@reflag/node-sdk 1.0.0-alpha.1
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.
- package/LICENSE +21 -0
- package/README.md +835 -0
- package/dist/package.json +49 -0
- package/dist/src/batch-buffer.js +85 -0
- package/dist/src/batch-buffer.js.map +1 -0
- package/dist/src/cache.js +71 -0
- package/dist/src/cache.js.map +1 -0
- package/dist/src/client.js +1141 -0
- package/dist/src/client.js.map +1 -0
- package/dist/src/config.js +98 -0
- package/dist/src/config.js.map +1 -0
- package/dist/src/edgeClient.js +32 -0
- package/dist/src/edgeClient.js.map +1 -0
- package/dist/src/fetch-http-client.js +85 -0
- package/dist/src/fetch-http-client.js.map +1 -0
- package/dist/src/flusher.js +56 -0
- package/dist/src/flusher.js.map +1 -0
- package/dist/src/inRequestCache.js +72 -0
- package/dist/src/inRequestCache.js.map +1 -0
- package/dist/src/index.js +9 -0
- package/dist/src/index.js.map +1 -0
- package/dist/src/periodicallyUpdatingCache.js +78 -0
- package/dist/src/periodicallyUpdatingCache.js.map +1 -0
- package/dist/src/rate-limiter.js +52 -0
- package/dist/src/rate-limiter.js.map +1 -0
- package/dist/src/types.js +6 -0
- package/dist/src/types.js.map +1 -0
- package/dist/src/utils.js +207 -0
- package/dist/src/utils.js.map +1 -0
- package/dist/types/src/batch-buffer.d.ts +27 -0
- package/dist/types/src/batch-buffer.d.ts.map +1 -0
- package/dist/types/src/cache.d.ts +16 -0
- package/dist/types/src/cache.d.ts.map +1 -0
- package/dist/types/src/client.d.ts +410 -0
- package/dist/types/src/client.d.ts.map +1 -0
- package/dist/types/src/config.d.ts +20 -0
- package/dist/types/src/config.d.ts.map +1 -0
- package/dist/types/src/edgeClient.d.ts +25 -0
- package/dist/types/src/edgeClient.d.ts.map +1 -0
- package/dist/types/src/fetch-http-client.d.ts +20 -0
- package/dist/types/src/fetch-http-client.d.ts.map +1 -0
- package/dist/types/src/flusher.d.ts +4 -0
- package/dist/types/src/flusher.d.ts.map +1 -0
- package/dist/types/src/inRequestCache.d.ts +7 -0
- package/dist/types/src/inRequestCache.d.ts.map +1 -0
- package/dist/types/src/index.d.ts +4 -0
- package/dist/types/src/index.d.ts.map +1 -0
- package/dist/types/src/periodicallyUpdatingCache.d.ts +16 -0
- package/dist/types/src/periodicallyUpdatingCache.d.ts.map +1 -0
- package/dist/types/src/rate-limiter.d.ts +14 -0
- package/dist/types/src/rate-limiter.d.ts.map +1 -0
- package/dist/types/src/types.d.ts +639 -0
- package/dist/types/src/types.d.ts.map +1 -0
- package/dist/types/src/utils.d.ts +65 -0
- package/dist/types/src/utils.d.ts.map +1 -0
- package/package.json +50 -0
package/README.md
ADDED
|
@@ -0,0 +1,835 @@
|
|
|
1
|
+
# Reflag Node.js SDK
|
|
2
|
+
|
|
3
|
+
Node.js, JavaScript/TypeScript client for [Reflag.com](https://reflag.com).
|
|
4
|
+
|
|
5
|
+
Reflag supports feature toggling, tracking feature usage, collecting feedback on features, and [remotely configuring features](#remote-config).
|
|
6
|
+
|
|
7
|
+
## Installation
|
|
8
|
+
|
|
9
|
+
Install using your favorite package manager:
|
|
10
|
+
|
|
11
|
+
{% tabs %}
|
|
12
|
+
{% tab title="npm" %}
|
|
13
|
+
|
|
14
|
+
```sh
|
|
15
|
+
npm i @reflag/node-sdk
|
|
16
|
+
```
|
|
17
|
+
|
|
18
|
+
{% endtab %}
|
|
19
|
+
|
|
20
|
+
{% tab title="yarn" %}
|
|
21
|
+
|
|
22
|
+
```sh
|
|
23
|
+
yarn add @reflag/node-sdk
|
|
24
|
+
```
|
|
25
|
+
|
|
26
|
+
{% endtab %}
|
|
27
|
+
|
|
28
|
+
{% tab title="bun" %}
|
|
29
|
+
|
|
30
|
+
```sh
|
|
31
|
+
bun add @reflag/node-sdk
|
|
32
|
+
```
|
|
33
|
+
|
|
34
|
+
{% endtab %}
|
|
35
|
+
|
|
36
|
+
{% tab title="pnpm" %}
|
|
37
|
+
|
|
38
|
+
```sh
|
|
39
|
+
pnpm add @reflag/node-sdk
|
|
40
|
+
```
|
|
41
|
+
|
|
42
|
+
{% endtab %}
|
|
43
|
+
|
|
44
|
+
{% tab title="deno" %}
|
|
45
|
+
|
|
46
|
+
```sh
|
|
47
|
+
deno add npm:@reflag/node-sdk
|
|
48
|
+
```
|
|
49
|
+
|
|
50
|
+
{% endtab %}
|
|
51
|
+
{% endtabs %}
|
|
52
|
+
|
|
53
|
+
Other supported languages/frameworks are in the [Supported languages](https://docs.reflag.com/quickstart/supported-languages) documentation pages.
|
|
54
|
+
|
|
55
|
+
You can also [use the HTTP API directly](https://docs.reflag.com/api/http-api)
|
|
56
|
+
|
|
57
|
+
## Basic usage
|
|
58
|
+
|
|
59
|
+
To get started you need to obtain your secret key from the [environment settings](https://app.reflag.com/env-current/settings/app-environments)
|
|
60
|
+
in Reflag.
