gc: reject import paths containing special characters

Also allow multiple invalid import statements in a
single file.

Fixes #3021. The changes to go/parser and the
language specifcation have already been committed.

R=rsc, gri
CC=golang-dev
https://golang.org/cl/5672084

src/cmd/gc/go.h

@@ -1171,6 +1171,7 @@ Type*	getthisx(Type *t);
 int	implements(Type *t, Type *iface, Type **missing, Type **have, int *ptr);
 void	importdot(Pkg *opkg, Node *pack);
 int	is64(Type *t);
+int	isbadimport(Strlit *s);
 int	isblank(Node *n);
 int	isblanksym(Sym *s);
 int	isfixedarray(Type *t);

src/cmd/gc/go.y

@@ -205,7 +205,15 @@ import_stmt:
 		my->lastlineno = $1;
 		my->block = 1;	// at top level
 	}
-
+|	import_here import_there
+	{
+		// When an invalid import path is passed to importfile,
+		// it calls yyerror and then sets up a fake import with
+		// no package statement. This allows us to test more
+		// than one invalid import statement in a single file.
+		if(nerrors == 0)
+			fatal("phase error in import");
+	}
 
 import_stmt_list:
 	import_stmt

src/cmd/gc/lex.c

@@ -573,6 +573,13 @@ findpkg(Strlit *name)
 	return 0;
 }
 
+static void
+fakeimport(void)
+{
+	importpkg = mkpkg(strlit("fake"));
+	cannedimports("fake.6", "$$\n");
+}
+
 void
 importfile(Val *f, int line)
 {
@@ -589,17 +596,19 @@ importfile(Val *f, int line)
 
 	if(f->ctype != CTSTR) {
 		yyerror("import statement not a string");
+		fakeimport();
 		return;
 	}
 
-	if(strlen(f->u.sval->s) != f->u.sval->len) {
-		yyerror("import path contains NUL");
-		errorexit();
+	if(f->u.sval->len == 0) {
+		yyerror("import path is empty");
+		fakeimport();
+		return;
 	}
-	
-	if(strchr(f->u.sval->s, '\\')) {
-		yyerror("import path contains backslash; use slash");
-		errorexit();
+
+	if(isbadimport(f->u.sval)) {
+		fakeimport();
+		return;
 	}
 
 	// The package name main is no longer reserved,

src/cmd/gc/subr.c

@@ -3559,11 +3559,9 @@ mkpkg(Strlit *path)
 {
 	Pkg *p;
 	int h;
-	
-	if(strlen(path->s) != path->len) {
-		yyerror("import path contains NUL byte");
+
+	if(isbadimport(path))
 		errorexit();
-	}
 	
 	h = stringhash(path->s) & (nelem(phash)-1);
 	for(p=phash[h]; p; p=p->link)
@@ -3612,3 +3610,41 @@ addinit(Node **np, NodeList *init)
 	n->ninit = concat(init, n->ninit);
 	n->ullman = UINF;
 }
+
+int
+isbadimport(Strlit *path)
+{
+	char *s;
+	Rune r;
+
+	if(strlen(path->s) != path->len) {
+		yyerror("import path contains NUL");
+		return 1;
+	}
+
+	s = path->s;
+	while(*s) {
+		s += chartorune(&r, s);
+		if(r == Runeerror) {
+			yyerror("import path contains invalid UTF-8 sequence");
+			return 1;
+		}
+		if(r < 0x20 || r == 0x7f) {
+			yyerror("import path contains control character");
+			return 1;
+		}
+		if(r == '\\') {
+			yyerror("import path contains backslash; use slash");
+			return 1;
+		}
+		if(isspacerune(r)) {
+			yyerror("import path contains space character");
+			return 1;
+		}
+		if(utfrune("!\"#$%&'()*,:;<=>?[]^`{|}~", r)) {
+			yyerror("import path contains invalid character '%C'", r);
+			return 1;
+		}
+	}
+	return 0;
+}

src/cmd/gc/y.tab.h

@@ -1,10 +1,8 @@
+/* A Bison parser, made by GNU Bison 2.5.  */
 
-/* A Bison parser, made by GNU Bison 2.4.1.  */
-
-/* Skeleton interface for Bison's Yacc-like parsers in C
+/* Bison interface for Yacc-like parsers in C
    
-      Copyright (C) 1984, 1989, 1990, 2000, 2001, 2002, 2003, 2004, 2005, 2006
-   Free Software Foundation, Inc.
+      Copyright (C) 1984, 1989-1990, 2000-2011 Free Software Foundation, Inc.
    
