Improve the documentation on unquoted words in query expressions

This mentions that $ and @ are allowed in unquoted words (previously omitted). The documentation of the characters allowed in unquoted words is also changed to list punctuation characters in the same relative order and format as the list of punctuation characters allowed in target names in that section of the documentation.

It also lists the full set of characters permitted in target expressions which are not permitted in unquoted words (the documentation previously mentioned + and = only in examples and ~ not at all). It emphasizes the implication of this: When writing code that generates query expressions, target names should be quoted.

A slight tweak is made to the bit about alphanumeric characters, since A-Za-z is not "the set of alphabetic characters" and 0-9 is not "numerals" unless you specify "in ASCII".

RELNOTES: None.
PiperOrigin-RevId: 247435415

site/docs/query.html

@@ -77,17 +77,21 @@ <h2 id='tokens'>Tokens: The Lexical Syntax</h2>
     <li>
       <p>
         <b>Words</b>, such as <code>foo/...</code> or
-        <code>".*test rule"</code> or
-        <code>//bar/baz:all</code>.
-        If a character sequence is "quoted" (begins and ends with a
-        single-quote <code>'</code>, or begins and ends with a
-        double-quote <code>&quot;</code>), it is a word.
-        If a character sequence is not quoted, it may still be parsed as a word.
-        Unquoted words are sequences of characters drawn from
-        the set of alphabet characters, numerals, slash <code>/</code>,
-        hyphen <code>-</code>, underscore <code>_</code>, star <code>*</code>, and
-        period <code>.</code>. Unquoted words may not start with a
-        hyphen or period.
+        <code>".*test rule"</code> or <code>//bar/baz:all</code>. If a
+        character sequence is "quoted" (begins and ends with a single-quote
+        <code>'</code>, or begins and ends with a double-quote
+        <code>&quot;</code>), it is a word. If a character sequence is not
+        quoted, it may still be parsed as a word. Unquoted words are sequences
+        of characters drawn from the alphabet characters <code>A-Za-z</code>,
+        the numerals <code>0-9</code>, and the punctuation <code>_/.-@*$</code>
+        (underscore, slash, period, hyphen, at sign, star, dollar sign).
+        Unquoted words may not start with a hyphen or period, even though
+        relative <a href="build-ref.html#name">target names</a> may start with
+        those characters. Unquoted words also may not include the characters
+        plus sign <code>+</code>, equals sign <code>=</code>, or tilde
+        <code>~</code>, even though those characters are permitted in target
+        names. Thus, when writing code that generates query expressions, target
+        names should be quoted.
       </p>
 
     <p>We chose this syntax so that quote marks aren't needed in most cases.