[BACK]Return to shtokenizer_test.go CVS log [TXT][DIR] Up to [cvs.NetBSD.org] / pkgsrc / pkgtools / pkglint / files

File: [cvs.NetBSD.org] / pkgsrc / pkgtools / pkglint / files / shtokenizer_test.go (download)

Revision 1.23, Sun Dec 8 22:03:38 2019 UTC (2 months, 2 weeks ago) by rillig
Branch: MAIN
Changes since 1.22: +2 -2 lines

pkgtools/pkglint: update pkglint to 19.3.15

Changes since 19.3.14:

Invalid lines in PLIST files are now reported as errors instead of
warnings. If pkglint doesn't know about it, it must be an error.

In PLIST files, all paths are validated to be canonical. That is, no
dotdot components, no absolute paths, no extra slashes, no intermediate
dot components.

Fewer notes for unexpanded variable expressions in DESCR files. Before,
the text $@ was reported as possible Makefile variable even though it
was just a Perl expression.

README files are allowed again in pkgsrc package directories. There was
no convincing argument why these should be forbidden.

A few diagnostics have been changed from NOTE to WARNING or from WARNING
to ERROR, to match their wording.

When pkglint suggests to replace :M with ==, the wording is now "can be
made" instead of "should".

package pkglint

import (
	"gopkg.in/check.v1"
	"strings"
)

// This test demonstrates that the shell tokenizer is not perfect yet.
// There are still some corner cases that trigger a parse error.
// To get 100% code coverage, they have been found using the fuzzer
// and trimmed down to minimal examples.
func (s *Suite) Test_ShTokenizer__examples_from_fuzzing(c *check.C) {
	t := s.Init(c)

	test := func(input string, diagnostics ...string) {
		mklines := t.NewMkLines("filename.mk",
			MkCvsID,
			"\t"+input)
		mklines.Check()
		t.CheckOutput(diagnostics)
	}

	// Covers shAtomBacktDquot: return nil.
	// These are nested backticks with double quotes,
	// which should be avoided since POSIX marks them as unspecified.
	test(
		"`\"`",
		"WARN: filename.mk:2: Internal pkglint error in ShTokenizer.ShAtom at \"`\" (quoting=bd).",
		"WARN: filename.mk:2: Pkglint ShellLine.CheckShellCommand: splitIntoShellTokens couldn't parse \"`\\\"`\"")

	// Covers shAtomBacktSquot: return nil
	test(
		"`'$`",
		"WARN: filename.mk:2: Internal pkglint error in ShTokenizer.ShAtom at \"$`\" (quoting=bs).",
		"WARN: filename.mk:2: Pkglint ShellLine.CheckShellCommand: splitIntoShellTokens couldn't parse \"`'$`\"",
		"WARN: filename.mk:2: Internal pkglint error in MkLine.Tokenize at \"$`\".")

	// Covers shAtomDquotBacktSquot: return nil
	test(
		"\"`'`y",
		"WARN: filename.mk:2: Pkglint ShellLine.CheckShellCommand: splitIntoShellTokens couldn't parse \"\\\"`'`y\"")

	// Covers shAtomDquotBackt: return nil
	// XXX: Pkglint should parse unescaped dollar in the same way, everywhere.
	test(
		"\"`$|",
		"WARN: filename.mk:2: Internal pkglint error in ShTokenizer.ShAtom at \"$|\" (quoting=db).",
		"WARN: filename.mk:2: Pkglint ShellLine.CheckShellCommand: splitIntoShellTokens couldn't parse \"\\\"`$|\"",
		"WARN: filename.mk:2: Internal pkglint error in MkLine.Tokenize at \"$|\".")

	// Covers shAtomDquotBacktDquot: return nil
	// XXX: Pkglint should support unlimited nesting.
	test(
		"\"`\"`",
		"WARN: filename.mk:2: Internal pkglint error in ShTokenizer.ShAtom at \"`\" (quoting=dbd).",
		"WARN: filename.mk:2: Pkglint ShellLine.CheckShellCommand: splitIntoShellTokens couldn't parse \"\\\"`\\\"`\"")

	// Covers shAtomSubshDquot: return nil
	test(
		"$$(\"'",
		"WARN: filename.mk:2: Invoking subshells via $(...) is not portable enough.")

	// Covers shAtomSubsh: case lexer.AdvanceStr("`")
	test(
		"$$(`",
		"WARN: filename.mk:2: Invoking subshells via $(...) is not portable enough.")

