Invisible red dot UNICODE character

[Skysnake 95]

'F‌inal' ; <========= problem between F and i 

'F‌inal' ; <========= problem between F and i 

'Final' ; <========= Notepad ANSI, no problem

F‌inal

PostgreSQL complains about that red dot.  Notepad++ marks the first word.

Notepad said it contains Unicode.  Saved as ANSI and put back. Red dot is gone.

 

the Text is inserted manually into an AutoIt Input box.  Things get done to it, and it ends up in a SQL database.  The data seems fine, but when I start generating reports, all kinds of funny problems show up.

 

If I can ID that character, I can remove it.

Any ideas?

Edited March 5, 2019 by Skysnake
Typo

[user4157124 23]

Possibly:

StringRegExpReplace('F‌inal', '[\x{200C}\x{200B}]', '')

Edited March 15, 2019 by user4157124

[Subz 826]

You could also replace non printable characters:

StringRegExpReplace(GuiCtrlRead($sInput), "[^[:print:]]", "")

 

[jchd 1,824]

With a bit broader range of removal:

StringRegExpReplace('F‌inal', '[\x{200B}-\x{200D}]', '')

Yet removing ZWJ (U+0200D) may change semantic of Unicode flux in some languages.

You may also discover other such special codepoints, depending on the nature of the input text.

[TheDcoder 1,842]

4 hours ago, Skysnake said:

'F‌inal' ; <========= problem between F and i 

'F‌inal' ; <========= problem between F and i 

'Final' ; <========= Notepad ANSI, no problem

Going by the character which my browser claims to be in between F and i, it is a Zero-width non-joiner character in unicode (code point 8204), it is used in some languages to represent more complex writing systems