@sun-asterisk/sungen 3.2.12-beta.3 → 3.2.12-beta.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/dist/orchestrator/templates/ai-src/skills/sungen-data-factory/SKILL.md +17 -6
  2. package/package.json +3 -3
  3. package/src/orchestrator/templates/ai-src/skills/sungen-data-factory/SKILL.md +17 -6
package/dist/orchestrator/templates/ai-src/skills/sungen-data-factory/SKILL.md CHANGED
@@ -85,12 +85,15 @@ This is the core discipline: *common is never enough; when blind, think and ask
85
85
  ## Customizing the catalog — `.overwrite` (survives `sungen update`)
86
86
 
87
87
  The shipped `common/` is a floor. Project rules go in `qa/data-factory/common.overwrite.yaml`,
88
- deep-merged over the common (**later-wins**, keyed by `type` + data-point `id`): same `id` replaces,
89
- new `id` appends, `constraints` merge key-by-key. `sungen update` refreshes `common/` but never
90
- touches your overlay. If an update makes an overlay entry stale (references a bank/constraint that no
91
- longer exists), the loader **warns you to prune it** — it never edits your file. Prefer a targeted
92
- `errorMap`/`constraints` in the field-map for per-screen tweaks; use `.overwrite` only for rules that
93
- apply project-wide.
88
+ deep-merged over the common (**later-wins**, keyed by `type` + data-point `id`) via three explicit
89
+ operations: **EXTEND** (a new `id` appends), **REPLACE** (same `id` wins; `constraints` merge
90
+ key-by-key), **DISABLE** (list ids under `disable:` to drop shipped cases that don't apply — a stale
91
+ target id warns). You can also override `security:` banks and `techniques:` the same way. `sungen
92
+ update` refreshes `common/` but never touches your overlay; a stale entry **warns you to prune it** —
93
+ it never edits your file. Nothing is silently overwritten: **`sungen data provenance`** (and a
94
+ `⟲ .overwrite →` line at `data gen`) shows exactly what your overlay changed, and each overlaid value
95
+ is tagged `via: .overwrite` in the test-data. Prefer a targeted `errorMap`/`constraints` in the
96
+ field-map for per-screen tweaks; use `.overwrite` only for rules that apply project-wide.
94
97
 
95
98
  ## Bulk-import fixtures (CSV)
96
99
 
@@ -117,6 +120,14 @@ constraints and every synthesized `invalid` value actually violates them. A boun
117
120
  a constraint — fix the field-map/constraints and re-run. This is what stops the "structurally right but
118
121
  semantically wrong" data (e.g. a "below minimum length" case that is actually long enough to pass).
119
122
 
123
+ **Applicability & profiles.** The generator only emits cases that apply, and reports the rest (never
124
+ silent). Set **`required: true|false`** correctly — an optional field drops the "empty (required)" case;
125
+ a numeric field with `min ≥ 0` drops the negative example. Security/injection payloads land in their own
126
+ **`adversarial:`** bucket, separate from `invalid:`. Dropped cases show up as `_not_applicable` in the
127
+ output + a `NOT_APPLICABLE` lint info. A `--profile` (`regression` default · `functional` · `smoke` ·
128
+ `security-min`) and `--locale`/`--channel`/`--sink` on `data gen|validate|lint` select a reduced set;
129
+ selection is deterministic. Prefer setting `required` in the field-map over disabling the empty case.
130
+
120
131
  ## Weaving into scenarios
121
132
 
