to list and collect the necessary files dynamically ...

Each interface could probably declare which files are taken as input, and then we'll take those.

[felixhekhorn]

Actually, you need the raw format (with set gf @GF@ etc.) to run pinefarm again, because the theory is coming only at creation time - so if we want to improve here, we would need to provide an additional key, which e.g. mg5 could fill with another base64 tar gz containing some files of interest