Instantly extract every user-visible string in precise order
JSON, YAML, CSV, TOML, INI, ENV
Tested on Ubuntu, macOS, and Windows for maximum compatibility.
If String-LE saves you time, a quick rating helps other developers discover it:
⭐ VS Code Marketplace • Open VSX
Extract strings from any file format — JSON, CSV, ENV, YAML, TOML — in one click. No regex. No scripts. Just results.
String-LE intelligently identifies user-visible text while filtering out numbers, IDs, URLs, and technical noise. Get a clean, ordered set of strings ready for i18n, documentation, or validation.
-
i18n without the hassle
Instantly extract and de-duplicate locale strings into
en.json,fr.json, or any language pack. Keep translation files lean and in sync across your codebase. -
Clarity across configs & APIs
Surface every user-facing message. Validation errors, system prompts, and API responses are cataloged in one place. -
Confident edits in complex specs
Flatten nested strings into a simple list you can safely edit without breaking structure or formatting. -
Stream massive CSVs
Work with millions of rows without locking up VS Code. Select only the columns you need and stream results directly to the editor. -
Automatic cleanup built-in
- Sort for stable diffs and reviews
- Dedupe to eliminate noise
-
Fast at any scale
Benchmarked at millions of lines per second, String-LE keeps up with large datasets and enterprise monorepos without slowing you down.
- Numbers-LE - Extract and analyze numeric data with statistics • Open VSX
- EnvSync-LE - Keep .env files in sync with visual diffs • Open VSX
- Paths-LE - Extract file paths from imports and dependencies • Open VSX
- URLs-LE - Audit API endpoints and external resources • Open VSX
- Scrape-LE - Validate scraper targets before debugging • Open VSX
- Colors-LE - Extract and analyze colors from stylesheets • Open VSX
- Dates-LE - Extract temporal data from logs and APIs • Open VSX
- i18n & Localization - Extract user-visible strings for translation files and language packs
- Content Management - Pull titles, descriptions, and messages from CMS exports for auditing
- API Validation - Extract user-facing messages and errors from API responses for documentation
- Documentation Audits - Get all text content from docs for reviews and updates
- Install from VS Code Marketplace or Open VSX
- VS Code Marketplace: Install here
- Open VSX (Cursor, VSCodium): Install here
- Open any supported file type (
Cmd/Ctrl + P String-LE). - Run Quick Extract (
Cmd+Alt+E/Ctrl+Alt+E/ Status Bar).
string-le.openResultsSideBySide– Open to the sidestring-le.csv.streamingEnabled– Toggle CSV streamingstring-le.dedupeEnabled– Auto-dedupe stringsstring-le.sortEnabled– Auto-sort output
- Safety Guards – File size warnings & thresholds
- Notification Levels – Control verbosity and alerts
i18n Workflow
{
"string-le.openResultsSideBySide": true,
"string-le.dedupeEnabled": true,
"string-le.sortEnabled": true,
"string-le.csv.streamingEnabled": false
}Large Dataset Processing
{
"string-le.csv.streamingEnabled": true,
"string-le.dedupeEnabled": false,
"string-le.sortEnabled": false,
"string-le.openResultsSideBySide": false
}Content Audit
{
"string-le.dedupeEnabled": true,
"string-le.sortEnabled": true,
"string-le.openResultsSideBySide": true,
"string-le.csv.streamingEnabled": false
}- CSV support assumes standard delimiter/quoting; unusual dialects not supported
- Large outputs can be slow to open — use Copy when prompted
- Streaming applies only to CSV; other formats load in memory
- Multi-line strings (e.g., YAML block scalars) are only partially supported
- CSV multi-line and all-column extracts stream to the editor only (no auto-copy)
- Sorting and deduplication apply to final strings, not their original positions
- Fallback mode uses quoted-string heuristics and may include false positives
See CONFIGURATION.md.
String-LE is built for speed and handles files from 100KB to 30MB+. See detailed benchmarks.
| Format | File Size | Throughput | Duration | Memory | Tested On |
|---|---|---|---|---|---|
| JSON | 0.13MB | 4703509 | ~1.14 | < 1MB | Apple Silicon |
| JSON | 1.31MB | 6337898 | ~8.47 | < 1MB | Apple Silicon |
| JSON | 6.55MB | 6181529 | ~43.42 | < 1MB | Apple Silicon |
| ENV | 5K lines | 1,144,828 | ~0.29 | < 1MB | Apple Silicon |
Note: CSV format tests encountered parsing errors with generated test data. Real-world CSV performance may vary.
Real-World Performance: Tested with actual data up to 20MB (practical limit: 1MB warning, 10MB error threshold)
Performance Monitoring: Built-in real-time tracking with configurable thresholds
Full Metrics: docs/PERFORMANCE.md • Test Environment: macOS, Bun 1.2.22, Node 22.x
- Sorting: Adds 10-20% processing time when enabled
- Hardware Requirements: Minimum 4GB RAM, recommended 8GB+ for large datasets
See PERFORMANCE.md.
13 languages: English, German, Spanish, French, Indonesian, Italian, Japanese, Korean, Portuguese (Brazil), Russian, Ukrainian, Vietnamese, Chinese (Simplified)
VS Code 1.70.0+ • Platform Windows, macOS, Linux
Memory 500MB recommended for large files
100% local processing. No data leaves your machine. Optional logging: string-le.telemetryEnabled
Not detecting strings?
Ensure file is saved with supported extension (.json, .yaml, .csv, .toml, .ini, .env)
Large files slow?
Enable CSV streaming: string-le.csv.streamingEnabled: true
Need help?
Check Issues or enable logging: string-le.telemetryEnabled: true
What strings are extracted?
User-visible text only (excludes IDs, numbers, technical values, variable names)
Can I get deduplication?
Yes, enable string-le.dedupeEnabled: true to remove duplicates automatically
Max file size?
Up to 500MB with CSV streaming. Practical limit: 10MB for other formats
Perfect for i18n?
Absolutely! Extract strings for translation files, locale management, and content audits
92 unit tests • 44.11% function coverage, 9.05% line coverage
Powered by Vitest • Run with bun test --coverage
See TESTING.md.
Copyright © 2025
@nolindnaidoo. All rights reserved.