Hash

Compute a cryptographic hash (digest) of text, for checksums, cache keys or comparing values. The text is hashed as UTF-8.

Base64

Encode text as Base64, or decode Base64 back to text. Used in data URLs, HTTP Basic authentication, JWTs and email attachments.

URL encode / decode

Percent-encode text so it can go safely in a URL query string, or decode an encoded string back to readable text.

Form encoding (application/x-www-form-urlencoded): spaces become +.

Hex

Convert text to its UTF-8 bytes in hexadecimal, or hex bytes back to text. Handy for inspecting encodings and binary data.

XML escape / unescape

Replace the characters that have a special meaning in XML and HTML with entities, or turn entities back into characters.

Escapes & < > " '. Unescape also handles numeric references like &#65; and &#x41;.

Case converter

Convert identifiers between naming conventions: camelCase, snake_case, kebab-case, CONSTANT_CASE and more.

Converts as you type. Each line is converted separately, so you can paste a list of names.

Lines → List

Turn a column of values (one per line) into a single delimited list, for example for an SQL IN clause or an array literal.

Split

Split text into parts on a delimiter or a regular expression and see every part with its index.

Substring

Extract part of a text by character position, counting from 0.

Negative indexes count from the end: start -3 gives the last 3 characters.

Transform

Apply a quick clean-up to text: sort or deduplicate lines, trim or remove whitespace, or reverse it.

Stats

Count characters, Unicode code points, UTF-8 bytes, words and lines. Useful to check length limits of fields and messages.

JSON validator

Check that text is valid JSON, see exactly where it breaks, and get it formatted or minified. Duplicate keys are reported as errors.

Regex tester

Test a Java regular expression against your text as you type: matches are highlighted, with their position and capture groups. Optionally try a replacement.

Java syntax (java.util.regex), as used by String.matches, Pattern and Spring. In the replacement, $1 or ${name} insert a group.


      

      
    


    
    

      
UUID generator

      
Generate universally unique identifiers (UUIDs), one or up to 100 at a time.

      

        
        
        
        
        
      

      
v7 UUIDs start with a timestamp, so they sort by creation time: a good choice for database keys.

    


    
    

      
Password generator

      
Generate a strong random password with a cryptographically secure generator. It is never cached.

      

        
        
        
        
        
        
        
      

      
The password contains at least one character of each selected type.

    


    
    

      
Random string

      
Generate a random string of the chosen length and characters, for tokens, test data or identifiers.

      

        
        
        
      

    


    
    

      
Java POJO from CSV

      
Paste a CSV header and get a Java field for every column, with a guessed type. Add a few data rows under the header and the types are taken from the values.

      
      

        
      

      

      
Values: true/false/yes/no → Boolean, whole numbers → BigInteger, decimals → BigDecimal, 2026-01-01 → LocalDate, 2026-01-01 09:30 → LocalDateTime, anything else → String. Wrapper types are used because an empty cell means null; choose primitive types when every row has a value.

    


    
    

      
Timestamp converter

      
Convert a Unix timestamp (epoch) to a readable date and back. Paste seconds, milliseconds or a date like 2026-01-01T09:00:00Z: the unit is recognised automatically.

      

        Now, Unix seconds
        
        
        Now, Unix milliseconds
        
        
      

      

        
        
        
      

      

      


      
How it works

      
What is a Unix timestamp? It counts the seconds since 1 January 1970 at 00:00 UTC (the "epoch"), the same number everywhere in the world. Time zones only matter when you turn it into a date.

      
Seconds or milliseconds? Linux (date +%s), PHP and most databases use seconds (10 digits today). Java (System.currentTimeMillis()) and JavaScript (Date.now()) use milliseconds (13 digits). Microseconds have 16 digits, nanoseconds 19.

      
Good to know: Unix time ignores leap seconds, and 32-bit systems that store it as a signed int overflow on 19 January 2038 at 03:14:07 UTC (the "year 2038 problem"). Use 64-bit values (long in Java).

    


    
    

      
Cron expression

      
Build a schedule and get the cron expression for Linux crontab, Spring (Java @Scheduled) and Quartz, with the next run times.

      

        
        
        
        
        
        
        
        
      

      
The end hour is not included: from 9 to 17 means the last run is before 17:00, so nothing runs in the evening or at night.

      

      

      

      
    
      
    Linux has 5 fields and no seconds. Spring puts seconds in front (6 fields). Quartz also starts with seconds, accepts an optional year and needs ? in day of month or day of week. Spring and Quartz days are written as names (MON-SUN) because the two number them differently (in Quartz 1 = Sunday).
    

      
    
        
    Check an expression
    
        
    
          
          
          
        
    
        
    Shows the next 5 runs, or what is wrong with the expression. Linux expressions follow cron's rule: when both day of month and day of week are set, either one matching is enough.
    
      
    
    
    

    
    
      
    
        Result
        
      
    
      
    
    
    

    
    
    How to use it
    Paste the text, enter the start position, counting from 0, and optionally the end. The end is exclusive: start 0 and end 5 on Hello, world give Hello. Leave the end empty to go to the end of the text.
    Frequently asked questions
    How do negative positions work?
    They count from the end of the text: start -5 on Hello, world gives world, and end -1 drops the last character. Handy when you know how much to keep from the end but not the length of the text.
    Is it the same as Java's substring?
    With positive positions, yes: it works like String.substring(start, end), with the same exclusive end. Java does not accept negative positions; here they work like slice in JavaScript or slicing in Python.
    Why does an emoji take two positions?
    Positions count UTF-16 units, like Java and JavaScript strings, and an emoji such as 😀 takes two. If a position falls in the middle of an emoji, the result shows a broken character.