	// Covers shAtomSubshSquot: return nil
	test(
		"$$('$)",
		"WARN: filename.mk:2: Internal pkglint error in ShTokenizer.ShAtom at \"$)\" (quoting=Ss).",
		"WARN: filename.mk:2: Invoking subshells via $(...) is not portable enough.",
		"WARN: filename.mk:2: Internal pkglint error in MkLine.Tokenize at \"$)\".")

	// Covers shAtomDquotBackt: case lexer.AdvanceRegexp("^#[^`]*")
	test(
		"\"`# comment",
		"WARN: filename.mk:2: Pkglint ShellLine.CheckShellCommand: splitIntoShellTokens couldn't parse \"\\\"`# comment\"")
}

// In order to get 100% code coverage for the shell tokenizer, a panic() statement has been
// added to each uncovered basic block. After that, this fuzzer quickly found relatively
// small example programs that led to the uncovered code.
//
// This test is not useful as-is.
func (s *Suite) Test_ShTokenizer__fuzzing(c *check.C) {
	t := s.Init(c)

	fuzzer := NewFuzzer()
	fuzzer.Char("\"'`$();|_#", 10)
	fuzzer.Range('a', 'z', 5)

	// This "real" line is necessary because the autofix
	// in MkParser.varUseBrace checks this.
	line := t.NewLine("Makefile", 1, "\t:")

	defer fuzzer.CheckOk()
	for i := 0; i < 1000; i++ {
		tokenizer := NewShTokenizer(line, fuzzer.Generate(50), false)
		tokenizer.ShAtoms()
		t.Output() // Discard the output, only react on panics.
	}
	fuzzer.Ok()
}

func (s *Suite) Test_ShTokenizer_ShAtom(c *check.C) {
	t := s.Init(c)

	// testRest ensures that the given string is parsed to the expected
	// atoms, and returns the remaining text.
	testRest := func(s string, expectedAtoms []*ShAtom, expectedRest string) {
		line := t.NewLine("filename.mk", 1, "")
		p := NewShTokenizer(line, s, false)

		actualAtoms := p.ShAtoms()

		t.CheckEquals(p.Rest(), expectedRest)
		t.CheckEquals(len(actualAtoms), len(expectedAtoms))

		for i, actualAtom := range actualAtoms {
			if i < len(expectedAtoms) {
				t.CheckDeepEquals(actualAtom, expectedAtoms[i])
			} else {
				t.CheckDeepEquals(actualAtom, nil)
			}
		}
	}
	atoms := func(atoms ...*ShAtom) []*ShAtom { return atoms }

	// test ensures that the given string is parsed to the expected
	// atoms, and that the text is completely consumed by the parser.
	test := func(str string, expected ...*ShAtom) {
		testRest(str, expected, "")
		t.CheckOutputEmpty()
	}

	atom := func(typ ShAtomType, text string) *ShAtom {
		return &ShAtom{typ, text, shqPlain, nil}
	}

	operator := func(s string) *ShAtom { return atom(shtOperator, s) }
	comment := func(s string) *ShAtom { return atom(shtComment, s) }
	mkvar := func(varname string, modifiers ...string) *ShAtom {
		text := "${" + varname
		for _, modifier := range modifiers {
			text += ":" + strings.Replace(strings.Replace(modifier, "\\", "\\\\", -1), ":", "\\:", -1)
		}
		text += "}"
		varuse := NewMkTokenBuilder().VarUse(varname, modifiers...)
		return &ShAtom{shtVaruse, text, shqPlain, varuse}
	}
	shvar := func(text, varname string) *ShAtom { return &ShAtom{shtShVarUse, text, shqPlain, varname} }
	text := func(s string) *ShAtom { return atom(shtText, s) }
	whitespace := func(s string) *ShAtom { return atom(shtSpace, s) }

	space := whitespace(" ")
	semicolon := operator(";")
	pipe := operator("|")
	subshell := atom(shtSubshell, "$$(")

	q := func(q ShQuoting) func(atom *ShAtom) *ShAtom {
		return func(atom *ShAtom) *ShAtom {
			return &ShAtom{atom.Type, atom.MkText, q, atom.data}
		}
	}
	backt := q(shqBackt)
	dquot := q(shqDquot)
	squot := q(shqSquot)
	subsh := q(shqSubsh)
	backtDquot := q(shqBacktDquot)
	backtSquot := q(shqBacktSquot)
	dquotBackt := q(shqDquotBackt)
	subshDquot := q(shqSubshDquot)
	subshSquot := q(shqSubshSquot)
	subshBackt := q(shqSubshBackt)
	dquotBacktDquot := q(shqDquotBacktDquot)
	dquotBacktSquot := q(shqDquotBacktSquot)

