Does regular Unicode have a character that looks like a space to a human yet is not treated as a space by software please?
asmusf at ix.netcom.com
Sat Mar 29 06:01:43 CDT 2014
On managing some types of spacing between elements in running text:
On 3/27/2014 8:04 AM, Jukka K. Korpela wrote:
> 2014-03-27 15:10, Kalvesmaki, Joel wrote:
>> William, try the U+2000..U+200A glyphs under General Punctuation--I
>> that's what you're looking for to manage precise widths of blank space.
> That range contains some “fixed-width spaces”, yes. Being
> “fixed-width” is rather relative here, though, and many fonts do not
> contain these characters. Rendering software could of course display
> them by just leaving suitable spacing, but that’s not common.
> The “fixed-width spaces” are mostly just legacy characters, holdover
> from old typography. They may have their uses, though, in contexts
> where they work and other spacing methods don’t (for example, I
> recently noticed that they seem to be the only way to create a little
> spacing between an inline equation and normal character in MS Word).
They are useful when the object is to create fixed offsets between
elements in running text. Unless these elements have a special nature
that is widely recognized, there usually isn't any styling or markup
available to create the same effect.
As noted ..
> But for the purposes of indenting text lines, I don’t think they are
> useful. In almost all cases, there are better tools for indentation.
.. they are usually not needed for indentation and they are also not
normally used for justification -- it seems somewhat of an unsettled
question whether they do or do not partake in expansion / contraction
based on justification and similar adjustments to the width of the
It's the fact that indentation and justification do not need specific
width for spaces that lead to the (incorrect) statement, oft repeated,
that they are not needed in digital typography -- which is nonsense, of
course, but unfortunately, by now, well-entrenched nonsense.
>> And many (most?) software routines do not treat these as part of the
>> of spacing characters (\s in regular expressions).
> Well, most regexp implementations are very Ascii-oriented: notations
> like \s, \w, \d, etc. match Ascii characters only.
Which is an entirely different issue.
> Unicode mailing list
> Unicode at unicode.org
More information about the Unicode