|
|
61
|
+
|
|
62
|
+
> [!CAUTION]
|
|
63
|
+
> Secret keys are meant for use in server side SDKs only. Secret keys offer the users the ability to obtain
|
|
64
|
+
> information that is often sensitive and thus should not be used in client-side applications.
|
|
65
|
+
|
|
66
|
+
Reflag will load settings through the various environment variables automatically (see [Configuring](#configuring) below).
|
|
67
|
+
|
|
68
|
+
1. Find the Reflag secret key for your development environment under [environment settings](https://app.reflag.com/env-current/settings/app-environments) in Reflag.
|
|
69
|
+
2. Set `REFLAG_SECRET_KEY` in your `.env` file
|
|
70
|
+
3. Create a `reflag.ts` file containing the following:
|
|
71
|
+
|
|
72
|
+
```typescript
|
|
73
|
+
import { ReflagClient } from "@reflag/node-sdk";
|
|
74
|
+
|
|
75
|
+
// Create a new instance of the client with the secret key. Additional options
|
|
76
|
+
// are available, such as supplying a logger and other custom properties.
|
|
77
|
+
//
|
|
78
|
+
// We recommend that only one global instance of `client` should be created
|
|
79
|
+
// to avoid multiple round-trips to our servers.
|
|
80
|
+
export const reflagClient = new ReflagClient();
|
|
81
|
+
|
|
82
|
+
// Initialize the client and begin fetching feature targeting definitions.
|
|
83
|
+
// You must call this method prior to any calls to `getFlags()`,
|
|
84
|
+
// otherwise an empty object will be returned.
|
|
85
|
+
reflagClient.initialize().then({
|
|
86
|
+
console.log("Reflag initialized!")
|
|
87
|
+
})
|
|
88
|
+
```
|
|
89
|
+
|
|
90
|
+
Once the client is initialized, you can obtain features along with the `isEnabled`
|
|
91
|
+
status to indicate whether the feature is targeted for this user/company:
|
|
92
|
+
|
|
93
|
+
> [!IMPORTANT]
|
|
94
|
+
> If `user.id` or `company.id` is not given, the whole `user` or `company` object is ignored.
|
|
95
|
+
|
|
96
|
+
```typescript
|
|
97
|
+
// configure the client
|
|
98
|
+
const boundClient = reflagClient.bindClient({
|
|
99
|
+
user: {
|
|
100
|
+
id: "john_doe",
|
|
101
|
+
name: "John Doe",
|
|
102
|
+
email: "john@acme.com",
|
|
103
|
+
avatar: "https://example.com/users/jdoe",
|
|
104
|
+
},
|
|
105
|
+
company: {
|
|
106
|
+
id: "acme_inc",
|
|
107
|
+
name: "Acme, Inc.",
|
|
108
|
+
avatar: "https://example.com/companies/acme",
|
|
109
|
+
},
|
|
110
|
+
});
|
|
111
|
+
|
|
112
|
+
// get the huddle feature using company, user and custom context to
|
|
113
|
+
// evaluate the targeting.
|
|
114
|
+
const { isEnabled, track, config } = boundClient.getFlag("huddle");
|
|
115
|
+
|
|
116
|
+
if (isEnabled) {
|
|
117
|
+
// this is your feature gated code ...
|
|
118
|
+
// send an event when the feature is used:
|
|
119
|
+
track();
|
|
120
|
+
|
|
121
|
+
if (config?.key === "zoom") {
|
|
122
|
+
// this code will run if a given remote configuration
|
|
123
|
+
// is set up.
|
|
124
|
+
}
|
|
125
|
+
|
|
126
|
+
// CAUTION: if you plan to use the event for automated feedback surveys
|
|
127
|
+
// call `flush` immediately after `track`. It can optionally be awaited
|
|
128
|
+
// to guarantee the sent happened.
|
|
129
|
+
boundClient.flush();
|
|
130
|
+
}
|
|
131
|
+
```
|
|
132
|
+
|
|
133
|
+
You can also use the `getFlags()` method which returns a map of all features:
|
|
134
|
+
|
|
135
|
+
```typescript
|
|
136
|
+
// get the current features (uses company, user and custom context to
|
|
137
|
+
// evaluate the features).
|
|
138
|
+
const features = boundClient.getFlags();
|
|
139
|
+
const bothEnabled =
|
|
140
|
+
features.huddle?.isEnabled && features.voiceHuddle?.isEnabled;
|
|
141
|
+
```
|
|
142
|
+
|
|
143
|
+
## High performance feature targeting
|
|
144
|
+
|
|
145
|
+
The SDK contacts the Reflag servers when you call `initialize()`
|
|
146
|
+
and downloads the features with their targeting rules.
|
|
147
|
+
These rules are then matched against the user/company information you provide
|
|
148
|
+
to `getFlags()` (or through `bindClient(..).getFlags()`). That means the
|
|
149
|
+
`getFlags()` call does not need to contact the Reflag servers once
|
|
150
|
+
`initialize()` has completed. `ReflagClient` will continue to periodically
|
|
151
|
+
download the targeting rules from the Reflag servers in the background.
|
|
152
|
+
|
|
153
|
+
### Batch Operations
|
|
154
|
+
|
|
155
|
+
The SDK automatically batches operations like user/company updates and feature tracking events to minimize API calls.
