of an error message on a given line to deduce the information it
needs. The program is syntactically valid regardless of whether each
name is a type or an ordinary identifier, so there will be no syntax
errors that might stop parsing early.

An error on not-declared:1 indicates that foo is undeclared.
An error on not-type:1 indicates that foo is not a type (if declared at all, it is an identifier).

// returned by a Split function to indicate that the scanning should stop
// with no error. If the token being delivered with this error is not nil,
// the token is the last token.
//
// The value is useful to stop processing early or when it is necessary to
// deliver a final empty token (which is different from a nil token).
// One could achieve the same behavior with a custom error value but
// providing one here is tidier.

Also, constants are always evaluated as 64-bit unsigned integers.
Thus <code>-2</code> is not the integer value minus two,
but the unsigned 64-bit integer with the same bit pattern.
The distinction rarely matters but
to avoid ambiguity, division or right shift where the right operand's
high bit is set is rejected.
</p>

<h3 id="symbols">Symbols</h3>

<p>

pkg syscall (linux-386), const PR_MCE_KILL = 33
pkg syscall (linux-386), const PR_MCE_KILL_CLEAR = 0
pkg syscall (linux-386), const PR_MCE_KILL_DEFAULT = 2
pkg syscall (linux-386), const PR_MCE_KILL_EARLY = 1
pkg syscall (linux-386), const PR_MCE_KILL_GET = 34
pkg syscall (linux-386), const PR_MCE_KILL_LATE = 0
pkg syscall (linux-386), const PR_MCE_KILL_SET = 1
pkg syscall (linux-386), const PR_SET_DUMPABLE = 4

func genLine(buf *bytes.Buffer, lineNum, n int, addNewline bool) {
	buf.Reset()
	doCR := lineNum%5 == 0
	if doCR {
		n--
	}
	for i := 0; i < n-1; i++ { // Stop early for \n.
		c := 'a' + byte(lineNum+i)
		if c == '\n' || c == '\r' { // Don't confuse us.
			c = 'N'
		}
		buf.WriteByte(c)
	}
	if addNewline {
		if doCR {
			buf.WriteByte('\r')
		}