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -148,7 +146,7 @@
 typedef union YYSTYPE
 {
 
-/* Line 1676 of yacc.c  */
+/* Line 2068 of yacc.c  */
 #line 28 "go.y"
 
 	Node*		node;
@@ -160,8 +158,8 @@ typedef union YYSTYPE
 
 
 
-/* Line 1676 of yacc.c  */
-#line 165 "y.tab.h"
+/* Line 2068 of yacc.c  */
+#line 163 "y.tab.h"
 } YYSTYPE;
 # define YYSTYPE_IS_TRIVIAL 1
 # define yystype YYSTYPE /* obsolescent; will be withdrawn */

src/cmd/gc/yerr.h

@@ -14,25 +14,25 @@ static struct {
 	// is converted by bisonerrors into the yystate and yychar caused
 	// by that token list.
 
-	220, ',',
+	221, ',',
 	"unexpected comma during import block",
 
-	376, ';',
+	377, ';',
 	"unexpected semicolon or newline before {",
 
-	397, ';',
+	398, ';',
 	"unexpected semicolon or newline before {",
 
-	236, ';',
+	237, ';',
 	"unexpected semicolon or newline before {",
 
-	473, LBODY,
+	474, LBODY,
 	"unexpected semicolon or newline before {",
 
 	22, '{',
 	"unexpected semicolon or newline before {",
 
-	143, ';',
+	144, ';',
 	"unexpected semicolon or newline in type declaration",
 
 	37, '}',
@@ -44,30 +44,30 @@ static struct {
 	37, ',',
 	"unexpected comma in channel type",
 
-	436, LELSE,
+	437, LELSE,
 	"unexpected semicolon or newline before else",
 
-	256, ',',
+	257, ',',
 	"name list not allowed in interface type",
 
-	236, LVAR,
+	237, LVAR,
 	"var declaration not allowed in for initializer",
 
 	65, '{',
 	"unexpected { at end of statement",
 
-	375, '{',
+	376, '{',
 	"unexpected { at end of statement",
 	
-	124, ';',
+	125, ';',
 	"argument to go/defer must be function call",
 	
-	424, ';',
+	425, ';',
 	"need trailing comma before newline in composite literal",
 	
-	111, LNAME,
+	112, LNAME,
 	"nested func not allowed",
 
-	614, ';',
+	615, ';',
 	"else must be followed by if or statement block"
 };

test/import5.go

@@ -4,10 +4,48 @@
 // Use of this source code is governed by a BSD-style
 // license that can be found in the LICENSE file.
 
-// Verify that imports with backslashes are rejected by the compiler.
+// Verify that invalid imports are rejected by the compiler.
 // Does not compile.
-// TODO: make more thorough.
 
 package main
 
-import `net\http`  // ERROR "backslash"
+// Correct import paths.
+import _ "fmt"
+import _ `time`
+import _ "m\x61th"
+import _ "go/parser"
+
+// Correct import paths, but the packages don't exist.
+// Don't test.
+//import "a.b"
+//import "greek/αβ"
+
+// Import paths must be strings.
+import 42    // ERROR "import statement"
+import 'a'   // ERROR "import statement"
+import 3.14  // ERROR "import statement"
+import 0.25i // ERROR "import statement"
+
+// Each of these pairs tests both `` vs "" strings
+// and also use of invalid characters spelled out as
+// escape sequences and written directly.
+// For example `"\x00"` tests import "\x00"
+// while "`\x00`" tests import `<actual-NUL-byte>`.
+import ""         // ERROR "import path"
+import ``         // ERROR "import path"
+import "\x00"     // ERROR "import path"
+import `\x00`     // ERROR "import path"
+import "\x7f"     // ERROR "import path"
+import `\x7f`     // ERROR "import path"
+import "a!"       // ERROR "import path"
+import `a!`       // ERROR "import path"
+import "a b"      // ERROR "import path"
+import `a b`      // ERROR "import path"
+import "a\\b"     // ERROR "import path"
+import `a\\b`     // ERROR "import path"
+import "\"`a`\""  // ERROR "import path"
+import `\"a\"`    // ERROR "import path"
+import "\x80\x80" // ERROR "import path"
+import `\x80\x80` // ERROR "import path"
+import "\xFFFD"   // ERROR "import path"
+import `\xFFFD`   // ERROR "import path"