|
|
156
|
+
The batch buffer is configurable through the client options:
|
|
157
|
+
|
|
158
|
+
```typescript
|
|
159
|
+
const client = new ReflagClient({
|
|
160
|
+
batchOptions: {
|
|
161
|
+
maxSize: 100, // Maximum number of events to batch
|
|
162
|
+
intervalMs: 1000, // Flush interval in milliseconds
|
|
163
|
+
},
|
|
164
|
+
});
|
|
165
|
+
```
|
|
166
|
+
|
|
167
|
+
You can manually flush the batch buffer at any time:
|
|
168
|
+
|
|
169
|
+
```typescript
|
|
170
|
+
await client.flush();
|
|
171
|
+
```
|
|
172
|
+
|
|
173
|
+
> [!TIP]
|
|
174
|
+
> It's recommended to call `flush()` before your application shuts down to ensure all events are sent.
|
|
175
|
+
|
|
176
|
+
### Rate Limiting
|
|
177
|
+
|
|
178
|
+
The SDK includes automatic rate limiting for feature events to prevent overwhelming the API.
|
|
179
|
+
Rate limiting is applied per unique combination of feature key and context. The rate limiter window size is configurable:
|
|
180
|
+
|
|
181
|
+
```typescript
|
|
182
|
+
const client = new ReflagClient({
|
|
183
|
+
rateLimiterOptions: {
|
|
184
|
+
windowSizeMs: 60000, // Rate limiting window size in milliseconds
|
|
185
|
+
},
|
|
186
|
+
});
|
|
187
|
+
```
|
|
188
|
+
|
|
189
|
+
### Flag definitions
|
|
190
|
+
|
|
191
|
+
Flag definitions include the rules needed to determine which features should be enabled and which config values should be applied to any given user/company.
|
|
192
|
+
Flag definitions are automatically fetched when calling `initialize()`.
|
|
193
|
+
They are then cached and refreshed in the background.
|
|
194
|
+
It's also possible to get the currently in use feature definitions:
|
|
195
|
+
|
|
196
|
+
```typescript
|
|
197
|
+
import fs from "fs";
|
|
198
|
+
|
|
199
|
+
const client = new ReflagClient();
|
|
200
|
+
|
|
201
|
+
const featureDefs = await client.getFlagDefinitions();
|
|
202
|
+
// [{
|
|
203
|
+
// key: "huddle",
|
|
204
|
+
// description: "Live voice conversations with colleagues."
|
|
205
|
+
// flag: { ... }
|
|
206
|
+
// config: { ... }
|
|
207
|
+
// }]
|
|
208
|
+
```
|
|
209
|
+
|
|
210
|
+
## Edge-runtimes like Cloudflare Workers
|
|
211
|
+
|
|
212
|
+
To use the Reflag NodeSDK with Cloudflare workers, set the `node_compat` flag [in your wrangler file](https://developers.cloudflare.com/workers/runtime-apis/nodejs/#get-started).
|
|
213
|
+
|
|
214
|
+
Instead of using `ReflagClient`, use `EdgeClient` and make sure you call `ctx.waitUntil(reflag.flush());` before returning from your worker function.
|
|
215
|
+
|
|
216
|
+
```typescript
|
|
217
|
+
import { EdgeClient } from "@reflag/node-sdk";
|
|
218
|
+
|
|
219
|
+
// set the REFLAG_SECRET_KEY environment variable or pass the secret key in the constructor
|
|
220
|
+
const reflag = new EdgeClient();
|
|
221
|
+
|
|
222
|
+
export default {
|
|
223
|
+
async fetch(request, _env, ctx): Promise<Response> {
|
|
224
|
+
// initialize the client and wait for it to complete
|
|
225
|
+
// if the client was initialized on a previous invocation, this is a no-op.
|
|
226
|
+
await reflag.initialize();
|
|
227
|
+
const features = reflag.getFlags({
|
|
228
|
+
user: { id: "userId" },
|
|
229
|
+
company: { id: "companyId" },
|
|
230
|
+
});
|
|
231
|
+
|
|
232
|
+
// ensure all events are flushed and any requests to refresh the feature cache
|
|
233
|
+
// have completed after the response is sent
|
|
234
|
+
ctx.waitUntil(reflag.flush());
|
|
235
|
+
|
|
236
|
+
return new Response(
|
|
237
|
+
`Flags for user ${userId} and company ${companyId}: ${JSON.stringify(features, null, 2)}`,
|
|
238
|
+
);
|
|
239
|
+
},
|
|
240
|
+
};
|
|
241
|
+
```
|
|
242
|
+
|
|
243
|
+
See [examples/cloudflare-worker](examples/cloudflare-worker/src/index.ts) for a deployable example.
|
|
244
|
+
|
|
245
|
+
Reflag maintains a cached set of feature definitions in the memory of your worker which it uses to decide which features to turn on for which users/companies.
|
|
246
|
+
|
|
247
|
+
The SDK caches feature definitions in memory for fast performance. The first request to a new worker instance fetches definitions from Reflag's servers, while subsequent requests use the cache. When the cache expires, it's updated in the background. `ctx.waitUntil(reflag.flush())` ensures completion of the background work, so response times are not affected. This background work may increase wall-clock time for your worker, but it will not measurably increase billable CPU time on platforms like Cloudflare.
