[PATCH 1/2] Handle shebang line, plus any whitespace and comment skipping in lexer

[Mark Wielaard <mark@klomp.org>]

The lexer tried to handle the shebang line but used loc directly,
instead of the current_column. And it assumed a '/' should immediately
follow the "#!". But if the "#!" is followed by whitespace and/or
comments and a '[' character, then the first line isn't see as a
shebang line (even if the kernel or shell would) but as the start of
an inner attribute.

Add various tests for when the first line starting with "#!" is seen
as a shebang line (and should be skipped). And some tests there is a
'[' character following some whitespace and/or comments and the "#!"
is seen as part of an inner attribute.
---
 gcc/rust/lex/rust-lex.cc                      | 79 ++++++++++++++-----
 .../rust/compile/torture/not_shebang.rs       |  3 +
 .../torture/not_shebang_block_comment.rs      |  1 +
 .../compile/torture/not_shebang_comment.rs    |  3 +
 .../torture/not_shebang_multiline_comment.rs  |  7 ++
 .../compile/torture/not_shebang_spaces.rs     |  6 ++
 gcc/testsuite/rust/compile/torture/shebang.rs |  3 +
 .../rust/compile/torture/shebang_plus_attr.rs |  3 +
 .../compile/torture/shebang_plus_attr2.rs     |  3 +
 9 files changed, 89 insertions(+), 19 deletions(-)
 create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang.rs
 create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang_block_comment.rs
 create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang_comment.rs
 create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang_multiline_comment.rs
 create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang_spaces.rs
 create mode 100755 gcc/testsuite/rust/compile/torture/shebang.rs
 create mode 100755 gcc/testsuite/rust/compile/torture/shebang_plus_attr.rs
 create mode 100755 gcc/testsuite/rust/compile/torture/shebang_plus_attr2.rs

diff --git a/gcc/rust/lex/rust-lex.cc b/gcc/rust/lex/rust-lex.cc
index d1384168731..ebd69de0fd1 100644
--- a/gcc/rust/lex/rust-lex.cc
+++ b/gcc/rust/lex/rust-lex.cc
@@ -237,28 +237,63 @@ Lexer::build_token ()
       current_char = peek_input ();
       skip_input ();
 
-      // return end of file token if end of file
-      if (current_char == EOF)
-	return Token::make (END_OF_FILE, loc);
-
       // detect shebang
-      if (loc == 1 && current_line == 1 && current_char == '#')
+      // Must be the first thing on the first line, starting with #!
+      // But since an attribute can also start with an #! we don't count it as a
+      // shebang line when after any whitespace or comments there is a [. If it
+      // is a shebang line we simple drop the line. Otherwise we don't consume
+      // any characters and fall through to the real tokenizer.
+      if (current_line == 1 && current_column == 1 && current_char == '#'
+	  && peek_input () == '!')
 	{
-	  current_char = peek_input ();
-
-	  if (current_char == '!')
+	  int n = 1;
+	  while (true)
 	    {
-	      skip_input ();
-	      current_char = peek_input ();
-
-	      if (current_char == '/')
+	      int next_char = peek_input (n);
+	      if (is_whitespace (next_char))
+		n++;
+	      else if (next_char == '/' && peek_input (n + 1) == '/')
 		{
-		  // definitely shebang
-
-		  skip_input ();
-
-		  // ignore rest of line
-		  while (current_char != '\n')
+		  // A single line comment
+		  n += 2;
+		  next_char = peek_input (n);
+		  while (next_char != '\n' && next_char != EOF)
+		    {
+		      n++;
+		      next_char = peek_input (n);
+		    }
+		  if (next_char == '\n')
+		    n++;
+		}
+	      else if (next_char == '/' && peek_input (n + 1) == '*')
+		{
+		  // Start of a block comment
+		  n += 2;
+		  int level = 1;
+		  while (level > 0)
+		    {
+		      if (peek_input (n) == EOF)
+			break;
+		      else if (peek_input (n) == '/'
+			       && peek_input (n + 1) == '*')
+			{
+			  n += 2;
+			  level += 1;
+			}
+		      else if (peek_input (n) == '*'
+			       && peek_input (n + 1) == '/')
+			{
+			  n += 2;
+			  level -= 1;
+			}
+		      else
+			n++;
+		    }
+		}
+	      else if (next_char != '[')
+		{
+		  // definitely shebang, ignore the first line
+		  while (current_char != '\n' && current_char != EOF)
 		    {
 		      current_char = peek_input ();
 		      skip_input ();
@@ -269,11 +304,17 @@ Lexer::build_token ()
 		  current_column = 1;
 		  // tell line_table that new line starts
 		  line_map->start_line (current_line, max_column_hint);
-		  continue;
+		  break;
 		}
+	      else
+		break; /* Definitely not a shebang line. */
 	    }
 	}
 
