@heliosgraphics/utils 6.0.0-alpha.17 → 6.0.0-alpha.19

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/package.json +2 -2
  2. package/sanitize.ts +15 -4
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@heliosgraphics/utils",
3
- "version": "6.0.0-alpha.17",
3
+ "version": "6.0.0-alpha.19",
4
4
  "type": "module",
5
5
  "sideEffects": false,
6
6
  "author": "Chris Puska <chris@puska.org>",
@@ -27,6 +27,6 @@
27
27
  "url": "https://github.com/heliosgraphics/helios-ui/issues"
28
28
  },
29
29
  "devDependencies": {
30
- "@types/node": "^25"
30
+ "@types/node": "^26"
31
31
  }
32
32
  }
package/sanitize.ts CHANGED
@@ -25,6 +25,9 @@ const ALLOWED_TAGS = new Set([
25
25
  const ALLOWED_ATTRIBUTES = new Set(["class", "id", "title"])
26
26
 
27
27
  const SAFE_PROTOCOLS = /^(https?|mailto|tel|ftp):/i
28
+ const UNSAFE_PROTOCOLS = new Set(["data", "javascript", "livescript", "mocha", "vbscript"])
29
+ const EMOJI_CLUSTER_LEFT_PATTERN: RegExp = /\p{Extended_Pictographic}(?:\uFE0F|[\u{1F3FB}-\u{1F3FF}])*$/u
30
+ const EMOJI_CLUSTER_RIGHT_PATTERN: RegExp = /^\p{Extended_Pictographic}/u
28
31
 
29
32
  interface SanitizeTextOptions {
30
33
  allowLinks?: boolean
@@ -85,6 +88,15 @@ const DANGEROUS_FUNCTIONS: Array<RegExp> = [
85
88
  /createElement\s*\(\s*["']script["']\s*\)/gi,
86
89
  ]
87
90
 
91
+ const stripUnsafeZeroWidthCharacter = (character: string, index: number, text: string): string => {
92
+ if (character !== "\u200D") return ""
93
+
94
+ const leftText: string = text.slice(0, index)
95
+ const rightText: string = text.slice(index + character.length)
96
+
97
+ return EMOJI_CLUSTER_LEFT_PATTERN.test(leftText) && EMOJI_CLUSTER_RIGHT_PATTERN.test(rightText) ? character : ""
98
+ }
99
+
88
100
  export const sanitizeText = (input: string = "", options: SanitizeTextOptions = {}): string => {
89
101
  if (!input) return ""
90
102
 
@@ -95,7 +107,6 @@ export const sanitizeText = (input: string = "", options: SanitizeTextOptions =
95
107
  .replace(/<!--[\s\S]*?-->/g, "")
96
108
  .replace(/\\[0-9a-f]{1,6}\s*/gi, "")
97
109
  .replace(/<img[^>]*src\s*=\s*["']data:[^"']*["'][^>]*>/gi, "")
98
- .replace(/data\s*:[^"'\s>)]*/gi, "")
99
110
 
100
111
  // Then decode entities
101
112
  sanitized = sanitized
@@ -140,7 +151,7 @@ export const sanitizeText = (input: string = "", options: SanitizeTextOptions =
140
151
  }
141
152
 
142
153
  sanitized = sanitized
143
- .replace(/[\u200B-\u200D\uFEFF]/g, "")
154
+ .replace(/[\u200B-\u200D\uFEFF]/g, stripUnsafeZeroWidthCharacter)
144
155
  // eslint-disable-next-line no-control-regex
145
156
  .replace(/[\u0000-\u0008\u000B\u000C\u000E-\u001F\u007F-\u009F]/g, "")
146
157
  .replace(/[\u202A-\u202E]/g, "")
@@ -154,11 +165,11 @@ export const sanitizeText = (input: string = "", options: SanitizeTextOptions =
154
165
  sanitized = sanitized.replace(/<style\b[^>]*>[\s\S]*$/gi, "")
155
166
 
156
167
  sanitized = sanitized.replace(
157
- /(\w+(?:[\s\u00A0]*\w)*)\s*(?:&#58;|&#x3a;|&colon;|%3a|%3A|\\:|:)/gi,
168
+ /(\w+(?:[\s\u00A0]*\w)*)\s*(?:&#58;|&#x3a;|&colon;|%3a|%3A|\\:|:)(?=\S)/gi,
158
169
  (_: string, protocol: string) => {
159
170
  const cleanProtocol = protocol.replace(/[\s\u00A0]/g, "").toLowerCase()
160
171
  const testString = cleanProtocol + ":"
161
- return SAFE_PROTOCOLS.test(testString) ? testString : ""
172
+ return SAFE_PROTOCOLS.test(testString) || !UNSAFE_PROTOCOLS.has(cleanProtocol) ? testString : ""
162
173
  },
163
174
  )
164
175