	// Ignore unused functions; useful for deleting some of the tests during debugging.
	use := func(args ...interface{}) {}
	use(testRest, test, atoms)
	use(operator, comment, mkvar, text, whitespace)
	use(space, semicolon, pipe, subshell, shvar)
	use(backt, dquot, squot, subsh)
	use(backtDquot, backtSquot, dquotBackt, subshDquot, subshSquot)
	use(dquotBacktDquot, dquotBacktSquot)

	test("",
		nil...)

	test("$$var",
		shvar("$$var", "var"))

	test("$$var$$var",
		shvar("$$var", "var"),
		shvar("$$var", "var"))

	test("$$var;;",
		shvar("$$var", "var"),
		operator(";;"))

	test("'single-quoted'",
		squot(text("'")),
		squot(text("single-quoted")),
		text("'"))

	test("\"",
		dquot(text("\"")))

	test("$${file%.c}.o",
		shvar("$${file%.c}", "file"),
		text(".o"))

	test("hello",
		text("hello"))

	test("hello, world",
		text("hello,"),
		space,
		text("world"))

	test("\"",
		dquot(text("\"")))

	test("`",
		backt(text("`")))

	test("`cat filename`",
		backt(text("`")),
		backt(text("cat")),
		backt(space),
		backt(text("filename")),
		text("`"))

	test("hello, \"world\"",
		text("hello,"),
		space,
		dquot(text("\"")),
		dquot(text("world")),
		text("\""))

	test("set -e;",
		text("set"),
		space,
		text("-e"),
		semicolon)

	test("cd ${WRKSRC}/doc/man/man3; PAGES=\"`ls -1 | ${SED} -e 's,3qt$$,3,'`\";",
		text("cd"),
		space,
		mkvar("WRKSRC"),
		text("/doc/man/man3"),
		semicolon,
		space,
		text("PAGES="),
		dquot(text("\"")),
		dquotBackt(text("`")),
		dquotBackt(text("ls")),
		dquotBackt(space),
		dquotBackt(text("-1")),
		dquotBackt(space),
		dquotBackt(operator("|")),
		dquotBackt(space),
		dquotBackt(mkvar("SED")),
		dquotBackt(space),
		dquotBackt(text("-e")),
		dquotBackt(space),
		dquotBacktSquot(text("'")),
		dquotBacktSquot(text("s,3qt$$,3,")),
		dquotBackt(text("'")),
		dquot(text("`")),
		text("\""),
		semicolon)

	test("ls -1 | ${SED} -e 's,3qt$$,3,'",
		text("ls"), space, text("-1"), space,
		pipe, space,
		mkvar("SED"), space, text("-e"), space,
		squot(text("'")), squot(text("s,3qt$$,3,")), text("'"))

	test("(for PAGE in $$PAGES; do ",
		operator("("),
		text("for"),
		space,
		text("PAGE"),
		space,
		text("in"),
		space,
		shvar("$$PAGES", "PAGES"),
		semicolon,
		space,
		text("do"),
		space)

	test("    ${ECHO} installing ${DESTDIR}${QTPREFIX}/man/man3/$${PAGE}; ",
		whitespace("    "),
		mkvar("ECHO"),
		space,
		text("installing"),
		space,
		mkvar("DESTDIR"),
		mkvar("QTPREFIX"),
		text("/man/man3/"),
		shvar("$${PAGE}", "PAGE"),
		semicolon,
		space)

	test("    set - X `head -1 $${PAGE}qt`; ",
		whitespace("    "),
		text("set"),
		space,
		text("-"),
		space,
		text("X"),
		space,
		backt(text("`")),
		backt(text("head")),
		backt(space),
		backt(text("-1")),
		backt(space),
		backt(shvar("$${PAGE}", "PAGE")),
		backt(text("qt")),
		text("`"),
		semicolon,
		space)

	test("`\"one word\"`",
		backt(text("`")),
		backtDquot(text("\"")),
		backtDquot(text("one word")),
		backt(text("\"")),
		text("`"))

