test(flagd): run the provider conformance suite against both resolvers by aepfli · Pull Request #411 · open-feature/python-sdk-contrib

feat(flagd): run the provider conformance suite against both resolvers

aepfli

committed

de11a61

fix(flagd): add the provider conformance suite to uv.lock

aepfli

committed

27bb722

test(flagd): follow the @numeric-coercion rename

aepfli

committed

d310f06

test(flagd): declare @large-integers and withdraw @numeric-coercion

aepfli

committed

225469a

test(flagd): the falsy flags were a naming gap, not a backend gap

aepfli

committed

7b98321

test(flagd): withhold @reinitialization, and say what was measured

aepfli

committed

67dac1c

test(flagd): declare @variants and @targeting, on a measured run

aepfli

committed

e4d166d

test(flagd): declare @disabled-flags on both resolvers, measured

aepfli

committed

38aed3e

test(flagd): declare the stack instead of wiring it

aepfli

committed

ad8422e

test(flagd): keep the conformance suite out of the default build, and say so

aepfli

committed

e937b3d

test(flagd): point at Appendix F for the policy, and keep the suite importing

aepfli

committed

a62b378

test(flagd): both resolvers claim the standard resolution reasons

aepfli

committed

2c35cae

test(flagd): say what the numeric-coercion scenarios actually do, measured

aepfli

committed

f515e6b

test(flagd): declare what a run supports, per resolver, and say what it measured

aepfli

committed

8e411e7

docs(flagd): cite the declaring rule instead of stating it again

aepfli

committed

4d58afd

chore(flagd): ignore the flag file every e2e run generates

aepfli

committed

0c21e4d

docs(flagd): cite rule six with the condition it was missing

aepfli

committed

56490ee

test(flagd): name the conformance modules for the resolver, not the suite

aepfli

committed

24169d3

test(flagd): let the collect step run even when the default suite is red

aepfli

committed

d4834d2

test(flagd): drop a gitignore this branch has no reason to add

aepfli

committed

ca2ddbf

test(flagd): one compose definition of the backend for the whole repo

aepfli

committed

2f997a0

docs(flagd): keep the evidence, link the rules, delete the retellings

aepfli

committed

1be4852

test(tck): run the conformance suite against flagd-testbed v3.10.1

aepfli

committed

be1638a

test(tck): stop naming one testbed release as the flag set's boundary

aepfli

committed

ca656ba

test(flagd): declare @string-typing on both resolvers, measured

aepfli

committed

ce5e7cc

test(flagd): declare @fully-typed-values on both resolvers, measured

aepfli

committed

2ccc945

docs(flagd): the standard-reasons note sits with the declarations it describes

aepfli

committed

6f8d796

test(flagd): sync the conformance assets where the scenarios execute

aepfli

committed

da6fdcf

docs(flagd): stop naming a testbed release in the README tally

aepfli

committed

969c87f

Navigation Menu

test(flagd): run the provider conformance suite against both resolvers - #411

test(flagd): run the provider conformance suite against both resolvers#411
aepfli wants to merge 29 commits into
feat/provider-tckfrom
feat/provider-tck-flagd

Commits on Oct 1, 2026

feat(flagd): run the provider conformance suite against both resolvers

fix(flagd): add the provider conformance suite to uv.lock

test(flagd): follow the @numeric-coercion rename

test(flagd): declare @large-integers and withdraw @numeric-coercion

test(flagd): the falsy flags were a naming gap, not a backend gap

test(flagd): withhold @reinitialization, and say what was measured

test(flagd): declare @variants and @targeting, on a measured run

test(flagd): declare @disabled-flags on both resolvers, measured

test(flagd): declare the stack instead of wiring it

test(flagd): keep the conformance suite out of the default build, and say so

test(flagd): point at Appendix F for the policy, and keep the suite importing

test(flagd): both resolvers claim the standard resolution reasons

test(flagd): say what the numeric-coercion scenarios actually do, measured

test(flagd): declare what a run supports, per resolver, and say what it measured

docs(flagd): cite the declaring rule instead of stating it again

chore(flagd): ignore the flag file every e2e run generates

docs(flagd): cite rule six with the condition it was missing

test(flagd): name the conformance modules for the resolver, not the suite

test(flagd): let the collect step run even when the default suite is red

test(flagd): drop a gitignore this branch has no reason to add

test(flagd): one compose definition of the backend for the whole repo

docs(flagd): keep the evidence, link the rules, delete the retellings

test(tck): run the conformance suite against flagd-testbed v3.10.1

test(tck): stop naming one testbed release as the flag set's boundary

test(flagd): declare @string-typing on both resolvers, measured

test(flagd): declare @fully-typed-values on both resolvers, measured

docs(flagd): the standard-reasons note sits with the declarations it describes

test(flagd): sync the conformance assets where the scenarios execute

docs(flagd): stop naming a testbed release in the README tally