122
133
  When a `.feature` exists, express invalid/boundary sets as data-driven `Scenario Outline` + `Examples`
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@sun-asterisk/sungen",
3
- "version": "3.2.12-beta.3",
3
+ "version": "3.2.12-beta.5",
4
4
  "description": "Deterministic E2E Test Compiler - Gherkin + Selectors → Playwright tests",
5
5
  "main": "src/index.ts",
6
6
  "types": "src/index.ts",
@@ -39,8 +39,8 @@
39
39
  "@babel/types": "^7.28.5",
40
40
  "@cucumber/gherkin": "^37.0.0",
41
41
  "@cucumber/messages": "^31.0.0",
42
- "@sungen/driver-data-factory": "3.2.12-beta.3",
43
- "@sungen/driver-ui": "3.2.12-beta.3",
42
+ "@sungen/driver-data-factory": "3.2.12-beta.5",
43
+ "@sungen/driver-ui": "3.2.12-beta.5",
44
44
  "chalk": "^5.6.2",
45
45
  "commander": "^14.0.2",
46
46
  "dotenv": "^17.2.3",
package/src/orchestrator/templates/ai-src/skills/sungen-data-factory/SKILL.md CHANGED
@@ -85,12 +85,15 @@ This is the core discipline: *common is never enough; when blind, think and ask
85
85
  ## Customizing the catalog — `.overwrite` (survives `sungen update`)
86
86
 
87
87
  The shipped `common/` is a floor. Project rules go in `qa/data-factory/common.overwrite.yaml`,
88
- deep-merged over the common (**later-wins**, keyed by `type` + data-point `id`): same `id` replaces,
89
- new `id` appends, `constraints` merge key-by-key. `sungen update` refreshes `common/` but never
90
- touches your overlay. If an update makes an overlay entry stale (references a bank/constraint that no
91
- longer exists), the loader **warns you to prune it** — it never edits your file. Prefer a targeted
92
- `errorMap`/`constraints` in the field-map for per-screen tweaks; use `.overwrite` only for rules that
93
- apply project-wide.
88
+ deep-merged over the common (**later-wins**, keyed by `type` + data-point `id`) via three explicit
89
+ operations: **EXTEND** (a new `id` appends), **REPLACE** (same `id` wins; `constraints` merge
90
+ key-by-key), **DISABLE** (list ids under `disable:` to drop shipped cases that don't apply — a stale
91
+ target id warns). You can also override `security:` banks and `techniques:` the same way. `sungen
92
+ update` refreshes `common/` but never touches your overlay; a stale entry **warns you to prune it** —
93
+ it never edits your file. Nothing is silently overwritten: **`sungen data provenance`** (and a
94
+ `⟲ .overwrite →` line at `data gen`) shows exactly what your overlay changed, and each overlaid value
95
+ is tagged `via: .overwrite` in the test-data. Prefer a targeted `errorMap`/`constraints` in the
96
+ field-map for per-screen tweaks; use `.overwrite` only for rules that apply project-wide.
94
97
 
95
98
  ## Bulk-import fixtures (CSV)
96
99
 
@@ -117,6 +120,14 @@ constraints and every synthesized `invalid` value actually violates them. A boun
117
120
  a constraint — fix the field-map/constraints and re-run. This is what stops the "structurally right but
118
121
  semantically wrong" data (e.g. a "below minimum length" case that is actually long enough to pass).
119
122
 
123
+ **Applicability & profiles.** The generator only emits cases that apply, and reports the rest (never
124
+ silent). Set **`required: true|false`** correctly — an optional field drops the "empty (required)" case;
125
+ a numeric field with `min ≥ 0` drops the negative example. Security/injection payloads land in their own
126
+ **`adversarial:`** bucket, separate from `invalid:`. Dropped cases show up as `_not_applicable` in the
127
+ output + a `NOT_APPLICABLE` lint info. A `--profile` (`regression` default · `functional` · `smoke` ·
128
+ `security-min`) and `--locale`/`--channel`/`--sink` on `data gen|validate|lint` select a reduced set;
129
+ selection is deterministic. Prefer setting `required` in the field-map over disabling the empty case.
130
+
120
131
  ## Weaving into scenarios
121
132
 
122
133
  When a `.feature` exists, express invalid/boundary sets as data-driven `Scenario Outline` + `Examples`