add option to explicitly by-pass check for identically configured components with different names

Allow opt-out of checking for identical configured components with different names.

This will allow eg. multiple lines to have the identically configured up-front filters with their own distinct and dedicated name -- which will introduce a bit of overhead, but it makes things logically independent (and as they were in Run1 + Run2) and more stable under changes in configuration (i.e. the configuration of one line will not change just because another line with which it spuriously shared some component instance changed). Note that these clashes are typically 'spurious', i.e. there is just a large likelihood of multiple unrelated lines filtering on the same set of lines at a previous stage, without this being intentional, and without this having to be the case 'by construction'.

Note that the default behaviour remains to not allow this, and one has to explicitly 'opt-out' of this check.

Goes together with Moore!2336 (merged)

requested review from @rmatev

changed title from explicitly add flag to by-pass check for identically configured components with different names to add option to explicitly by-pass check for identically configured components with different names

mentioned in merge request Moore!2336 (merged)

added RTA label

added 1 commit

• 307814e5 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

changed the description

• [2023-05-23 15:06] Validation started with lhcb-master-mr#7987
• [2023-05-23 22:51] Validation started with lhcb-master-mr#7998
• [2023-05-24 12:05] Validation started with lhcb-master-mr#8002
• [2023-05-24 18:13] Validation started with lhcb-master-mr#8004

added 5 commits

• 307814e5...fb8ab3e8 - 4 commits from branch master
• 53338385 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

I would like to have a pytest for this functionality. (we tried to have a comprehensive set of unit tests for PyConf, which helps a lot when touching it for any reason, and I'd like to preserve this)

approved this merge request

added 1 commit

• 2bf05f9f - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• eae98e1b - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• 41d2eca6 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• 564d17c2 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• c5d79d69 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• e5044411 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• 80b97e9a - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• 30a4a12d - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

mentioned in merge request Moore!1605 (merged)

@rmatev: you may be interested in this discussion which directly pertains to this functionality. I am actually not sure why the clash between make_merged_neutral_pions and make_merged_pi0s has not been causing problems already...

added 1 commit

• fe8aede0 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added 1 commit

• 97dbf2ad - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

/ci-test Moore!2336 (merged)

added ci-test-triggered label

• [2023-05-26 18:17] Validation started with lhcb-master-mr#8044
• [2023-05-26 18:17] Validation started with lhcb-master-mr#8045
• [2023-05-30 20:12] Validation started with lhcb-master-mr#8071
• [2023-05-31 11:39] Validation started with lhcb-master-mr#8077
• [2023-06-01 15:22] Validation started with lhcb-master-mr#8106
• [2023-06-01 21:32] Validation started with lhcb-master-mr#8118

added 3 commits

• 97dbf2ad...676abf33 - 2 commits from branch master
• 8b9f2127 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

/ci-test Moore!2336 (merged)

added 5 commits

• 8b9f2127...fac15526 - 4 commits from branch master
• 4a8c3eeb - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

/ci-test Moore!2336 (merged)

added 5 commits

• 4a8c3eeb...9b4e8dcd - 4 commits from branch master
• 5644d0d4 - explicitly add flag to by-pass check for identically configured components with different names

Compare with previous version

added needs ref update label

Is this ready for the next round of test ?

assigned to @sponce

mentioned in merge request !4116 (merged)

resolved all threads

mentioned in commit Allen@2685b052

mentioned in commit Moore@3074db90

mentioned in commit Panoptes@17cf0300

mentioned in commit Boole@a0b54de5

mentioned in merge request Alignment!381 (merged)

mentioned in merge request Panoptes!277 (closed)

mentioned in commit Alignment@07e95f59

mentioned in merge request Moore!2386 (merged)

mentioned in merge request Boole!478 (closed)

mentioned in merge request Allen!1229 (closed)

mentioned in commit Moore@c45e22f5

• [2023-06-02 15:12] Validation started with lhcb-master-mr#8128

added hlt1-throughput-decreased label

merged

mentioned in commit f62b8355

mentioned in commit Moore@7946a307

mentioned in commit Alignment@c286be84

mentioned in commit Moore@efc4f95f

mentioned in commit Moore@47563ec2

mentioned in issue Moore#598 (closed)

mentioned in merge request !4191 (merged)