GitLab

TestLHE commented out for now

See merge request !46379

ART tests - correct to reduce the time needed for running

See merge request !46377

add hdamp steering params for Powheg ttbb module

See merge request !46372

Sweeping !46078 from master to 21.6.
Fix possible bug in modify_param_card

See merge request !46112

MadGraph 2.95

See merge request !46346

Johannes Elmsheuser authored Aug 26, 2021 and Atlas Nightlybuild committed Aug 27, 2021

Fix possible bug in modify_param_card

See merge request !46078

(cherry picked from commit 3489b821)

e1324fe4 Update Generators/MadGraphControl/python/MadGraphUtils.py
62b86ad9 Fix leftovers bug

Update systematics tool

See merge request !46080

MadGraph: switch to version 2.9.3.atlas2

See merge request atlas/athena!45916

MadGraph update to 3.1.1.atlas

See merge request !45852

Handling of log messages: AGENE-2016, AGENE-2024; Adding ttj MiNNLO: AGENE-2016

See merge request !45670

PowhegControl: fixing reweighting issues

See merge request !45648

Removing duplicated properties; cleaned SuppressSmallPT code

See merge request !45611

Pythia8.245p3lhcb7

See merge request !45570

improve PowhegControl VBF W,Z interface AGENE-1831

See merge request !45492

PowhegControl: fix reweighting issue and expose parallelstage in some processes

See merge request !45526

PowhegControl: fix reweighting issue due to #pdf comments, and expose parallelstage option in remaining V2 processes

Add four-top filter

See merge request !45501

patch tau filters for decay-less Sherpa status code 20 taus

See merge request !45453

Pythia8 version 305

See merge request !45437

Setting the default sde_strategy in MadGraphControl

See merge request !45406

In versions of MadGraph newer than 2.9, there's an sde_strategy run
parameter that defines how the job does the integration. It's a change
of strategy that may behave better under some circumstances. We have
seen in the validation, AGENE-2038, that this strategy is automatically
set based on the process being used. This seems confusing, and the
automatic setting doesn't always align with what is naively expected.

After some discussion, we decided to always set the default strategy,
and of course to allow the user to change the strategy if they choose to
do so. This MR updates the run card as soon as a new process is defined
to ensure that the default strategy is used. If the user chooses to set
the new strategy they'll be able to do so, and their setting will not be
overridden.

I took the opportunity to clean up a couple other cases where the run
card was being modified and we were being overlay prescriptive about
what to set.