Code

Singletons setting unit tests fix#442

Code

Merged

scijones merged 37 commits intodevelopmentSoarGroup/Soar:developmentfrom singletons-setting-unit-tests-fixSoarGroup/Soar:singletons-setting-unit-tests-fixCopy head branch name to clipboardMay 1, 2024

+266-216

Commits

Commits on Mar 22, 2024

• Revert "Revert "new setting that makes soar turn any string attribute into a singleton pattern, set to "on" by default""

  Show description for ab93647
  scijones

  committed
  ab93647

  Copy full SHA for ab93647

• updated expected chunks for ChunkingDemoTests_Demo_Arithmetic test to not have a redundant second test for the same wme.

  scijones

  committed
  9ae9987

  Copy full SHA for 9ae9987

• updated the ChunkingDemoTests_Demo_Blocks_World_Look_Ahead_expected.soar to no longer have three extra redundant ontop relations in the chunks.

  scijones

  committed
  2cc8942

  Copy full SHA for 2cc8942

• Updated ChunkingDemoTests_Demo_Blocks_World_Operator_Subgoaling_expected.soar to no longer redundantly test the same bottom block twice.

  scijones

  committed
  7bf2d13

  Copy full SHA for 7bf2d13

• Updated ChunkingDemoTests_Demo_Eight_Puzzle_expected.soar to no longer expect chunks with redundant testing of the same ^binding

  scijones

  committed
  b4e7ebb

  Copy full SHA for b4e7ebb

• Updated ChunkingDemoTests_Demo_Water_Jug_Look_Ahead_expected.soar to not expect a redundant test for the water jug problem space.

  scijones

  committed
  c72fdbc

  Copy full SHA for c72fdbc

• Updated ChunkingTests_Maintain_Instantiation_Specific_Identity_expected.soar to not redundantly test ^item1

  scijones

  committed
  b56a1dc

  Copy full SHA for b56a1dc

• updated ChunkingTests_Opaque_State_Barrier_expected.soar not to redundantly test ds, dd, and d

  scijones

  committed
  032fd02

  Copy full SHA for 032fd02

• Updated ChunkingTests_Superstate_Identity_Opaque_expected.soar to test for a single foo satisfying both negations. technically this is a slightly less general chunk, but still correct.

  scijones

  committed
  7bf05b7

  Copy full SHA for 7bf05b7

• Updated ChunkingTests_NCC_from_Backtrace_expected.soar to expect a chunk with combined tests where they used to be split.

  scijones

  committed
  7e2b911

  Copy full SHA for 7e2b911

• added test that old singletons behavior is preserved when command is used to disable new singletons-by-default behavior

  Show description for d269cc9
  scijones

  committed
  d269cc9

  Copy full SHA for d269cc9

• updated literalization_simple to not expect redundant foo and bar tests. note that the point of the test was to examine whether things were literalized that should be, so this is actually "more cor…

  Show description for 3720f37
  scijones

  committed
  3720f37

  Copy full SHA for 3720f37

• updated ChunkingTests_Literalization_with_BT_Constraints_expected to not redundantly test word twice

  scijones

  committed
  e44a5c5

  Copy full SHA for e44a5c5

• Updated ChunkingTests_Literalization_with_BT_Constraints2_expected to not redundantly test the same object1 twice. it seems like this is another case where maybe the default to have singletons lead…

  Show description for b93dc69
  scijones

  committed
  b93dc69

  Copy full SHA for b93dc69

• updated ChunkingTests_Literalization_with_Constraints to not expect a chunk with an extra test for word.

  scijones

  committed
  4bc5dba

  Copy full SHA for 4bc5dba

Commits on Mar 25, 2024

• removed extra tests that only make sense if there were multiple operators present with the same names. They weren't really there, though. but yes, two logic paths, so it once again comes down to wh…

  Show description for 231ec2c
  scijones

  committed
  231ec2c

  Copy full SHA for 231ec2c

• updated Chunk_Operator_Tie_Item_Links test to not expect extra operator tests.

