@binary-balance/seshat 0.6.0 → 0.8.0

package/README.md

@@ -10,9 +10,11 @@ insufficiently tested code.
 ## What Seshat does
 
 - Calculates function complexity and CRAP scores from branch coverage.
-- Changes comparison, logical and arithmetic operators one at a time and checks
-  whether tests detect them.
+- Selects whole test files for verified Vitest Istanbul setups, with full-run fallbacks and an opt-out.
+- Changes comparison, logical, arithmetic and boolean expressions one at a time
+  and checks whether tests detect them.
 - Replaces function bodies to find covered functions whose removal tests miss.
+- Removes expression statements and bare returns from eligible functions.
 - Runs tests in isolated copies of the captured project.
 - Produces readable terminal output and versioned JSON, with optional CI thresholds.
 - Supports explicit source selection, multiple test setups and parallel mutation workers.
@@ -84,8 +86,11 @@ The [examples](https://github.com/Binary-Balance/seshat/blob/main/examples/READM
 and Jest/Expo projects. The [configuration guide](https://github.com/Binary-Balance/seshat/blob/main/docs/configuration.md) covers
 every field, including `workers` for parallel mutation runs.
 
-Comparison, logical and arithmetic mutants are enabled by default. Set
+Comparison, logical, arithmetic and boolean mutants are enabled by default. Set
 `"mutationOperators": ["comparison"]` to retain the operator scope of 0.4.0.
+Use `["comparison", "logical", "arithmetic"]` to retain the scope of 0.5.0 and
+0.6.0. The `boolean` family flips runtime `true`/`false` literals and removes
+prefix `!`, including its boolean conversion.
 The broader default can add test executions and change mutation scores.
 
 ## Use
@@ -125,7 +130,8 @@ leaving the non-member outcome untested. Its statement coverage is 100%,
 but its branch coverage is 50%. Both measurements appear in the report.
 
 The mutation score is the percentage of mutants that made a test fail. A mutant
-is one changed operator, such as `>` to `>=` or `&&` to `||`. If tests catch two
+makes one deliberate code change, such as `>` to `>=` or `true` to `false`.
+If tests catch two
 of three mutants, the score is `66.67%`. A timeout or execution error leaves the run
 incomplete and withholds the score. Neither score proves correctness.
 
@@ -142,7 +148,9 @@ Add optional thresholds to `seshat.json`:
 ```json
 "thresholds": {
   "maxCrap": 30,
-  "minMutationScore": 80
+  "minMutationScore": 80,
+  "maxInferredAny": 10,
+  "maxTypeEscapes": 0
 }
 ```
 
@@ -153,6 +161,10 @@ Add optional thresholds to `seshat.json`:
 | `2` | Invalid input, failed baseline or incomplete run. |
 | `130` / `143` | Cancelled by `SIGINT` / `SIGTERM` on Unix. |
 
+`maxInferredAny` limits any-typed runtime expression sites separately in each
+setup. `maxTypeEscapes` limits explicit `any` and TypeScript suppression
+comments across selected source. Both apply in all three commands.
+
 See [quality thresholds](https://github.com/Binary-Balance/seshat/blob/main/docs/configuration.md#quality-thresholds) for the full
 rules.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@binary-balance/seshat",
-  "version": "0.6.0",
+  "version": "0.8.0",
   "description": "A native CLI for TypeScript and TSX complexity analysis and mutation testing.",
   "license": "MIT",
   "repository": {
@@ -20,10 +20,10 @@
     "LICENSE"
   ],
   "optionalDependencies": {
-    "@binary-balance/seshat-linux-x64": "0.6.0",
-    "@binary-balance/seshat-linux-arm64": "0.6.0",
-    "@binary-balance/seshat-darwin-x64": "0.6.0",
-    "@binary-balance/seshat-darwin-arm64": "0.6.0",
-    "@binary-balance/seshat-win32-x64": "0.6.0"
+    "@binary-balance/seshat-linux-x64": "0.8.0",
+    "@binary-balance/seshat-linux-arm64": "0.8.0",
+    "@binary-balance/seshat-darwin-x64": "0.8.0",
+    "@binary-balance/seshat-darwin-arm64": "0.8.0",
+    "@binary-balance/seshat-win32-x64": "0.8.0"
   }
 }