Initial lbexec support

When reviewing this merge request the individual commits likely make the most sense. In summary:

• 6843d61b Adds the two needed parts for lbexec, more specifically:
  • A DaVinci.Options class which is used internally by lbexec to parse the YAML
  • A DaVinci.make_config function which is the nominal way people should be using DaVinci
• 931f5149 makes it so that the global DaVinci.options object is never relied upon (as it will be removed once the transition to lbexec is complete)
• 7d2f007b contains an example of porting to lbexec with the test_davinci_tupling_All test. If everyone is happy with it I'll port the remaining tests and start cleaning up.

To see how this might actually look in reality, make a yaml file named options.yaml containing (most of these keys probably aren't needed):

input_files:
    - root://eoslhcb.cern.ch//eos/lhcb/wg/dpa/wp3/tests/spruce_all_lines_realtimereco_newPacking.dst'
annsvc_config: 'root://eoslhcb.cern.ch//eos/lhcb/wg/dpa/wp3/tests/spruce_all_lines_realtime_newPacking.tck.json'
data_type: Upgrade
input_type: ROOT
simulation: true
conddb_tag: sim-20171127-vc-md100
dddb_tag: dddb-20171126
histo_file: 'sprucing_histos.root'
input_raw_format: 0.3
lumi: false
ntuple_file: 'sprucing_tuple.root'
print_freq: 1
process: 'Spruce'
stream: 'default'

You can then run davinci using: lbexec DaVinciExamples.tupling.AllFunctors:alg_config options.yaml

I think it would be good to sit down and discuss what should be stored in the options.yaml and what the arguments to DaVinci.make_config should be. In particular I think things like process/stream/lumi/evt_pre_filters might be worth moving to the Python side.

Work towards DPA grand task https://gitlab.cern.ch/lhcb-dpa/project/-/issues/180 on configuration. Necessary for lhcb-dpa/analysis-productions/LbAnalysisProductions#68 (closed).

DaVinciExamples/python/DaVinciExamples/tupling/AllFunctors.py

@@ -10,9 +10,6 @@
 ###############################################################################
 """
 Example of a DaVinci job filling all available functors. This is obviously a stress test and not realistic.
-
- This example is meant to be run with
-    $ ./run davinci --inputfiledb Spruce_all_lines_dst Phys/DaVinci/options/DaVinciDB-Example.yaml --joboptfile DaVinciExamples/python/DaVinciExamples/tupling/option_davinci_tupling_from_spruce.yaml --user_algorithms  DaVinciExamples/python/DaVinciExamples/tupling/example-tupling-AllFunctors:alg_config --evt_max 100 |& cat | tee log
 """
 __author__ = "P. Koppenburg"
 __date__ = "2021-11-23"
@@ -27,7 +24,7 @@ from DecayTreeFitter import DTFAlg
 from DaVinci.truth_matching import configured_MCTruthAndBkgCatAlg
 from PyConf.Algorithms import PrintDecayTree
 
-from DaVinci import options
+from DaVinci import make_config
 
 #
 # Definition of strucing line
@@ -39,9 +36,6 @@ _basic = 'basic'
 _composite = 'composite'
 _toplevel = 'toplevel'
 
-options.ntuple_file = "DV_example_allFunctors_ntp.root"
-options.histo_file = "DV_example_allFunctors_his.root"
-
 
 def all_variables(pvs, DTFR, mctruth, ptype):
     """
@@ -255,7 +249,7 @@ def event_variables(PVs, ODIN, decreports):
     return evt_vars
 
 
-def alg_config():
+def alg_config(options):
     """
     Algorithm configuration function called from the comad line
     """
@@ -312,7 +306,8 @@ def alg_config():
     #
     # Sprucing filter
     #
-    my_filter = add_filter("HDRFilter_B0DsK", f"HLT_PASS('{bd2dsk_line}')")
+    my_filter = add_filter(options, "HDRFilter_B0DsK",
+                           f"HLT_PASS('{bd2dsk_line}')")
 
     #
     # FunTuple
@@ -329,6 +324,5 @@ def alg_config():
     #
     # Algorithms to be run
     #
-    return {
-        "UserAlgs": [PrintDecayTree(Input=bd2dsk_data), my_filter, my_tuple]
-    }, []
+    return make_config(
+        options, [PrintDecayTree(Input=bd2dsk_data), my_filter, my_tuple])