|
|
248
|
+
|
|
249
|
+
## Error Handling
|
|
250
|
+
|
|
251
|
+
The SDK is designed to fail gracefully and never throw exceptions to the caller. Instead, it logs errors and provides
|
|
252
|
+
fallback behavior:
|
|
253
|
+
|
|
254
|
+
1. **Flag Evaluation Failures**:
|
|
255
|
+
|
|
256
|
+
```typescript
|
|
257
|
+
const { isEnabled } = client.getFlag("my-feature");
|
|
258
|
+
// If feature evaluation fails, isEnabled will be false
|
|
259
|
+
```
|
|
260
|
+
|
|
261
|
+
2. **Network Errors**:
|
|
262
|
+
|
|
263
|
+
```typescript
|
|
264
|
+
// Network errors during tracking are logged but don't affect your application
|
|
265
|
+
const { track } = client.getFlag("my-feature");
|
|
266
|
+
if (isEnabled) {
|
|
267
|
+
try {
|
|
268
|
+
await track();
|
|
269
|
+
} catch (error) {
|
|
270
|
+
// The SDK already logged this error
|
|
271
|
+
// Your application can continue normally
|
|
272
|
+
}
|
|
273
|
+
}
|
|
274
|
+
```
|
|
275
|
+
|
|
276
|
+
3. **Missing Context**:
|
|
277
|
+
|
|
278
|
+
```typescript
|
|
279
|
+
// The SDK tracks missing context fields but continues operation
|
|
280
|
+
const features = client.getFlags({
|
|
281
|
+
user: { id: "user123" },
|
|
282
|
+
// Missing company context will be logged but won't cause errors
|
|
283
|
+
});
|
|
284
|
+
```
|
|
285
|
+
|
|
286
|
+
4. **Offline Mode**:
|
|
287
|
+
|
|
288
|
+
```typescript
|
|
289
|
+
// In offline mode, the SDK uses feature overrides
|
|
290
|
+
const client = new ReflagClient({
|
|
291
|
+
offline: true,
|
|
292
|
+
featureOverrides: () => ({
|
|
293
|
+
"my-feature": true,
|
|
294
|
+
}),
|
|
295
|
+
});
|
|
296
|
+
```
|
|
297
|
+
|
|
298
|
+
The SDK logs all errors with appropriate severity levels. You can customize logging by providing your own logger:
|
|
299
|
+
|
|
300
|
+
```typescript
|
|
301
|
+
const client = new ReflagClient({
|
|
302
|
+
logger: {
|
|
303
|
+
debug: (msg) => console.debug(msg),
|
|
304
|
+
info: (msg) => console.info(msg),
|
|
305
|
+
warn: (msg) => console.warn(msg),
|
|
306
|
+
error: (msg, error) => {
|
|
307
|
+
console.error(msg, error);
|
|
308
|
+
// Send to your error tracking service
|
|
309
|
+
errorTracker.capture(error);
|
|
310
|
+
},
|
|
311
|
+
},
|
|
312
|
+
});
|
|
313
|
+
```
|
|
314
|
+
|
|
315
|
+
## Remote config
|
|
316
|
+
|
|
317
|
+
Remote config is a dynamic and flexible approach to configuring feature behavior outside of your app – without needing to re-deploy it.
|
|
318
|
+
|
|
319
|
+
Similar to `isEnabled`, each feature has a `config` property. This configuration is managed from within Reflag.
|
|
320
|
+
It is managed similar to the way access to features is managed, but instead of the binary `isEnabled` you can have
|
|
321
|
+
multiple configuration values which are given to different user/companies.
|
|
322
|
+
|
|
323
|
+
```ts
|
|
324
|
+
const features = reflagClient.getFlags();
|
|
325
|
+
// {
|
|
326
|
+
// huddle: {
|
|
327
|
+
// isEnabled: true,
|
|
328
|
+
// targetingVersion: 42,
|
|
329
|
+
// config: {
|
|
330
|
+
// key: "gpt-3.5",
|
|
331
|
+
// payload: { maxTokens: 10000, model: "gpt-3.5-beta1" }
|
|
332
|
+
// }
|
|
333
|
+
// }
|
|
334
|
+
// }
|
|
335
|
+
```
|
|
336
|
+
|
|
337
|
+
`key` is mandatory for a config, but if a feature has no config or no config value was matched against the context, the `key` will be `undefined`. Make sure to check against this case when trying to use the configuration in your application. `payload` is an optional JSON value for arbitrary configuration needs.
|
|
338
|
+
|
|
339
|
+
Just as `isEnabled`, accessing `config` on the object returned by `getFlags` does not automatically
|
|
340
|
+
generate a `check` event, contrary to the `config` property on the object returned by `getFlag`.
|
|
341
|
+
|
|
342
|
+
## Configuring
|
|
343
|
+
|
|
344
|
+
The Reflag `Node.js` SDK can be configured through environment variables,
|
|
345
|
+
a configuration file on disk or by passing options to the `ReflagClient`
|
|
346
|
+
constructor. By default, the SDK searches for `reflag.config.json` in the
|
|
347
|
+
current working directory.
|
|
348
|
+
|
|
349
|
+
| Option | Type | Description | Env Var |
|
|
350
|
+
| ------------------ | ----------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------- |
|
|
351
|
+
| `secretKey` | string | The secret key used for authentication with Reflag's servers. | REFLAG_SECRET_KEY |
|
|
352
|
+
| `logLevel` | string | The log level for the SDK (e.g., `"DEBUG"`, `"INFO"`, `"WARN"`, `"ERROR"`). Default: `INFO` | REFLAG_LOG_LEVEL |
|
|
353
|
+
| `offline` | boolean | Operate in offline mode. Default: `false`, except in tests it will default to `true` based off of the `TEST` env. var. | REFLAG_OFFLINE |
|
|
354
|
+
| `apiBaseUrl` | string | The base API URL for the Reflag servers. | REFLAG_API_BASE_URL |
|
|
355
|
+
| `featureOverrides` | Record<string, boolean> | An object specifying feature overrides for testing or local development. See [examples/express/app.test.ts](https://github.com/reflagcom/javascript/tree/main/packages/node-sdk/examples/express/app.test.ts) for how to use `featureOverrides` in tests. | REFLAG_FLAGS_ENABLED, REFLAG_FLAGS_DISABLED |
|
|
356
|
+
| `configFile` | string | Load this config file from disk. Default: `reflag.config.json` | REFLAG_CONFIG_FILE |
|
|
357
|
+
|
|
358
|
+
> [!NOTE] > `REFLAG_FLAGS_ENABLED` and `REFLAG_FLAGS_DISABLED` are comma separated lists of features which will be enabled or disabled respectively.
