@the-convocation/twitter-scraper 0.18.3 → 0.19.0
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/README.md +35 -0
- package/dist/cycletls/cjs/index.cjs +99 -0
- package/dist/cycletls/cjs/index.cjs.map +1 -0
- package/dist/cycletls/esm/index.mjs +96 -0
- package/dist/cycletls/esm/index.mjs.map +1 -0
- package/dist/cycletls/index.d.ts +11 -0
- package/dist/default/cjs/index.js +75 -18
- package/dist/default/cjs/index.js.map +1 -1
- package/dist/default/esm/index.mjs +75 -18
- package/dist/default/esm/index.mjs.map +1 -1
- package/dist/node/cjs/index.cjs +75 -18
- package/dist/node/cjs/index.cjs.map +1 -1
- package/dist/node/esm/index.mjs +75 -18
- package/dist/node/esm/index.mjs.map +1 -1
- package/examples/cycletls/README.md +54 -0
- package/examples/cycletls/package.json +12 -0
- package/package.json +25 -9
- package/rollup.config.mjs +34 -0
|
@@ -0,0 +1,54 @@
|
|
|
1
|
+
# CycleTLS Cloudflare Bypass Example
|
|
2
|
+
|
|
3
|
+
This example demonstrates how to use the `@the-convocation/twitter-scraper/cycletls` entrypoint to bypass Cloudflare bot detection when authenticating with Twitter.
|
|
4
|
+
|
|
5
|
+
## Problem
|
|
6
|
+
|
|
7
|
+
Twitter's authentication endpoints may be protected by Cloudflare's advanced bot detection, which analyzes TLS fingerprints to detect non-browser clients. Standard Node.js TLS handshakes can trigger `403 Forbidden` errors during login.
|
|
8
|
+
|
|
9
|
+
## Solution
|
|
10
|
+
|
|
11
|
+
This example uses [CycleTLS](https://github.com/Danny-Dasilva/CycleTLS), which leverages golang to mimic Chrome browser TLS fingerprints, allowing requests to pass through Cloudflare's protection.
|
|
12
|
+
|
|
13
|
+
## Installation
|
|
14
|
+
|
|
15
|
+
```sh
|
|
16
|
+
yarn install
|
|
17
|
+
```
|
|
18
|
+
|
|
19
|
+
## Configuration
|
|
20
|
+
|
|
21
|
+
Create a `.env` file in the root of the repository (two levels up) with your Twitter credentials:
|
|
22
|
+
|
|
23
|
+
```
|
|
24
|
+
TWITTER_USERNAME=your_username
|
|
25
|
+
TWITTER_PASSWORD=your_password
|
|
26
|
+
TWITTER_EMAIL=your_email
|
|
27
|
+
```
|
|
28
|
+
|
|
29
|
+
## Usage
|
|
30
|
+
|
|
31
|
+
```sh
|
|
32
|
+
yarn start
|
|
33
|
+
```
|
|
34
|
+
|
|
35
|
+
## How it works
|
|
36
|
+
|
|
37
|
+
The example imports the `cycleTLSFetch` function from the `/cycletls` subpath:
|
|
38
|
+
|
|
39
|
+
```ts
|
|
40
|
+
import { Scraper } from '@the-convocation/twitter-scraper';
|
|
41
|
+
import { cycleTLSFetch, cycleTLSExit } from '@the-convocation/twitter-scraper/cycletls';
|
|
42
|
+
|
|
43
|
+
const scraper = new Scraper({
|
|
44
|
+
fetch: cycleTLSFetch as any,
|
|
45
|
+
});
|
|
46
|
+
```
|
|
47
|
+
|
|
48
|
+
This replaces the default fetch implementation with one that uses Chrome-like TLS fingerprints, bypassing Cloudflare's detection.
|
|
49
|
+
|
|
50
|
+
## Important Notes
|
|
51
|
+
|
|
52
|
+
- **Node.js only**: The `/cycletls` entrypoint requires Node.js and will not work in browsers
|
|
53
|
+
- **Cleanup required**: Always call `cycleTLSExit()` when done to cleanup golang resources
|
|
54
|
+
- **Optional dependency**: `cycletls` must be explicitly installed alongside the main package
|
package/package.json
CHANGED
|
@@ -7,18 +7,25 @@
|
|
|
7
7
|
"scraper",
|
|
8
8
|
"crawler"
|
|
9
9
|
],
|
|
10
|
-
"version": "0.
