v3.1.4 - Black Friday 🎁, Nov 28th, 2025

What's New in v3.1.4

🔧 Critical Fix: URL_DOMAIN_FIXES Processing Order

This release fixes a critical bug in URL_DOMAIN_FIXES processing that could cause domains with existing protocols to be incorrectly modified.

The Problem:
In v3.1.3, the regex pattern for domain fixes used a negative lookbehind ((?<!https?:\/\/)) which is not supported in ES5 runtime. While the script compiled without errors, this caused:

• Domains with existing https:// or http:// protocols to be processed incorrectly
• Potential duplicate protocol prefixes in some edge cases
• Unpredictable behavior when domains appeared multiple times in content

The Solution:
Replaced the negative lookbehind with a two-pattern approach that explicitly handles both cases:

1. Pattern 1 (runs FIRST): Protects already valid URLs

   • Matches: https://domain or http://domain
   • Action: Replaces with itself (no modification)
   • Purpose: Prevents double-processing of valid URLs

2. Pattern 2 (runs SECOND): Adds https:// to bare domains

   • Matches: Bare domains at word boundaries
   • Action: Adds https:// prefix
   • Purpose: Fixes domains missing protocol

Technical Implementation:

// Pattern 1: Protect valid URLs (https://domain or http://domain)
domainPatterns.push({
  pattern: "(https?:\\/\\/)" + escapedDomain + "(\\/[^\\s]*|[\\s]|$)",
  replacement: "$1" + domain + "$2",
  flags: "gm",
  literal: false
});

// Pattern 2: Add https:// to bare domains at boundaries
domainPatterns.push({
  pattern: "(^|[\\s\\(\\[\\{<\"'])" + escapedDomain + "(\\/[^\\s\\)\\]\\}>\"',;]*|[\\s\\)\\]\\}>\"',;]|$)",
  replacement: "$1https://" + domain + "$2",
  flags: "gm",
  literal: false
});

Key Features:

• ✅ ES5 Compatible: No negative lookbehind required
• ✅ Order-Dependent: Pattern 1 runs before Pattern 2 to prevent conflicts
• ✅ Boundary-Aware: Pattern 2 respects word boundaries to avoid matching substrings
• ✅ Path-Preserving: Both patterns preserve URL paths correctly

🔤 Unicode Normalization

• Fixed mojibake characters in comments and string literals throughout the codebase
• Corrected display of Czech characters (např., příspěvek)
• Fixed ellipsis character (…) representation in comments
• Normalized emoji rendering (𝕏, 🦋, 📝, 💬, 📤, 🖼️, 🔗)
• Improved character encoding for ampersand safe character (⅋)

These are cosmetic fixes that improve code readability without affecting functionality.

Migration from v3.1.3

This is a bug fix release. No configuration changes are required.

Recommended Actions:

1. Replace your current script with v3.1.4
2. If you use URL_DOMAIN_FIXES, test with content containing:
   • Domains with existing https:// prefixes
   • Bare domains without protocols
   • Multiple occurrences of the same domain
3. Verify that URLs are processed correctly

Breaking Changes:

• None. This release maintains full backward compatibility.

Testing

All existing tests pass with 100% success rate. Additional test coverage added for:

• URL_DOMAIN_FIXES with existing protocols
• Multiple occurrences of domains in content
• Mixed protocol scenarios

Installation

1. Copy the entire content of example-ifttt-filter-x-xcom-3_1_4.ts
2. Paste into your IFTTT applet filter code
3. Configure SETTINGS according to your needs
4. Test with beta bot before production deployment

File Information

• Filename: example-ifttt-filter-x-xcom-3_1_4.ts
• Size: ~64KB (within IFTTT limits)
• TypeScript Version: 2.9.2 (ES5 compatible)
• IFTTT Runtime: ES5

Known Issues

None at this time.

Next Steps

• Monitor production deployment for 24-48 hours
• Consider URL_DOMAIN_FIXES enhancements for future versions
• Continue optimization of regex patterns for performance

---

Related Files

• Main script: example-ifttt-filter-x-xcom-3_1_4.ts
• Documentation: See Unified_Filter_Guide__-_v3_1_0.md for complete usage guide
• Settings reference: See Settings_for_IFTTT_filter_script_-_v3_1_0.md

Credits

Developed and maintained by Daniel for Zprávobot.news

License

This script is part of the Zprávobot.news project.