|
|
359
|
+
|
|
360
|
+
`reflag.config.json` example:
|
|
361
|
+
|
|
362
|
+
```json
|
|
363
|
+
{
|
|
364
|
+
"secretKey": "...",
|
|
365
|
+
"logLevel": "warn",
|
|
366
|
+
"offline": true,
|
|
367
|
+
"apiBaseUrl": "https://proxy.slick-demo.com",
|
|
368
|
+
"flagOverrides": {
|
|
369
|
+
"huddles": true,
|
|
370
|
+
"voiceChat": { "isEnabled": false },
|
|
371
|
+
"aiAssist": {
|
|
372
|
+
"isEnabled": true,
|
|
373
|
+
"config": {
|
|
374
|
+
"key": "gpt-4.0",
|
|
375
|
+
"payload": {
|
|
376
|
+
"maxTokens": 50000
|
|
377
|
+
}
|
|
378
|
+
}
|
|
379
|
+
}
|
|
380
|
+
}
|
|
381
|
+
}
|
|
382
|
+
```
|
|
383
|
+
|
|
384
|
+
When using a `reflag.config.json` for local development, make sure you add it to your
|
|
385
|
+
`.gitignore` file. You can also set these options directly in the `ReflagClient`
|
|
386
|
+
constructor. The precedence for configuration options is as follows, listed in the
|
|
387
|
+
order of importance:
|
|
388
|
+
|
|
389
|
+
1. Options passed along to the constructor directly,
|
|
390
|
+
2. Environment variable,
|
|
391
|
+
3. The config file.
|
|
392
|
+
|
|
393
|
+
## Type safe flags
|
|
394
|
+
|
|
395
|
+
To get type checked flags, install the Reflag CLI:
|
|
396
|
+
|
|
397
|
+
```
|
|
398
|
+
npm i --save-dev @reflag/cli
|
|
399
|
+
```
|
|
400
|
+
|
|
401
|
+
then generate the types:
|
|
402
|
+
|
|
403
|
+
```
|
|
404
|
+
npx reflag features types
|
|
405
|
+
```
|
|
406
|
+
|
|
407
|
+
This will generate a `reflag.d.ts` containing all your features.
|
|
408
|
+
Any feature look ups will now be checked against the features that exist in Reflag.
|
|
409
|
+
|
|
410
|
+
Here's an example of a failed type check:
|
|
411
|
+
|
|
412
|
+
```typescript
|
|
413
|
+
import { ReflagClient } from "@reflag/node-sdk";
|
|
414
|
+
|
|
415
|
+
export const reflagClient = new ReflagClient();
|
|
416
|
+
|
|
417
|
+
reflagClient.initialize().then(() => {
|
|
418
|
+
console.log("Reflag initialized!");
|
|
419
|
+
|
|
420
|
+
// TypeScript will catch this error: "invalid-feature" doesn't exist
|
|
421
|
+
reflagClient.getFlag("invalid-feature");
|
|
422
|
+
|
|
423
|
+
const {
|
|
424
|
+
isEnabled,
|
|
425
|
+
config: { payload },
|
|
426
|
+
} = reflagClient.getFlag("create-todos");
|
|
427
|
+
});
|
|
428
|
+
```
|
|
429
|
+
|
|
430
|
+

|
|
431
|
+
|
|
432
|
+
This is an example of a failed config payload check:
|
|
433
|
+
|
|
434
|
+
```typescript
|
|
435
|
+
reflagClient.initialize().then(() => {
|
|
436
|
+
// TypeScript will catch this error as well: "minLength" is not part of the payload.
|
|
437
|
+
if (isEnabled && todo.length > config.payload.minLength) {
|
|
438
|
+
// ...
|
|
439
|
+
}
|
|
440
|
+
});
|
|
441
|
+
```
|
|
442
|
+
|
|
443
|
+

|
|
444
|
+
|
|
445
|
+
## Testing
|
|
446
|
+
|
|
447
|
+
When writing tests that cover code with flags, you can toggle features on/off programmatically to test the different behavior.
|
|
448
|
+
|
|
449
|
+
`reflag.ts`:
|
|
450
|
+
|
|
451
|
+
```typescript
|
|
452
|
+
import { ReflagClient } from "@reflag/node-sdk";
|
|
453
|
+
|
|
454
|
+
export const reflag = new ReflagClient();
|
|
455
|
+
```
|
|
456
|
+
|
|
457
|
+
`app.test.ts`:
|
|
458
|
+
|
|
459
|
+
```typescript
|
|
460
|
+
import { reflag } from "./reflag.ts";
|
|
461
|
+
|
|
462
|
+
beforeAll(async () => await reflag.initialize());
|
|
463
|
+
afterEach(() => {
|
|
464
|
+
reflag.clearFlagOverrides();
|
|
465
|
+
});
|
|
466
|
+
|
|
467
|
+
describe("API Tests", () => {
|
|
468
|
+
it("should return 200 for the root endpoint", async () => {
|
|
469
|
+
reflag.featureOverrides = {
|
|
470
|
+
"show-todo": true,
|
|
471
|
+
};
|
|
472
|
+
|
|
473
|
+
const response = await request(app).get("/");
|
|
474
|
+
expect(response.status).toBe(200);
|
|
475
|
+
expect(response.body).toEqual({ message: "Ready to manage some TODOs!" });
|
|
476
|
+
});
|
|
477
|
+
});
|
|
478
|
+
```
|
|
479
|
+
|
|
480
|
+
See more on feature overrides in the section below.