|
|
10
|
+
"version": "0.19.0",
|
|
11
11
|
"main": "dist/default/cjs/index.js",
|
|
12
12
|
"types": "./dist/types/index.d.ts",
|
|
13
13
|
"exports": {
|
|
14
|
-
"
|
|
15
|
-
|
|
16
|
-
"
|
|
17
|
-
|
|
14
|
+
".": {
|
|
15
|
+
"types": "./dist/types/index.d.ts",
|
|
16
|
+
"node": {
|
|
17
|
+
"import": "./dist/node/esm/index.mjs",
|
|
18
|
+
"require": "./dist/node/cjs/index.cjs"
|
|
19
|
+
},
|
|
20
|
+
"default": {
|
|
21
|
+
"import": "./dist/default/esm/index.mjs",
|
|
22
|
+
"require": "./dist/default/cjs/index.js"
|
|
23
|
+
}
|
|
18
24
|
},
|
|
19
|
-
"
|
|
20
|
-
"
|
|
21
|
-
"
|
|
25
|
+
"./cycletls": {
|
|
26
|
+
"types": "./dist/cycletls/index.d.ts",
|
|
27
|
+
"import": "./dist/cycletls/esm/index.mjs",
|
|
28
|
+
"require": "./dist/cycletls/cjs/index.cjs"
|
|
22
29
|
}
|
|
23
30
|
},
|
|
24
31
|
"repository": "https://github.com/the-convocation/twitter-scraper.git",
|
|
@@ -33,7 +40,7 @@
|
|
|
33
40
|
"commit": "cz",
|
|
34
41
|
"docs:generate": "typedoc --options typedoc.json",
|
|
35
42
|
"docs:deploy": "yarn docs:generate && gh-pages -d docs",
|
|
36
|
-
"format": "prettier --write src/**/*.ts",
|
|
43
|
+
"format": "prettier --write src/**/*.ts examples/**/*.{ts,js,tsx}",
|
|
37
44
|
"prepare": "husky install",
|
|
38
45
|
"test": "jest"
|
|
39
46
|
},
|
|
@@ -48,6 +55,14 @@
|
|
|
48
55
|
"tough-cookie": "^4.1.2",
|
|
49
56
|
"tslib": "^2.5.2"
|
|
50
57
|
},
|
|
58
|
+
"peerDependencies": {
|
|
59
|
+
"cycletls": "^2.0.4"
|
|
60
|
+
},
|
|
61
|
+
"peerDependenciesMeta": {
|
|
62
|
+
"cycletls": {
|
|
63
|
+
"optional": true
|
|
64
|
+
}
|
|
65
|
+
},
|
|
51
66
|
"devDependencies": {
|
|
52
67
|
"@commitlint/cli": "^17.6.3",
|
|
53
68
|
"@commitlint/config-conventional": "^17.6.3",
|
|
@@ -59,6 +74,7 @@
|
|
|
59
74
|
"@types/tough-cookie": "^4.0.2",
|
|
60
75
|
"@typescript-eslint/eslint-plugin": "^5.59.7",
|
|
61
76
|
"@typescript-eslint/parser": "^5.59.7",
|
|
77
|
+
"cycletls": "^2.0.4",
|
|
62
78
|
"cz-conventional-changelog": "^3.3.0",
|
|
63
79
|
"dotenv": "^16.3.1",
|
|
64
80
|
"esbuild": "^0.21.5",
|
package/rollup.config.mjs
CHANGED
|
@@ -58,4 +58,38 @@ export default [
|
|
|
58
58
|
format: 'es',
|
|
59
59
|
},
|
|
60
60
|
},
|
|
61
|
+
// CycleTLS build (Node-only)
|
|
62
|
+
{
|
|
63
|
+
input: 'src/_cycletls.ts',
|
|
64
|
+
external: ['cycletls'],
|
|
65
|
+
plugins: [
|
|
66
|
+
esbuild({
|
|
67
|
+
define: {
|
|
68
|
+
PLATFORM_NODE: 'true',
|
|
69
|
+
PLATFORM_NODE_JEST: 'false',
|
|
70
|
+
},
|
|
71
|
+
}),
|
|
72
|
+
],
|
|
73
|
+
output: [
|
|
74
|
+
{
|
|
75
|
+
file: 'dist/cycletls/cjs/index.cjs',
|
|
76
|
+
format: 'cjs',
|
|
77
|
+
sourcemap: true,
|
|
78
|
+
},
|
|
79
|
+
{
|
|
80
|
+
file: 'dist/cycletls/esm/index.mjs',
|
|
81
|
+
format: 'es',
|
|
82
|
+
sourcemap: true,
|
|
83
|
+
},
|
|
84
|
+
],
|
|
85
|
+
},
|
|
86
|
+
{
|
|
87
|
+
input: 'src/_cycletls.ts',
|
|
88
|
+
external: ['cycletls'],
|
|
89
|
+
plugins: [dts()],
|
|
90
|
+
output: {
|
|
91
|
+
file: 'dist/cycletls/index.d.ts',
|
|
92
|
+
format: 'es',
|
|
93
|
+
},
|
|
94
|
+
},
|
|
61
95
|
];
|