feat!: allow custom seed when using targetingKey override for fractio…

…nal op (open-feature#1266)

<!-- Please use this template for your pull request. -->
<!-- Please use the sections that you need and delete other sections -->

## This PR
<!-- add the description of the PR here -->

- no longer injects flagKey as seed for fractional op when user has
provided custom targeting
- updates schema to allow `cat` and other operations so that custom
targeting can be properly seeded

### Related Issues
<!-- add here the GitHub issue that this PR resolves if applicable -->

open-feature#1264

### Notes
<!-- any additional notes for this PR -->

### Follow-up Tasks
<!-- anything that is related to this PR but not done here should be
noted under this section -->
<!-- if there is a need for a new issue, please link it here -->

### How to test
<!-- if applicable, add testing instructions under this section -->

```bash
# unit tests
make test

# gherkin tests
./bin/flagd start -f file:test-harness/flags/testing-flags.json -f file:test-harness/flags/custom-ops.json -f file:test-harness/flags/evaluator-refs.json -f file:test-harness/flags/zero-flags.json
make flagd-integration-test
```

---------

Signed-off-by: Cole Bailey <cole.bailey@deliveryhero.com>
Signed-off-by: Cole Bailey <cole.bailey.one@gmail.com>
Signed-off-by: Kavindu Dodanduwa <kavindudodanduwa@gmail.com>
Co-authored-by: Michael Beemer <beeme1mr@users.noreply.github.com>
Co-authored-by: Kavindu Dodanduwa <kavindudodanduwa@gmail.com>

.github/workflows/build.yaml

@@ -137,7 +137,8 @@ jobs:
           -f file:${{ github.workspace }}/test-harness/flags/testing-flags.json \
           -f file:${{ github.workspace }}/test-harness/flags/custom-ops.json \
           -f file:${{ github.workspace }}/test-harness/flags/evaluator-refs.json \
-          -f file:${{ github.workspace }}/test-harness/flags/zero-flags.json &
+          -f file:${{ github.workspace }}/test-harness/flags/zero-flags.json \
+          -f file:${{ github.workspace }}/test-harness/flags/edge-case-flags.json &
 
       - name: Run evaluation test suite
         run: go clean -testcache && go test -cover ./test/integration

core/go.mod

@@ -11,7 +11,7 @@ require (
 	connectrpc.com/otelconnect v0.7.0
 	github.com/diegoholiveira/jsonlogic/v3 v3.4.0
 	github.com/fsnotify/fsnotify v1.7.0
-	github.com/open-feature/flagd-schemas v0.2.9-0.20240215170351-8c72c14eebff
+	github.com/open-feature/flagd-schemas v0.2.9-0.20240408192555-ea4f119d2bd7
 	github.com/open-feature/open-feature-operator/apis v0.2.39
 	github.com/prometheus/client_golang v1.19.0
 	github.com/robfig/cron v1.2.0

