@pionne/node 0.3.3 → 0.3.5

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 (2) hide show
  1. package/README.md +28 -1
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -77,6 +77,33 @@ Pionne.setTags({ tier: 'pro' });
77
77
  Pionne.setEnabled(false);
78
78
  ```
79
79
 
80
+ ### Profiling — preview (coming soon)
81
+
82
+ Continuous-ish CPU profiling is **shipped on `@pionne/react-native@0.8.0`**
83
+ (Hermes sampler) and is on the roadmap for `@pionne/node` next. The Node
84
+ implementation will use the V8 inspector profiler (`node:inspector`,
85
+ `Profiler.start`/`Profiler.stop`) and ship the resulting CPU profile to
86
+ the same `POST /api/profiles` endpoint.
87
+
88
+ The API will mirror RN exactly:
89
+
90
+ ```ts
91
+ // Coming in @pionne/node ~v0.4.0
92
+ await Pionne.profile('checkoutHandler', async () => {
93
+ await processOrder(orderId);
94
+ }, { route: 'POST /api/checkout' });
95
+ ```
96
+
97
+ Same backend, same retention (raw 7 d, aggregates 90 d), same flame graph
98
+ view + cross-release regression chart in the mobile dashboard. If you want
99
+ profiling **today** on the server, you can post CPU profiles to the
100
+ endpoint directly — the JSON shape is documented at
101
+ [pionne.agkgcreations.fr/profiling/intro](https://pionne.agkgcreations.fr/profiling/intro).
102
+
103
+ Heads-up : V8 profiling has a 3–8 % CPU overhead during capture. We'll
104
+ default to manual transaction-scoped (start/stop) rather than continuous
105
+ to keep that cost off your hot path.
106
+
80
107
  ### Bundle ID pinning — N/A on Node
81
108
 
82
109
  The "Bundle ID" anti-token-theft check on Pionne projects is **mobile only**
@@ -137,7 +164,7 @@ Voir les types TypeScript pour la liste complète.
137
164
 
138
165
  ## Rate limit serveur
139
166
 
140
- L'API Pionne cap **600 req/min/token** (= 10/sec) sur tous les endpoints publics (`/ingest`, `/sessions`, `/feedback`). Au-delà → `HTTP 429` avec un header `Retry-After`. Le SDK fait silencieusement échouer (try/catch interne). Empêche un token leaké (ou un worker qui boucle) de drainer ton infra ou ton quota mensuel. Voir [doc rate limits](https://pionne.agkgcreations.fr/security/rate-limits).
167
+ L'API Pionne applique un rate-limit par token sur tous les endpoints publics. Au-delà → `HTTP 429` avec un header `Retry-After`. Le SDK fait silencieusement échouer (try/catch interne). Empêche un token leaké (ou un worker qui boucle) de drainer ton quota mensuel. Voir [doc rate limits](https://pionne.agkgcreations.fr/security/rate-limits).
141
168
 
142
169
  ## License
143
170
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@pionne/node",
3
- "version": "0.3.3",
3
+ "version": "0.3.5",
4
4
  "description": "Error monitoring SDK for Node.js \u2014 Pionne. Auto-captures uncaught exceptions and unhandled rejections, ships runtime context (Node, OS, hostname, pid), and tracks Release Health (crash-free user rate per release).",
5
5
  "license": "MIT",
6
6
  "author": "AGKG Creations",