[#charref-attributes]
= Predefined Attributes for Character Replacements

[#charref-attributes-table]
// tag::table[]
.Predefined attributes for character replacements ^[1][2][3]^
[width="75%", cols="^4m,^3l,^3"]
|===
|Attribute name |Replacement text |Appearance

|blank
e|nothing
|{empty}

|empty
e|nothing
|{empty}

|sp
e|single space
|{sp}

|nbsp
| 
|{nbsp}

|zwsp^[4]^
|​
|{zwsp}

|wj^[5]^
|⁠
|{wj}

|apos
|'
|{apos}

|quot
|"
|{quot}

|lsquo
|‘
|{lsquo}

|rsquo
|’
|{rsquo}

|ldquo
|“
|{ldquo}

|rdquo
|”
|{rdquo}

|deg
|°
|{deg}

|plus
|+
|{plus}

|brvbar
|¦
|¦

|vbar
|\|
|{vbar}

|amp
|&
|&

|lt
|<
|<

|gt
|>
|>

|startsb
|[
|[

|endsb
|]
|]

|caret
|^
|^

|asterisk
|*
|*

|tilde
|~
|~

|backslash
|\
|\

|backtick
|`
|`

|two-colons
|::
|::

|two-semicolons
|;;
|;;

|cpp
|C++
|C++
|===

^[1]^ Some replacements are Unicode characters, whereas others are numeric character references (e.g., \&#34;).
These character references are used whenever the use of the Unicode character could interfere with the AsciiDoc syntax or confuse the renderer (i.e., the browser).
It's up to the converter to transform the reference into something the renderer understands (something both the man page and PDF converter handle).

^[2]^ Asciidoctor does not prevent you from reassigning predefined attributes.
However, it's best to treat them as read-only unless the output format requires the use of a different encoding scheme.
These attributes are an effective tool for decoupling content and presentation.

^[3]^ Asciidoctor allows you to use any of the named character references (aka named entities) defined in HTML (e.g., \&euro; resolves to &euro;).
However, using named character references can cause problems when generating non-HTML output such as PDF because the lookup table needed to resolve these names may not be defined.
Our recommendation is avoid using named character references--with the exception of those defined in XML (i.e., lt, gt, amp, quot and apos).
Instead, use numeric character references (e.g., \&#8364;).

^[4]^ The Zero Width Space (ZWSP) is a code point in Unicode that shows where a long word can be split if necessary.

^[5]^ The word joiner (WJ) is a code point in Unicode that prevents a line break at its position.
// end::table[]