Question

I know that there are many types of space (em space, en space, thin space, non-breaking space, etc), but, all these, that I refered, have HTML entities (at least, PHP’s htmlentities() return something like  .

But, what about those spaces that have no HTML entities?
Example: [example URL not valid anymore]
Look at the nickname of this account. It has many ” ” (spaces) at the front, which are visible for us (this doesn’t happen with the  ).

I tried already filter with regular expressions, using \x escape, filter with str_replace(), with the space as the argument, and no luck at all!

Do you have any suggestion on how to filter ALL types of whitespace?

Answer 1

\s by default, will not match whitespace characters with values greater than 128. To get at those, you can instead make good use of other UTF-8-aware sequences.

---

^{(Standard disclaimer: I’m skimming the PCRE source code to compile the lists below, I may miss a character or type something incorrectly. Please forgive me.)}

\p{Zs} matches:

• U+0020 Space
• U+00A0 No-break space
• U+1680 Ogham space mark
• U+180E Mongolian vowel separator
• U+2000 En quad
• U+2001 Em quad
• U+2002 En space
• U+2003 Em space
• U+2004 Three-per-em space
• U+2005 Four-per-em space
• U+2006 Six-per-em space
• U+2007 Figure space
• U+2008 Punctuation space
• U+2009 Thin space
• U+200A Hair space
• U+202F Narrow no-break space
• U+205F Medium mathematical space
• U+3000 Ideographic space

\h (Horizontal whitespace) matches the same as \p{Zs} above, plus

• U+0009 Horizontal tab.

Similarly for matching vertical whitespace there are a few options.

\p{Zl} matches U+2028 Line separator.

\p{Zp} matches U+2029 Paragraph separator.

\v (Vertical whitespace) matches \p{Zl}, \p{Zp} and the following

• U+000A Linefeed
• U+000B Vertical tab
• U+000C Formfeed
• U+000D Carriage return
• U+0085 Next line

Going back to the beginning, in UTF-8 mode (i.e. using the u pattern modifier) \s will match any character that \p{Z} matches (which is anything that \p{Zs}, \p{Zl} and \p{Zp} will match), plus

• U+0009 Horizontal tab
• U+000A Linefeed
• U+000C Formfeed
• U+000D Carriage return

---

To cut a long story short (I bet you read all of the above, didn’t you?) you might want to use \s but make sure to be in UTF-8 mode like /\s/u. Putting that to some practical use, to filter out those matching whitespace characters from a string you would do something like

$new_string = preg_replace('/\s/u', '', $old_string);

Finally, if you really, really care about the vertical whitespaces which aren’t included in \s (LF and NEL) then you can use the character class [\s\v] to match all 26 of the whitespace characters listed above.