node-autotask 0.33.0 → 0.37.0

package/docs/LIVE-TESTING.md

@@ -128,25 +128,25 @@ DRY-RUN-ONLY verdict (the only no-credentials path, exit 0).
 
       npm run live-coverage -- --only tickets          # one resource's ops
       npm run live-coverage -- --only configuration_item_ssl  # the cisan node-parity op
-      npm run live-coverage                             # FULL sweep: 1182 rows, --max-calls 1600
+      npm run live-coverage                             # FULL sweep: 1281 rows, --max-calls 1600
 
 Full-sweep budget (worst case, every table non-empty — the machine-readable
-number is the report's `budget` block): PLAN-HONEST 930 bounded op calls
+number is the report's `budget` block): PLAN-HONEST 1016 bounded op calls
 (every read/helper row = at most ONE bounded call — reads use
 `queryBounded` limit 2, a TOTAL cap; the served-artifact rows —
 operations.blindSurfaces (issue #11), operations.createGuidance (issue
 #12) — plan the op call at ZERO wire, but still count as one op call in
-this number: the budget counts ROW calls, not raw wire, so the 930 is the
-conservative row-based figure, 907 if the zero-wire rows are subtracted)
-+ <=77 best-effort update diff GETs
-+ <=121 cached limit-1 id reads (one per resource; the parent chain shares
-the cache) = <= 1128. WIRE-HONEST: the dry-run verification reads ride on
+this number: the budget counts ROW calls, not raw wire, so the 1016 is the
+conservative row-based figure)
++ <=82 best-effort update diff GETs
++ <=129 cached limit-1 id reads (one per resource; the parent chain shares
+the cache) = <= 1227. WIRE-HONEST: the dry-run verification reads ride on
 top — 54 delete current-record GETs + 63 scoped-create parent-verify GETs
-+ 20 getContext part GETs + <=15 special-op reads (incl. the issue #9
++ 20 getContext part GETs + <=18 special-op reads (incl. the issue #9
 `operations.recovery` dry-run find — ONE bounded query, zero mutating
-wire) = <=152 — so the worst case is <= 1280 raw wire requests. The 1600
---max-calls cap bounds the row-based plan (actualCalls <= 1128); the wire
-count cannot exceed plan + the stated verification set (and 1280 is far
+wire) = <=155 — so the worst case is <= 1382 raw wire requests. The 1600
+--max-calls cap bounds the row-based plan (actualCalls <= 1227); the wire
+count cannot exceed plan + the stated verification set (and 1382 is far
 under the API's 10k req/hour). ZERO mutating requests: every mutation travels
 `dryRun: true` (the sweep flag + the SDK governor's dry-run-first default,
 double-locked; pinned per-resource in the generated test suite), and EVERY

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "node-autotask",
-  "version": "0.33.0",
+  "version": "0.37.0",
   "description": "Fully-typed TypeScript SDK for the Autotask (Datto) PSA REST API. Zero runtime dependencies. Built for building MCP servers and integrations.",
   "type": "module",
   "sideEffects": false,