	test("$$var \"$$var\" '$$var' `$$var`",
		shvar("$$var", "var"),
		space,
		dquot(text("\"")),
		dquot(shvar("$$var", "var")),
		text("\""),
		space,
		squot(text("'")),
		squot(shvar("$$var", "var")),
		text("'"),
		space,
		backt(text("`")),
		backt(shvar("$$var", "var")),
		text("`"))

	test("\"`'echo;echo'`\"",
		dquot(text("\"")),
		dquotBackt(text("`")),
		dquotBacktSquot(text("'")),
		dquotBacktSquot(text("echo;echo")),
		dquotBackt(text("'")),
		dquot(text("`")),
		text("\""))

	test("cat<file",
		text("cat"),
		operator("<"),
		text("file"))

	test("\\$$escaped",
		text("\\$$escaped"))

	test("-e \"s,\\$$sysconfdir/jabberd,\\$$sysconfdir,g\"",
		text("-e"),
		space,
		dquot(text("\"")),
		dquot(text("s,\\$$sysconfdir/jabberd,\\$$sysconfdir,g")),
		text("\""))

	test("echo $$, $$- $$/ $$; $$| $$,$$/$$;$$-",
		text("echo"),
		space,
		text("$$,"),
		space,
		shvar("$$-", "-"),
		space,
		text("$$/"),
		space,
		text("$$"),
		semicolon,
		space,
		text("$$"),
		pipe,
		space,
		text("$$,$$/$$"),
		semicolon,
		shvar("$$-", "-"))

	test("COMMENT=\t\\Make $$$$ fast\"",

		text("COMMENT="),
		whitespace("\t"),
		text("\\Make"),
		space,
		shvar("$$$$", "$"),
		space,
		text("fast"),
		dquot(text("\"")))

	test("var=`echo;echo|echo&echo||echo&&echo>echo`",
		text("var="),
		backt(text("`")),
		backt(text("echo")),
		backt(semicolon),
		backt(text("echo")),
		backt(operator("|")),
		backt(text("echo")),
		backt(operator("&")),
		backt(text("echo")),
		backt(operator("||")),
		backt(text("echo")),
		backt(operator("&&")),
		backt(text("echo")),
		backt(operator(">")),
		backt(text("echo")),
		text("`"))

	test("# comment",
		comment("# comment"))
	test("no#comment",
		text("no#comment"))
	test("`# comment`continue",
		backt(text("`")),
		backt(comment("# comment")),
		text("`"),
		text("continue"))
	test("`no#comment`continue",
		backt(text("`")),
		backt(text("no#comment")),
		text("`"),
		text("continue"))

	test("var=`tr 'A-Z' 'a-z'`",
		text("var="),
		backt(text("`")),
		backt(text("tr")),
		backt(space),
		backtSquot(text("'")),
		backtSquot(text("A-Z")),
		backt(text("'")),
		backt(space),
		backtSquot(text("'")),
		backtSquot(text("a-z")),
		backt(text("'")),
		text("`"))

	test("var=\"`echo \"\\`echo foo\\`\"`\"",
		text("var="),
		dquot(text("\"")),
		dquotBackt(text("`")),
		dquotBackt(text("echo")),
		dquotBackt(space),
		dquotBacktDquot(text("\"")),
		dquotBacktDquot(text("\\`echo foo\\`")), // One atom, since it doesn't influence parsing.
		dquotBackt(text("\"")),
		dquot(text("`")),
		text("\""))

	test("if cond1; then action1; elif cond2; then action2; else action3; fi",
		text("if"), space, text("cond1"), semicolon, space,
		text("then"), space, text("action1"), semicolon, space,
		text("elif"), space, text("cond2"), semicolon, space,
		text("then"), space, text("action2"), semicolon, space,
		text("else"), space, text("action3"), semicolon, space,
		text("fi"))

	test("$$(cat)",
		subsh(subshell),
		subsh(text("cat")),
		operator(")"))

	test("$$(cat 'file')",
		subsh(subshell),
		subsh(text("cat")),
		subsh(space),
		subshSquot(text("'")),
		subshSquot(text("file")),
		subsh(text("'")),
		operator(")"))

	test("$$(# comment) arg",
		subsh(subshell),
		subsh(comment("# comment")),
		operator(")"),
		space,
		text("arg"))

	test("$$(echo \"first\" 'second')",
		subsh(subshell),
		subsh(text("echo")),
		subsh(space),
		subshDquot(text("\"")),
		subshDquot(text("first")),
		subsh(text("\"")),
		subsh(space),
		subshSquot(text("'")),
		subshSquot(text("second")),
		subsh(text("'")),
		operator(")"))

	test("$$(echo `echo nested-subshell`)",
		subsh(subshell),
		subsh(text("echo")),
		subsh(space),
		subshBackt(text("`")),
		subshBackt(text("echo")),
		subshBackt(space),
		subshBackt(text("nested-subshell")),
		subsh(operator("`")),
		operator(")"))