  scijones

  committed
  b187bff

  Copy full SHA for b187bff

• Updated RHS_Math_expected to not test both a literalization and a variable referring to the same thing.

  scijones

  committed
  749e405

  Copy full SHA for 749e405

• Updated RHS_Math_Abs test to not expect a redundant test.

  scijones

  committed
  6e52efb

  Copy full SHA for 6e52efb

• Updated RHS_Math_Mixed to not expect redundant tests and to not use a variable in the RHS where I think it is supposed to use a literalization.

  scijones

  committed
  a9a7ac0

  Copy full SHA for a9a7ac0

• Updated RHS_Math_Children_Force_Learn to expect a chunk with only the test for ^value 3 and not variablized ^values.

  scijones

  committed
  6952bf8

  Copy full SHA for 6952bf8

• I think I removed the redundant goal tests, so I can't easily figure out why this one still fails.

  scijones

  committed
  cec9918

  Copy full SHA for cec9918

Commits on Mar 26, 2024

• i'm reading a bit into the name of the test "disjunction merge", but the point here is that with the singletons setting on, the intersection of the disjunctive sets is really what becomes the condi…

  Show description for 804f36e
  scijones

  committed
  804f36e

  Copy full SHA for 804f36e

• it's not that I think the new setting makes a bad chunk, but it seems like in this test, it defeats the point of the test. so, updated justifications_get_new_identities to turn automatic singletons…

  Show description for 21921a2
  scijones

  committed
  21921a2

  Copy full SHA for 21921a2

• updated BUNCPS_0 to not expect a redundant testr for name, instead a single merged test.

  scijones

  committed
  da1bfc9

  Copy full SHA for da1bfc9

• updated BUNCPS_3 to expect a single condition with both attributes (as it really is in the state)

  scijones

  committed
  19db11d

  Copy full SHA for 19db11d

• updated BUNCPS_6_Four_Level not to redundantly test attr1

  scijones

  committed
  f01d026

  Copy full SHA for f01d026

• I admit I don't entirely know what aspect of deep-copy this is supposed to be testing exactly, but all I had to do was eliminate redundant tests, so, I think this is fine.

  scijones

  committed
  f75acd7

  Copy full SHA for f75acd7

• teach_soar_90_games is too complicated for me to figure out easily what the right chunks should be, so for now this has the new setting turned off.

  scijones

  committed
  b453e25

  Copy full SHA for b453e25

• the chunks didn't look bad, but this learns two chunks instead of one (a chunk learned from a chunk), and it seemed like it could suck up a lot of time to determine the correct behavior, so the aut…

  Show description for 7a1d0df
  scijones

  committed
  7a1d0df

  Copy full SHA for 7a1d0df

• the operator selection knowledge "mega test" is complicated -- no way to easily determine correct chunks in short period of time, so automatically-create-singletons off

  scijones

  committed
  b8b6be4

  Copy full SHA for b8b6be4

• for the preexisting singletons tests, I assume they expect an automatic singleton creation setting to not be on.

  scijones

  committed
  d01f90b

  Copy full SHA for d01f90b

• again, for preexisting singletons tests, I assume they would not expect the automatically-create-singletons tests to be on.

  scijones

  committed
  298bc4c

  Copy full SHA for 298bc4c

• could not eyeball what the precise difference was, but in the original (earlier commit) expectation, there was redundant testing, so I'm just using the new chunk as the "correct" expectation for th…

  Show description for 92451ab
  scijones

  committed
  92451ab

  Copy full SHA for 92451ab

• updated unify_through_two_traces_four_deep not to expect redundant "word" tests.

  scijones

  committed
  4887e08

  Copy full SHA for 4887e08

• even though the ToH agent is actually a broken agent as it is, leaving this test in for now. does give old behavior with the automatically-create-singletons off setting.

  scijones

  committed
  ede5e5b

  Copy full SHA for ede5e5b

• the 90 games test is the most problematic. Even with the setting switched to off, the old behavior is not precisely replicated, but it seems to be a brittle test in general. I suppose just having t…

  Show description for 4fc44ac
  scijones

  committed
  4fc44ac

  Copy full SHA for 4fc44ac