|
|
481
|
+
|
|
482
|
+
## Flag Overrides
|
|
483
|
+
|
|
484
|
+
Flag overrides allow you to override flags and their configurations locally. This is particularly useful for development and testing. You can specify overrides in three ways:
|
|
485
|
+
|
|
486
|
+
1. Through environment variables:
|
|
487
|
+
|
|
488
|
+
```bash
|
|
489
|
+
REFLAG_FLAGS_ENABLED=feature1,feature2
|
|
490
|
+
REFLAG_FLAGS_DISABLED=feature3,feature4
|
|
491
|
+
```
|
|
492
|
+
|
|
493
|
+
2. Through `reflag.config.json`:
|
|
494
|
+
|
|
495
|
+
```json
|
|
496
|
+
{
|
|
497
|
+
"flagOverrides": {
|
|
498
|
+
"delete-todos": {
|
|
499
|
+
"isEnabled": true,
|
|
500
|
+
"config": {
|
|
501
|
+
"key": "dev-config",
|
|
502
|
+
"payload": {
|
|
503
|
+
"requireConfirmation": true,
|
|
504
|
+
"maxDeletionsPerDay": 5
|
|
505
|
+
}
|
|
506
|
+
}
|
|
507
|
+
}
|
|
508
|
+
}
|
|
509
|
+
}
|
|
510
|
+
```
|
|
511
|
+
|
|
512
|
+
3. Programmatically through the client options:
|
|
513
|
+
|
|
514
|
+
You can use a simple `Record<string, boolean>` and pass it either in the constructor or by setting `client.featureOverrides`:
|
|
515
|
+
|
|
516
|
+
```typescript
|
|
517
|
+
// pass directly in the constructor
|
|
518
|
+
const client = new ReflagClient({ featureOverrides: { myFlag: true } });
|
|
519
|
+
// or set on the client at a later time
|
|
520
|
+
client.featureOverrides = { myFlag: false };
|
|
521
|
+
|
|
522
|
+
// clear feature overrides. Same as setting to {}.
|
|
523
|
+
client.clearFlagOverrides();
|
|
524
|
+
```
|
|
525
|
+
|
|
526
|
+
To get dynamic overrides, use a function which takes a context and returns a boolean or an object with the shape of `{isEnabled, config}`:
|
|
527
|
+
|
|
528
|
+
```typescript
|
|
529
|
+
import { ReflagClient, Context } from "@reflag/node-sdk";
|
|
530
|
+
|
|
531
|
+
const featureOverrides = (context: Context) => ({
|
|
532
|
+
"delete-todos": {
|
|
533
|
+
isEnabled: true,
|
|
534
|
+
config: {
|
|
535
|
+
key: "dev-config",
|
|
536
|
+
payload: {
|
|
537
|
+
requireConfirmation: true,
|
|
538
|
+
maxDeletionsPerDay: 5,
|
|
539
|
+
},
|
|
540
|
+
},
|
|
541
|
+
},
|
|
542
|
+
});
|
|
543
|
+
|
|
544
|
+
const client = new ReflagClient({
|
|
545
|
+
featureOverrides,
|
|
546
|
+
});
|
|
547
|
+
```
|
|
548
|
+
|
|
549
|
+
## Remote Flag Evaluation
|
|
550
|
+
|
|
551
|
+
In addition to local feature evaluation, Reflag supports remote evaluation using stored context. This is useful when you want to evaluate features using user/company attributes that were previously sent to Reflag:
|
|
552
|
+
|
|
553
|
+
```typescript
|
|
554
|
+
// First, update user and company attributes
|
|
555
|
+
await client.updateUser("user123", {
|
|
556
|
+
attributes: {
|
|
557
|
+
role: "admin",
|
|
558
|
+
subscription: "premium",
|
|
559
|
+
},
|
|
560
|
+
});
|
|
561
|
+
|
|
562
|
+
await client.updateCompany("company456", {
|
|
563
|
+
attributes: {
|
|
564
|
+
tier: "enterprise",
|
|
565
|
+
employees: 1000,
|
|
566
|
+
},
|
|
567
|
+
});
|
|
568
|
+
|
|
569
|
+
// Later, evaluate features remotely using stored context
|
|
570
|
+
const features = await client.getFlagsRemote("company456", "user123");
|
|
571
|
+
// Or evaluate a single feature
|
|
572
|
+
const feature = await client.getFlagRemote(
|
|
573
|
+
"create-todos",
|
|
574
|
+
"company456",
|
|
575
|
+
"user123",
|
|
576
|
+
);
|
|
577
|
+
|
|
578
|
+
// You can also provide additional context
|
|
579
|
+
const featuresWithContext = await client.getFlagsRemote(
|
|
580
|
+
"company456",
|
|
581
|
+
"user123",
|
|
582
|
+
{
|
|
583
|
+
other: {
|
|
584
|
+
location: "US",
|
|
585
|
+
platform: "mobile",
|
|
586
|
+
},
|
|
587
|
+
},
|
|
588
|
+
);
|
|
589
|
+
```
|
|
590
|
+
|
|
591
|
+
Remote evaluation is particularly useful when:
|
|
592
|
+
|
|
593
|
+
- You want to use the most up-to-date user/company attributes stored in Reflag
|
|
594
|
+
- You don't want to pass all context attributes with every evaluation
|
|
595
|
+
- You need to ensure consistent feature evaluation across different services
|
|
596
|
+
|
|
597
|
+
## Using with Express
|
|
598
|
+
|
|
599
|
+
A popular way to integrate the Reflag Node.js SDK is through an express middleware.
|
|
600
|
+
|
|
601
|
+
```typescript
|
|
602
|
+
import reflag from "./reflag";
|
|
603
|
+
import express from "express";
|
|
604
|
+
import { BoundReflagClient } from "@reflag/node-sdk";
|
|
605
|
+
|
|
606
|
+
// Augment the Express types to include a `boundReflagClient` property on the
|
|
607
|
+
// `res.locals` object.