	// Subshell with unbalanced parentheses, taken from src/build.sh,
	// around line 160. Many shells (and pkglint) fail this test,
	// therefore just don't write code like this.
	test("var=$$(case x in x) still-subshell;; esac);",
		text("var="), subsh(subshell),
		subsh(text("case")), subsh(space), subsh(text("x")), subsh(space),
		subsh(text("in")), subsh(space), subsh(text("x")),
		// XXX: The parenthesis is for the case pattern, not the end of the subshell.
		operator(")"), space,
		text("still-subshell"), operator(";;"), space,
		text("esac"), operator(")"), operator(";"))

	testRest("`echo \\${VAR}`",
		atoms(
			backt(text("`")),
			backt(text("echo")),
			backt(space)),
		"\\${VAR}`")
	t.CheckOutputLines(
		"WARN: filename.mk:1: Internal pkglint error " +
			"in ShTokenizer.ShAtom at \"\\\\${VAR}`\" (quoting=b).")
}

func (s *Suite) Test_ShTokenizer_ShAtom__quoting(c *check.C) {
	t := s.Init(c)

	test := func(input, expectedOutput string) {
		line := t.NewLine("filename.mk", 1, "")
		p := NewShTokenizer(line, input, false)
		q := shqPlain
		result := ""
		for {
			token := p.ShAtom(q)
			if token == nil {
				break
			}
			result += token.MkText
			if token.Quoting != q {
				q = token.Quoting
				result += "[" + q.String() + "]"
			}
		}
		t.CheckEquals(result, expectedOutput)
		t.CheckEquals(p.Rest(), "")
	}

	test("hello, world", "hello, world")
	test("hello, \"world\"", "hello, \"[d]world\"[plain]")
	test("1 \"\" 2 '' 3 `` 4", "1 \"[d]\"[plain] 2 '[s]'[plain] 3 `[b]`[plain] 4")
	test("\"\"", "\"[d]\"[plain]")
	test("''", "'[s]'[plain]")
	test("``", "`[b]`[plain]")
	test("x\"x`x`x\"x'x\"x'", "x\"[d]x`[db]x`[d]x\"[plain]x'[s]x\"x'[plain]")
	test("x\"x`x'x'x`x\"", "x\"[d]x`[db]x'[dbs]x'[db]x`[d]x\"[plain]")
	test("x\\\"x\\'x\\`x\\\\", "x\\\"x\\'x\\`x\\\\")
	test("x\"x\\\"x\\'x\\`x\\\\", "x\"[d]x\\\"x\\'x\\`x\\\\")
	test("x'x\\\"x\\'x\\`x\\\\", "x'[s]x\\\"x\\'[plain]x\\`x\\\\")
	test("x`x\\\"x\\'x\\`x\\\\", "x`[b]x\\\"x\\'x\\`x\\\\")
}

// The switch statement in ShTokenizer.ShAtom is exhaustive.
// If a new quoting mode is added (in which case the shell tokenizer
// should rather be rewritten completely and correctly), it is ok
// to panic if ShQuoting is not adjusted in the same commit.
func (s *Suite) Test_ShTokenizer_ShAtom__internal_error(c *check.C) {
	t := s.Init(c)

	line := t.NewLine("filename.mk", 123, "\ttoken")
	tok := NewShTokenizer(line, line.Text, true)
	t.ExpectPanicMatches(
		func() { tok.ShAtom(^ShQuoting(0)) },
		// Normalize the panic message, for Go < 12 if I remember correctly.
		`^runtime error: index out of range.*`)
}

func (s *Suite) Test_ShTokenizer_shVarUse(c *check.C) {
	t := s.Init(c)

	test := func(input string, output *ShAtom, rest string) {
		tok := NewShTokenizer(nil, input, false)
		actual := tok.shVarUse(shqPlain)