+      // return end of file token if end of file
+      if (current_char == EOF)
+	return Token::make (END_OF_FILE, loc);
+
       // if not end of file, start tokenising
       switch (current_char)
 	{
diff --git a/gcc/testsuite/rust/compile/torture/not_shebang.rs b/gcc/testsuite/rust/compile/torture/not_shebang.rs
new file mode 100644
index 00000000000..37e01b65940
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/not_shebang.rs
@@ -0,0 +1,3 @@
+#!
+[allow(unused)]
+fn main () { }
diff --git a/gcc/testsuite/rust/compile/torture/not_shebang_block_comment.rs b/gcc/testsuite/rust/compile/torture/not_shebang_block_comment.rs
new file mode 100644
index 00000000000..662f6506749
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/not_shebang_block_comment.rs
@@ -0,0 +1 @@
+#!/*/this/is/a/comment*/[allow(unused)] fn main () { }
diff --git a/gcc/testsuite/rust/compile/torture/not_shebang_comment.rs b/gcc/testsuite/rust/compile/torture/not_shebang_comment.rs
new file mode 100644
index 00000000000..273ae4e8e2a
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/not_shebang_comment.rs
@@ -0,0 +1,3 @@
+#!//this/is/a/comment
+[allow(unused)]   
+fn main () { }
diff --git a/gcc/testsuite/rust/compile/torture/not_shebang_multiline_comment.rs b/gcc/testsuite/rust/compile/torture/not_shebang_multiline_comment.rs
new file mode 100644
index 00000000000..86800b14cb3
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/not_shebang_multiline_comment.rs
@@ -0,0 +1,7 @@
+#!//this/is/a/comment
+
+/* Also a /* nested */
+   multiline // comment
+   with some more whitespace after, but then finally a [, so not a real #! line.  */
+
+[allow(unused)] fn main () { }
diff --git a/gcc/testsuite/rust/compile/torture/not_shebang_spaces.rs b/gcc/testsuite/rust/compile/torture/not_shebang_spaces.rs
new file mode 100644
index 00000000000..6b94a69111a
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/not_shebang_spaces.rs
@@ -0,0 +1,6 @@
+#!   
+
+    [allow(unused)]   
+
+        fn main () { }
+    
diff --git a/gcc/testsuite/rust/compile/torture/shebang.rs b/gcc/testsuite/rust/compile/torture/shebang.rs
new file mode 100755
index 00000000000..1c8b9c9a955
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/shebang.rs
@@ -0,0 +1,3 @@
+#!/usr/bin/env cat 
+
+fn main () { }
diff --git a/gcc/testsuite/rust/compile/torture/shebang_plus_attr.rs b/gcc/testsuite/rust/compile/torture/shebang_plus_attr.rs
new file mode 100755
index 00000000000..075bc6cf594
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/shebang_plus_attr.rs
@@ -0,0 +1,3 @@
+#!/usr/bin/env cat 
+#![allow(unused)]
+fn main () { }
diff --git a/gcc/testsuite/rust/compile/torture/shebang_plus_attr2.rs b/gcc/testsuite/rust/compile/torture/shebang_plus_attr2.rs
new file mode 100755
index 00000000000..ece8a52381c
--- /dev/null
+++ b/gcc/testsuite/rust/compile/torture/shebang_plus_attr2.rs
@@ -0,0 +1,3 @@
+#!//usr/bin/env cat 
+#![allow(unused)]
+fn main () { }
-- 
2.32.0