|
|
608
|
+
// This will allow us to access the ReflagClient instance in our route handlers
|
|
609
|
+
// without having to pass it around manually
|
|
610
|
+
declare global {
|
|
611
|
+
namespace Express {
|
|
612
|
+
interface Locals {
|
|
613
|
+
boundReflagClient: BoundReflagClient;
|
|
614
|
+
}
|
|
615
|
+
}
|
|
616
|
+
}
|
|
617
|
+
|
|
618
|
+
// Add express middleware
|
|
619
|
+
app.use((req, res, next) => {
|
|
620
|
+
// Extract the user and company IDs from the request
|
|
621
|
+
// You'll want to use a proper authentication and identification
|
|
622
|
+
// mechanism in a real-world application
|
|
623
|
+
const user = {
|
|
624
|
+
id: req.user?.id,
|
|
625
|
+
name: req.user?.name
|
|
626
|
+
email: req.user?.email
|
|
627
|
+
}
|
|
628
|
+
|
|
629
|
+
const company = {
|
|
630
|
+
id: req.user?.companyId
|
|
631
|
+
name: req.user?.companyName
|
|
632
|
+
}
|
|
633
|
+
|
|
634
|
+
// Create a new BoundReflagClient instance by calling the `bindClient`
|
|
635
|
+
// method on a `ReflagClient` instance
|
|
636
|
+
// This will create a new instance that is bound to the user/company given.
|
|
637
|
+
const boundReflagClient = reflag.bindClient({ user, company });
|
|
638
|
+
|
|
639
|
+
// Store the BoundReflagClient instance in the `res.locals` object so we
|
|
640
|
+
// can access it in our route handlers
|
|
641
|
+
res.locals.boundReflagClient = boundReflagClient;
|
|
642
|
+
next();
|
|
643
|
+
});
|
|
644
|
+
|
|
645
|
+
// Now use res.locals.boundReflagClient in your handlers
|
|
646
|
+
app.get("/todos", async (_req, res) => {
|
|
647
|
+
const { track, isEnabled } = res.locals.reflagUser.getFlag("show-todos");
|
|
648
|
+
|
|
649
|
+
if (!isEnabled) {
|
|
650
|
+
res.status(403).send({"error": "flag inaccessible"})
|
|
651
|
+
return
|
|
652
|
+
}
|
|
653
|
+
|
|
654
|
+
...
|
|
655
|
+
}
|
|
656
|
+
```
|
|
657
|
+
|
|
658
|
+
See [examples/express/app.ts](https://github.com/reflagcom/javascript/tree/main/packages/node-sdk/example/express/app.ts) for a full example.
|
|
659
|
+
|
|
660
|
+
## Remote flag evaluation with stored context
|
|
661
|
+
|
|
662
|
+
If you don't want to provide context each time when evaluating flags but
|
|
663
|
+
rather you would like to utilize the attributes you sent to Reflag previously
|
|
664
|
+
(by calling `updateCompany` and `updateUser`) you can do so by calling `getFlagsRemote`
|
|
665
|
+
(or `getFlagRemote` for a specific feature) with providing just `userId` and `companyId`.
|
|
666
|
+
These methods will call Reflag's servers and flags will be evaluated remotely
|
|
667
|
+
using the stored attributes.
|
|
668
|
+
|
|
669
|
+
```typescript
|
|
670
|
+
// Update user and company attributes
|
|
671
|
+
client.updateUser("john_doe", {
|
|
672
|
+
attributes: {
|
|
673
|
+
name: "John O.",
|
|
674
|
+
role: "admin",
|
|
675
|
+
},
|
|
676
|
+
});
|
|
677
|
+
|
|
678
|
+
client.updateCompany("acme_inc", {
|
|
679
|
+
attributes: {
|
|
680
|
+
name: "Acme, Inc",
|
|
681
|
+
tier: "premium"
|
|
682
|
+
},
|
|
683
|
+
});
|
|
684
|
+
...
|
|
685
|
+
|
|
686
|
+
// This will evaluate flags with respecting the attributes sent previously
|
|
687
|
+
const features = await client.getFlagsRemote("acme_inc", "john_doe");
|
|
688
|
+
```
|
|
689
|
+
|
|
690
|
+
> [!IMPORTANT]
|
|
691
|
+
> User and company attribute updates are processed asynchronously, so there might
|
|
692
|
+
> be a small delay between when attributes are updated and when they are available
|
|
693
|
+
> for evaluation.
|
|
694
|
+
|
|
695
|
+
## Opting out of tracking
|
|
696
|
+
|
|
697
|
+
There are use cases in which you not want to be sending `user`, `company` and
|
|
698
|
+
`track` events to [Reflag.com](https://reflag.com). These are usually cases where you could be impersonating
|
|
699
|
+
another user in the system and do not want to interfere with the data being
|
|
700
|
+
collected by Reflag.
|
|
701
|
+
|
|
702
|
+
To disable tracking, bind the client using `bindClient()` as follows:
|
|
703
|
+
|
|
704
|
+
```typescript
|
|
705
|
+
// binds the client to a given user and company and set `enableTracking` to `false`.
|
|
706
|
+
const boundClient = client.bindClient({ user, company, enableTracking: false });
|
|
707
|
+
|
|
708
|
+
boundClient.track("some event"); // this will not actually send the event to Reflag.
|
|
709
|
+
|
|
710
|
+
// the following code will not update the `user` nor `company` in Reflag and will
|
|
711
|
+
// not send `track` events either.
|
|
712
|
+
const { isEnabled, track } = boundClient.getFlag("user-menu");
|
|
713
|
+
if (isEnabled) {
|
|
714
|
+
track();
|
|
715
|
+
}
|
|
716
|
+
```
|
|
717
|
+
|
|
718
|
+
Another way way to disable tracking without employing a bound client is to call `getFlag()`
|
|
719
|
+
or `getFlags()` by supplying `enableTracking: false` in the arguments passed to
|
|
720
|
+
these functions.
|
|
721
|
+
|
|
722
|
+
> [!IMPORTANT]
|
|
723
|
+
> Note, however, that calling `track()`, `updateCompany()` or `updateUser()` in the `ReflagClient`
|
|
724
|
+
> will still send tracking data. As such, it is always recommended to use `bindClient()`
|
|
725
|
+
> when using this SDK.