core/go.sum

@@ -435,6 +435,7 @@ github.com/cncf/xds/go v0.0.0-20230105202645-06c439db220b/go.mod h1:eXthEFrGJvWH
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc h1:U9qPSI2PIWSS1VwoXQT9A3Wy9MM3WgvqSxFWenqJduM=
+github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/diegoholiveira/jsonlogic/v3 v3.4.0 h1:TN++nRmEMA5UHzKl8MJ1kbF5SSzWtKHE0PZ6ITbJeH4=
 github.com/diegoholiveira/jsonlogic/v3 v3.4.0/go.mod h1:9oE8z9G+0OMxOoLHF3fhek3KuqD5CBqM0B6XFL08MSg=
 github.com/emicklei/go-restful/v3 v3.12.0 h1:y2DdzBAURM29NFF94q6RaY4vjIH1rtwDapwQtU84iWk=
@@ -626,8 +627,8 @@ github.com/onsi/ginkgo/v2 v2.14.0 h1:vSmGj2Z5YPb9JwCWT6z6ihcUvDhuXLc3sJiqd3jMKAY
 github.com/onsi/ginkgo/v2 v2.14.0/go.mod h1:JkUdW7JkN0V6rFvsHcJ478egV3XH9NxpD27Hal/PhZw=
 github.com/onsi/gomega v1.30.0 h1:hvMK7xYz4D3HapigLTeGdId/NcfQx1VHMJc60ew99+8=
 github.com/onsi/gomega v1.30.0/go.mod h1:9sxs+SwGrKI0+PWe4Fxa9tFQQBG5xSsSbMXOI8PPpoQ=
-github.com/open-feature/flagd-schemas v0.2.9-0.20240215170351-8c72c14eebff h1:ZJwqlDjz+vfMzs+pYXqThCg3YV8wXxON8YztweiuaQ0=
-github.com/open-feature/flagd-schemas v0.2.9-0.20240215170351-8c72c14eebff/go.mod h1:WKtwo1eW9/K6D+4HfgTXWBqCDzpvMhDa5eRxW7R5B2U=
+github.com/open-feature/flagd-schemas v0.2.9-0.20240408192555-ea4f119d2bd7 h1:oP+BH8RiNEmSWTffKEXz2ciwen7wbvyX0fESx0aoJ80=
+github.com/open-feature/flagd-schemas v0.2.9-0.20240408192555-ea4f119d2bd7/go.mod h1:WKtwo1eW9/K6D+4HfgTXWBqCDzpvMhDa5eRxW7R5B2U=
 github.com/open-feature/open-feature-operator/apis v0.2.39 h1:uzZplxAes8/KQWVpC3AE0ME2CNNks2gaaufSZUBkVnM=
 github.com/open-feature/open-feature-operator/apis v0.2.39/go.mod h1:I/4tLd5D4JpWpaFZxe2o8R2S1isWGNwHDSC/H5h7o3A=
 github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
@@ -637,6 +638,7 @@ github.com/pkg/sftp v1.10.1/go.mod h1:lYOWFsE0bwd1+KfKJaKeuokY15vzFx25BLbzYYoAxZ
 github.com/pkg/sftp v1.13.1/go.mod h1:3HaPG6Dq1ILlpPZRO0HVMrsydcdLt6HRDccSgb87qRg=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2 h1:Jamvg5psRIccs7FGNTlIRMkT8wgtp5eCXdBlqhYGL6U=
+github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/prometheus/client_golang v1.19.0 h1:ygXvpU1AoN1MhdzckN+PyD9QJOSD4x7kmXYlnfbA6JU=
 github.com/prometheus/client_golang v1.19.0/go.mod h1:ZRM9uEAypZakd+q/x7+gmsvXdURP+DABIEIjnmDdp+k=
 github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=

core/pkg/evaluator/fractional.go

@@ -56,18 +56,20 @@ func parseFractionalEvaluationData(values, data any) (string, []fractionalEvalua
 	if ok {
 		valuesArray = valuesArray[1:]
 	} else {
-		bucketBy, ok = dataMap[targetingKeyKey].(string)
+		targetingKey, ok := dataMap[targetingKeyKey].(string)
 		if !ok {
 			return "", nil, errors.New("bucketing value not supplied and no targetingKey in context")
 		}
+
+		bucketBy = fmt.Sprintf("%s%s", properties.FlagKey, targetingKey)
 	}
 
 	feDistributions, err := parseFractionalEvaluationDistributions(valuesArray)
 	if err != nil {
 		return "", nil, err
 	}
 
-	return fmt.Sprintf("%s%s", properties.FlagKey, bucketBy), feDistributions, nil
+	return bucketBy, feDistributions, nil
 }
 
 func parseFractionalEvaluationDistributions(values []any) ([]fractionalEvaluationDistribution, error) {

core/pkg/evaluator/fractional_test.go

@@ -32,7 +32,7 @@ func TestFractionalEvaluation(t *testing.T) {
 											  },
 											  {
 												"fractional": [
-												  {"var": "email"},
+												  {"cat": [{"var": "$flagd.flagKey"}, {"var": "email"}]},
 												  [
 												  "red",
 												  25
@@ -54,6 +54,34 @@ func TestFractionalEvaluation(t *testing.T) {
 											]
 										  }`),
 			},
+			"customSeededHeaderColor": {
+				State:          "ENABLED",
+				DefaultVariant: "red",
+				Variants: map[string]any{
+					"red":    "#FF0000",
+					"blue":   "#0000FF",
+					"green":  "#00FF00",
+					"yellow": "#FFFF00",
+				},
+				Targeting: []byte(`{
+					"if": [
+						{
+						"in": ["@faas.com", {
+								"var": ["email"]
+								}]
+						},
+						{
+						"fractional": [
+							{"cat": ["my-seed", {"var": "email"}]},
+							["red",25],
+							["blue",25],
+							["green",25],
+							["yellow",25]
+						]
+						}, null
+					]				  
+				}`),
+			},
 		},
 	}
 
