[PATCH 107/125] gccrs: Improve parsing of raw string literals

[Arthur Cohen]

From: Owen Avery <powerboat9.ga...@gmail.com>

gcc/rust/ChangeLog:

        * lex/rust-lex.cc
        (Lexer::parse_raw_string):
        Bring handling of edge cases to par with parse_raw_byte_string.

gcc/testsuite/ChangeLog:

        * rust/compile/raw-string-loc.rs: New test.

Signed-off-by: Owen Avery <powerboat9.ga...@gmail.com>
---
 gcc/rust/lex/rust-lex.cc                     | 21 +++++++++++++++++---
 gcc/testsuite/rust/compile/raw-string-loc.rs |  6 ++++++
 2 files changed, 24 insertions(+), 3 deletions(-)
 create mode 100644 gcc/testsuite/rust/compile/raw-string-loc.rs

diff --git a/gcc/rust/lex/rust-lex.cc b/gcc/rust/lex/rust-lex.cc
index 7c37e83d6cb..e5c9148976c 100644
--- a/gcc/rust/lex/rust-lex.cc
+++ b/gcc/rust/lex/rust-lex.cc
@@ -2152,6 +2152,9 @@ Lexer::parse_raw_string (location_t loc, int 
initial_hash_count)
   str.reserve (16); // some sensible default
 
   int length = 1 + initial_hash_count;
+  current_column += length;
+
+  const location_t string_begin_locus = get_current_location ();
 
   if (initial_hash_count > 0)
     skip_input (initial_hash_count - 1);
@@ -2162,10 +2165,11 @@ Lexer::parse_raw_string (location_t loc, int 
initial_hash_count)
     rust_error_at (get_current_location (), "raw string has no opening 
%<\"%>");
 
   length++;
+  current_column++;
   skip_input ();
   current_char = peek_input ();
 
-  while (!current_char.is_eof ())
+  while (true)
     {
       if (current_char.value == '"')
        {
@@ -2186,19 +2190,30 @@ Lexer::parse_raw_string (location_t loc, int 
initial_hash_count)
              skip_input (initial_hash_count);
              current_char = peek_input ();
              length += initial_hash_count + 1;
+             current_column += initial_hash_count + 1;
              break;
            }
        }
+      else if (current_char.is_eof ())
+       {
+         rust_error_at (string_begin_locus, "unended raw string literal");
+         return Token::make (END_OF_FILE, get_current_location ());
+       }
 
       length++;
+      current_column++;
+      if (current_char == '\n')
+       {
+         current_line++;
+         current_column = 1;
+         start_line (current_line, max_column_hint);
+       }
 
       str += current_char.as_string ();
       skip_input ();
       current_char = peek_input ();
     }
 
-  current_column += length;
-
   loc += length - 1;
 
   str.shrink_to_fit ();
diff --git a/gcc/testsuite/rust/compile/raw-string-loc.rs 
b/gcc/testsuite/rust/compile/raw-string-loc.rs
new file mode 100644
index 00000000000..70977510ba3
--- /dev/null
+++ b/gcc/testsuite/rust/compile/raw-string-loc.rs
@@ -0,0 +1,6 @@
+const X: &'static str = r#"12
+12"#;
+
+BREAK
+// { dg-error "unrecognised token" "" { target *-*-* } .-1 }
+// { dg-excess-errors "error 'failed to parse item' does not have location" }
-- 
2.45.2