I was doing a bunch of search-replace operations in notepad++ to effectively minify my css - mostly removing whitespace/tabs etc...) This ended up breaking much of my css.
Apparently a strange character (​) was inserted all over the place) Using notepad++ in UTF-8 without BOM, I cannot see these, but they appeared in a view-source.
I was able to remove these by doing a search replace in ANSI encoding, but my question is, what is this character, and why might it have appeared?
The string “​” is the UTF-8 encoded form of ZWSP when misinterpreted as windows-1252 encoded data. (Checked this using a nice UTF-8 decoder.) This explains why you don’t see it in Notepad++ in UTF-8 mode; ZWSP (zero-width space) is an invisible character with no width.
Apparently browsers are interpreting the style sheet as windows-1252 encoded. Saving the file with BOM might help, since then browsers would probably guess the encoding better. The real fix is to make sure (in a server-dependent manner) that the server sends appropriate Content-Type header for the CSS file.
But if this is the only non-Ascii character in your CSS file, it does not matter in practice, after you have removed the offending data.
I don’t know of any simple way to make Notepad++ insert ZWSP (you could of course use general character insertion utilities in the system), so it’s a bit of mystery where it came from. Perhaps via copy and paste from somewhere.
Using the web developer plug in or ext in Firefox you can see the problem character in the css document.
In Visual Studio all I could see was:
Web developer showed an unwanted hidden character, an "a" with a caret on top:
The utf encoder link above revealed this
and this
but simply fix the problem by deleting and retyping.