@@ -106,6 +134,46 @@ func TestFractionalEvaluation(t *testing.T) {
 			expectedValue:   "#00FF00",
 			expectedReason:  model.TargetingMatchReason,
 		},
+		"rachel@faas.com with custom seed": {
+			flags:   flags,
+			flagKey: "customSeededHeaderColor",
+			context: map[string]any{
+				"email": "rachel@faas.com",
+			},
+			expectedVariant: "green",
+			expectedValue:   "#00FF00",
+			expectedReason:  model.TargetingMatchReason,
+		},
+		"monica@faas.com with custom seed": {
+			flags:   flags,
+			flagKey: "customSeededHeaderColor",
+			context: map[string]any{
+				"email": "monica@faas.com",
+			},
+			expectedVariant: "red",
+			expectedValue:   "#FF0000",
+			expectedReason:  model.TargetingMatchReason,
+		},
+		"joey@faas.com with custom seed": {
+			flags:   flags,
+			flagKey: "customSeededHeaderColor",
+			context: map[string]any{
+				"email": "joey@faas.com",
+			},
+			expectedVariant: "green",
+			expectedValue:   "#00FF00",
+			expectedReason:  model.TargetingMatchReason,
+		},
+		"ross@faas.com with custom seed": {
+			flags:   flags,
+			flagKey: "customSeededHeaderColor",
+			context: map[string]any{
+				"email": "ross@faas.com",
+			},
+			expectedVariant: "green",
+			expectedValue:   "#00FF00",
+			expectedReason:  model.TargetingMatchReason,
+		},
 		"ross@faas.com with different flag key": {
 			flags: Flags{
 				Flags: map[string]model.Flag{

docs/reference/custom-operations/fractional-operation.md

@@ -10,7 +10,13 @@ OpenFeature allows clients to pass contextual information which can then be used
 // Factional evaluation property name used in a targeting rule
 "fractional": [
   // Evaluation context property used to determine the split
-  { "var": "email" },
+  // Note using `cat` and `$flagd.flagKey` is the suggested default to seed your hash value and prevent bucketing collisions
+  {
+    "cat": [
+      { "var": "$flagd.flagKey" },
+      { "var": "email" }
+    ]
+  },
   // Split definitions contain an array with a variant and percentage
   // Percentages must add up to 100
   [
@@ -33,7 +39,7 @@ The `defaultVariant` is `red`, but it contains a [targeting rule](../flag-defini
 
 In this case, `25%` of the evaluations will receive `red`, `25%` will receive `blue`, and so on.
 
-Assignment is deterministic (sticky) based on the expression supplied as the first parameter (`{ "var": "email" }`, in this case).
+Assignment is deterministic (sticky) based on the expression supplied as the first parameter (`{ "cat": [{ "var": "$flagd.flagKey" }, { "var": "email" }]}`, in this case).
 The value retrieved by this expression is referred to as the "bucketing value".
 The bucketing value expression can be omitted, in which case a concatenation of the `targetingKey` and the `flagKey` will be used.
 
@@ -46,8 +52,10 @@ is selected.
 As hashing is deterministic we can be sure to get the same result every time for the same data point.
 
 The `fractional` operation can be added as part of a targeting definition.
-The value is an array and the first element is the name of the property to use from the evaluation context.
+The value is an array and the first element is a nested JsonLogic rule which resolves to the hash key.
+This rule should typically consist of a seed concatenated with a session variable to use from the evaluation context.
 This value should typically be something that remains consistent for the duration of a users session (e.g. email or session ID).
+The seed is typically the flagKey so that experiments running across different flags are statistically independent, however, you can also specify another seed to either align or further decouple your allocations across different feature flags or use-cases.
 The other elements in the array are nested arrays with the first element representing a variant and the second being the percentage that this option is selected.
 There is no limit to the number of elements but the configured percentages must add up to 100.
 
@@ -69,7 +77,12 @@ Flags defined as such:
       "state": "ENABLED",
       "targeting": {
         "fractional": [
-          { "var": "email" },
+          { 
+            "cat": [
+              { "var": "$flagd.flagKey" },
+              { "var": "email" }
+            ]
+          },
           [
             "red",
             50

docs/reference/specifications/custom-operations/fractional-operation-spec.md

@@ -34,7 +34,12 @@ JSON object. Below is an example of a targeting rule containing a `fractional`:
       "state": "ENABLED",
       "targeting": {
         "fractional": [
-          {"var":"email"},
+          {
+            "cat": [
+              { "var": "$flagd.flagKey" },
+              { "var": "email" }
+            ]
+          },
           [
             "red",
             50