Visual text pipeline

Your input is processed only in this browser. See the privacy policy.

How to use it

  1. Paste a line-based list or delimited text into the input.
  2. Choose a step type, add it, and edit its options.
  3. Check the line count after every stage and copy the final result.
  4. Click a recipe above to load a whole pipeline at once. If the input is empty, sample text is filled in too.
  5. Click a step name to see the output right after that step and how the line count changed. Untick a step to skip it temporarily.
  6. Copy recipe link stores only the steps and options in the URL fragment. Your input text is never included.

Examples

Build a clean unique list

Stack trim → remove blank lines → sort → unique to produce a readable deduplicated list.

Input
 pear \nApple\n\napple\npear
Result
apple\nApple\npear

Cut and filter a delimited list

Select the second comma-delimited field and add an include filter to visualize a flow similar to cut followed by grep.

Pull out email addresses

The Extract emails recipe runs extract (email) → lowercase → unique, so the same address written in different cases collapses into one line.

Input
Kim: kim@example.com, lee@test.co.kr\nCC: KIM@example.com
Result
kim@example.com\nlee@test.co.kr

Build an SQL IN list

Remove blanks and duplicates, wrap each line in quotes with the regex replace ^(.*)$ → '$1', then join with ", ".

Input
1001\n1002\n1001
Result
'1001', '1002'

Details

Visual Text Pipeline replaces one opaque regular expression with small, explainable steps that run from top to bottom. Combine sorting, deduplication, blank-line removal, trimming, case conversion, literal or regular-expression replacement, delimited field selection, include or exclude filters, numbering, joining and splitting.

Common jobs load as one-click recipes: dedupe and sort, extract emails, URLs or numbers, clean whitespace, unwrap text copied from a PDF, pick a CSV column, count word frequency, make slugs and build an SQL IN list. After loading you can still edit, disable or reorder any step, and Undo (Ctrl+Z) restores the previous pipeline.

Every card reports the number of lines after that stage. Order changes meaning: converting case before deduplication can merge Apple and apple, while deduplicating first may preserve both. Intermediate counts make that difference visible before you copy the final result.

The generated shell pipeline is educational and useful as a starting point. Browser and operating-system versions of sort, sed, awk, grep, cut, tr and paste can differ in locale sorting, Unicode case handling and regular-expression syntax. Test important automation with representative data.

Frequently asked questions

Does step order matter?

Yes. Each step consumes the previous step's output.

What happens when a regular expression is invalid?

The pipeline stops at that step, shows an error, and preserves the result from the preceding stage.

Will the shell command always produce identical output?

Not always. Locale sorting, Unicode case conversion and sed implementations vary, so treat it as an educational starting point.

Is my input stored?

No. It is processed in memory in this tab and is not put into URLs, browser storage or analytics events.

Does the recipe link contain my text?

No. Only step types and options go after the # in the link, and the fragment is never sent to a server. The recipient pastes their own text to run the same steps.

How do I find the step that broke my data?

Click the step names one by one. The output switches to the result right after that step, and every step shows its line count and how many lines it added or removed. A failing step turns red and later steps do not run.