|
|
726
|
+
|
|
727
|
+
## Flushing
|
|
728
|
+
|
|
729
|
+
ReflagClient employs a batching technique to minimize the number of calls that are sent to
|
|
730
|
+
Reflag's servers.
|
|
731
|
+
|
|
732
|
+
By default, the SDK automatically subscribes to process exit signals and attempts to flush
|
|
733
|
+
any pending events. This behavior is controlled by the `flushOnExit` option in the client configuration:
|
|
734
|
+
|
|
735
|
+
```typescript
|
|
736
|
+
const client = new ReflagClient({
|
|
737
|
+
batchOptions: {
|
|
738
|
+
flushOnExit: false, // disable automatic flushing on exit
|
|
739
|
+
},
|
|
740
|
+
});
|
|
741
|
+
```
|
|
742
|
+
|
|
743
|
+
## Tracking custom events and setting custom attributes
|
|
744
|
+
|
|
745
|
+
Tracking allows events and updating user/company attributes in Reflag.
|
|
746
|
+
For example, if a customer changes their plan, you'll want Reflag to know about it,
|
|
747
|
+
in order to continue to provide up-do-date targeting information in the Reflag interface.
|
|
748
|
+
|
|
749
|
+
The following example shows how to register a new user, associate it with a company
|
|
750
|
+
and finally update the plan they are on.
|
|
751
|
+
|
|
752
|
+
```typescript
|
|
753
|
+
// registers the user with Reflag using the provided unique ID, and
|
|
754
|
+
// providing a set of custom attributes (can be anything)
|
|
755
|
+
client.updateUser("user_id", {
|
|
756
|
+
attributes: { longTimeUser: true, payingCustomer: false },
|
|
757
|
+
});
|
|
758
|
+
client.updateCompany("company_id", { userId: "user_id" });
|
|
759
|
+
|
|
760
|
+
// the user started a voice huddle
|
|
761
|
+
client.track("user_id", "huddle", { attributes: { voice: true } });
|
|
762
|
+
```
|
|
763
|
+
|
|
764
|
+
It's also possible to achieve the same through a bound client in the following manner:
|
|
765
|
+
|
|
766
|
+
```typescript
|
|
767
|
+
const boundClient = client.bindClient({
|
|
768
|
+
user: { id: "user_id", longTimeUser: true, payingCustomer: false },
|
|
769
|
+
company: { id: "company_id" },
|
|
770
|
+
});
|
|
771
|
+
|
|
772
|
+
boundClient.track("huddle", { attributes: { voice: true } });
|
|
773
|
+
```
|
|
774
|
+
|
|
775
|
+
Some attributes are used by Reflag to improve the UI, and are recommended
|
|
776
|
+
to provide for easier navigation:
|
|
777
|
+
|
|
778
|
+
- `name` -- display name for `user`/`company`,
|
|
779
|
+
- `email` -- the email of the user,
|
|
780
|
+
- `avatar` -- the URL for `user`/`company` avatar image.
|
|
781
|
+
|
|
782
|
+
Attributes cannot be nested (multiple levels) and must be either strings,
|
|
783
|
+
integers or booleans.
|
|
784
|
+
|
|
785
|
+
## Managing `Last seen`
|
|
786
|
+
|
|
787
|
+
By default `updateUser`/`updateCompany` calls automatically update the given
|
|
788
|
+
user/company `Last seen` property on Reflag servers.
|
|
789
|
+
|
|
790
|
+
You can control if `Last seen` should be updated when the events are sent by setting
|
|
791
|
+
`meta.active = false`. This is often useful if you
|
|
792
|
+
have a background job that goes through a set of companies just to update their
|
|
793
|
+
attributes but not their activity.
|
|
794
|
+
|
|
795
|
+
Example:
|
|
796
|
+
|
|
797
|
+
```typescript
|
|
798
|
+
client.updateUser("john_doe", {
|
|
799
|
+
attributes: { name: "John O." },
|
|
800
|
+
meta: { active: true },
|
|
801
|
+
});
|
|
802
|
+
|
|
803
|
+
client.updateCompany("acme_inc", {
|
|
804
|
+
attributes: { name: "Acme, Inc" },
|
|
805
|
+
meta: { active: false },
|
|
806
|
+
});
|
|
807
|
+
```
|
|
808
|
+
|
|
809
|
+
`bindClient()` updates attributes on the Reflag servers but does not automatically
|
|
810
|
+
update `Last seen`.
|
|
811
|
+
|
|
812
|
+
## Zero PII
|
|
813
|
+
|
|
814
|
+
The Reflag SDK doesn't collect any metadata and HTTP IP addresses are _not_ being
|
|
815
|
+
stored. For tracking individual users, we recommend using something like database
|
|
816
|
+
ID as userId, as it's unique and doesn't include any PII (personal identifiable
|
|
817
|
+
information). If, however, you're using e.g. email address as userId, but prefer
|
|
818
|
+
not to send any PII to Reflag, you can hash the sensitive data before sending
|
|
819
|
+
it to Reflag:
|
|
820
|
+
|
|
821
|
+
```typescript
|
|
822
|
+
import { sha256 } from 'crypto-hash';
|
|
823
|
+
|
|
824
|
+
client.updateUser({ userId: await sha256("john_doe"), ... });
|
|
825
|
+
```
|
|
826
|
+
|
|
827
|
+
## Typescript
|
|
828
|
+
|
|
829
|
+
Types are bundled together with the library and exposed automatically when importing
|
|
830
|
+
through a package manager.
|
|
831
|
+
|
|
832
|
+
## License
|
|
833
|
+
|
|
834
|
+
> MIT License
|
|
835
|
+
> Copyright (c) 2025 Bucket ApS
|