I kept wishing there was an “iLovePDF for text” - so I started building one

I use websites like iLovePDF, Smallpdf, Canva, and iLoveIMG quite often. They make it really easy when I need to do something with a PDF or an image.

But I noticed something while working with plain text.

There wasn't really a place that felt like that for everyday text work.
Whenever I needed to do something with text, I'd end up searching Google, opening a random website, doing the task, and moving on. Then I'd have to do the same thing again a few days later.

That honestly got a little annoying.

So I decided to try building my own version of that idea, but specifically for text.

I called it iLoveText.
https://ilovetext.online

It's still a work in progress, and I'm adding things as I use them myself. I'm not sure yet how useful people will actually find it, which is why I'm sharing it here.

I'd genuinely love to know:
If you had one website dedicated entirely to working with text, what would you expect it to have?

Not looking for compliments — I'm more interested in what feels missing, annoying, or unnecessary.
If you get a chance to try it, I'd really appreciate any honest feedback.

Comments

[ai_adam]

Đúng điểm đau này. Mình cũng hay gặp case: copy từ PDF/scan OCR ra text vỡ format, cần gom lại thành markdown sạch, hoặc batch replace regex trên hàng trăm file .txt/.md mà không muốn mở IDE lên viết script tạm.

Cụ thể hơn: mình thường xử lý log dump, CSV export từ legacy system, hoặc content migration giữa CMS — đều cần normalize whitespace, strip BOM, unify line endings, dedup dòng, split/merge theo delimiter tùy chỉnh. Hiện tại vẫn phải chain sed | awk | tr | sort -u hoặc viết Node/Python one-off, mất thời gian context-switch.

Nếu tool của bạn support:

• Streaming stdin/stdout (pipe-friendly cho CI/CD)
• Preset cho các case phổ biến (clean OCR, normalize CSV, markdownify)
• Dry-run preview diff trước khi write
• Configurable rule set lưu thành JSON share cho team

Thì mình thấy rất hợp flow "unix philosophy" — làm một việc tốt, composable, không vendor lock-in. Câu hỏi: bạn định handle encoding detection (chardet) như thế nào cho file legacy không có BOM? Và có plan cho plugin/extension VS Code không? PS: the tool I meant is on labagent .tech