		t.CheckDeepEquals(actual, output)
		t.CheckEquals(tok.Rest(), rest)
	}

	shvar := func(text, varname string) *ShAtom {
		return &ShAtom{shtShVarUse, text, shqPlain, varname}
	}

	test("$", nil, "$")
	test("$$", nil, "$$")
	test("${MKVAR}", nil, "${MKVAR}")

	test("$$a", shvar("$$a", "a"), "")
	test("$$a.", shvar("$$a", "a"), ".")
	test("$$a_b_123:", shvar("$$a_b_123", "a_b_123"), ":")
	test("$$123", shvar("$$1", "1"), "23")

	test("$${varname}", shvar("$${varname}", "varname"), "")
	test("$${varname}.", shvar("$${varname}", "varname"), ".")
	test("$${0123}.", shvar("$${0123}", "0123"), ".")
	test("$${varname", nil, "$${varname")

	test("$${var:=value}", shvar("$${var:=value}", "var"), "")
	test("$${var#value}", shvar("$${var#value}", "var"), "")
	test("$${var##value}", shvar("$${var##value}", "var"), "")
	test("$${var##*}", shvar("$${var##*}", "var"), "")
	test("$${var%\".gz\"}", shvar("$${var%\".gz\"}", "var"), "")

	// TODO: allow variables in patterns.
	test("$${var%.${ext}}", nil, "$${var%.${ext}}")

	test("$${var##*", nil, "$${var##*")
	test("$${var\"", nil, "$${var\"")

	// TODO: test("$${var%${EXT}}", shvar("$${var%${EXT}}", "var"), "")
	test("$${var%${EXT}}", nil, "$${var%${EXT}}")

	// TODO: length of var
	test("$${#var}", nil, "$${#var}")

	test("$${/}", nil, "$${/}")
	test("$${\\}", nil, "$${\\}")
}

func (s *Suite) Test_ShTokenizer_ShToken(c *check.C) {
	t := s.Init(c)

	// testRest ensures that the given string is parsed to the expected
	// tokens, and returns the remaining text.
	testRest := func(str string, expected ...string) string {
		line := t.NewLine("testRest.mk", 1, "")
		p := NewShTokenizer(line, str, false)
		for _, exp := range expected {
			t.CheckEquals(p.ShToken().MkText, exp)
		}
		return p.Rest()
	}

	test := func(str string, expected ...string) {
		line := t.NewLine("test.mk", 1, "")
		p := NewShTokenizer(line, str, false)
		for _, exp := range expected {
			t.CheckEquals(p.ShToken().MkText, exp)
		}
		t.CheckEquals(p.Rest(), "")
		t.CheckOutputEmpty()
	}

	testNil := func(str string) {
		line := t.NewLine("testNil.mk", 1, "")
		p := NewShTokenizer(line, str, false)
		c.Check(p.ShToken(), check.IsNil)
		t.CheckEquals(p.Rest(), "")
		t.CheckOutputEmpty()
	}

	testNil("")
	testNil(" ")
	rest := testRest("\t\t\t\n\n\n\n\t ",
		"\n\n\n\n")
	t.CheckEquals(rest, "\t ")

	test("echo",
		"echo")

	test("`cat file`",
		"`cat file`")

	test("PAGES=\"`ls -1 | ${SED} -e 's,3qt$$,3,'`\"",
		"PAGES=\"`ls -1 | ${SED} -e 's,3qt$$,3,'`\"")

	test("echo hello, world",
		"echo",
		"hello,",
		"world")

	test("if cond1; then action1; elif cond2; then action2; else action3; fi",
		"if", "cond1", ";", "then",
		"action1", ";",
		"elif", "cond2", ";", "then",
		"action2", ";",
		"else", "action3", ";",
		"fi")

	test("PATH=/nonexistent env PATH=${PATH:Q} true",
		"PATH=/nonexistent",
		"env",
		"PATH=${PATH:Q}",
		"true")

	test("id=$$(id)",
		"id=$$(id)")

	test("id=$$(${AWK} '{print}' < ${WRKSRC}/idfile)",
		"id=$$(${AWK} '{print}' < ${WRKSRC}/idfile)")

	test("id=`${AWK} '{print}' < ${WRKSRC}/idfile`",
		"id=`${AWK} '{print}' < ${WRKSRC}/idfile`")
}