
Use six invisible Unicode characters to check how your form, database and UI handle names that look empty. Includes regex fixes and byte counts.
A display-name field is marked required, your code trims whitespace, and the QA ticket says it works. Then someone pastes a single character that looks like nothing, passes validation, and your leaderboard now has a blank row. Invisible Text Generator produces exactly those characters so you can test your own forms and layouts before a user stumbles into the problem.
The clip below shows how the tool produces blank characters for bios and fields.
The tool offers six variants, each shown with its Unicode code point and a one-line note.
A slider sets how many to produce, from 1 to 20. Copy Invisible Text puts them on the clipboard, and a test box beneath lets you paste the result back and read the code points it contains, so you can prove what actually landed in a field.
Plain JavaScript trimming removes spaces, tabs, newlines and the non-breaking space. It does not remove any of the six characters above, and the whitespace class in a regular expression does not match them either. So a check such as "trimmed length is zero" lets all six through. To the user the field looks empty; to your code it is a valid three-byte string.
Byte counts matter for storage. Each of these characters takes three bytes in UTF-8 even though it shows as one character, so 20 of them occupy 60 bytes. A database column limited to a number of bytes can reject a name that looks short on screen, and a character limit in the interface may disagree with the one in the back end.
Run the following with the form you own or are authorised to test.
A blank name can leave a clickable link with nothing to click, a table row with zero visible content, or a mention that notifies someone but shows no text. Long runs of a width-bearing blank, such as Braille Blank or Hangul Filler, can push other elements sideways or add a horizontal scrollbar when the container is narrow. A zero-width space is also a permitted line-break point, so it can make text wrap in odd places, while the joiner characters do not allow a break at all.
Also check copy and search. Text containing a hidden character looks identical to the clean version but will not match a search, a duplicate detector or an exact comparison.
The usual remedy is to normalise input on the server and, ideally, in the browser too. Reject or strip a defined set of invisible code points, then require at least one visible letter or digit. A JavaScript pattern that matches a string made only of whitespace and these characters is:
^[\s-⠀ㅤ]*$
Paste it into Regex Tester, which runs the JavaScript engine of your browser, and try it against the test strings you generated. The pattern also includes the byte order mark U+FEFF, which the generator does not produce but which often arrives from copied files. To remove those characters from a longer string, use the same set without the anchors and the star, and replace matches with nothing. Unicode property escapes for the format category are broader still, but they need careful testing.
To see the raw bytes, encode a pasted blank in URL Encoder. Braille Blank becomes %E2%A0%80 and a zero-width space becomes %E2%80%8B, which makes an otherwise invisible character visible in a bug report.
The tool itself asks that blank text is not used to impersonate others, evade moderation or mislead people, and some platforms remove these characters anyway. Treat it as a test fixture, not a disguise. For more helpers of this kind, browse the text tools category.
Which field in your product worries you most, usernames, bios or comment bodies? Tell us on the contact page.