From 694673f1d7981907a6e0695ae1d9cb1bdec136f4 Mon Sep 17 00:00:00 2001
From: Yu-hsin Wang <yuhsingw@google.com>
Date: Tue, 30 May 2023 13:27:01 +0800
Subject: [PATCH] arch: set multiline re as default in isa_parser

In python3.11, it requires the global specifier should be the first
token of regex. However it's not possible when using ply library.
Instead, we set the rules are multiline regex by default and modifies
those single line rules.

Ref: https://github.com/dabeaz/ply/issues/282

Change-Id: I7bdbfeb97a9dd74f45c1890a76f8cc16100e5a42
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/71019
Reviewed-by: Richard Cooper <richard.cooper@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
---
 src/arch/isa_parser/isa_parser.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/src/arch/isa_parser/isa_parser.py b/src/arch/isa_parser/isa_parser.py
index 0f29840c3b..5be50a11bf 100755
--- a/src/arch/isa_parser/isa_parser.py
+++ b/src/arch/isa_parser/isa_parser.py
@@ -514,6 +514,7 @@ class InstObjParams(object):
 class ISAParser(Grammar):
     def __init__(self, output_dir):
         super().__init__()
+        self.lex_kwargs["reflags"] = int(re.MULTILINE)
         self.output_dir = output_dir
 
         self.filename = None  # for output file watermarking/scaremongering
@@ -851,7 +852,7 @@ class ISAParser(Grammar):
     # String literal.  Note that these use only single quotes, and
     # can span multiple lines.
     def t_STRLIT(self, t):
-        r"(?m)'([^'])+'"
+        r"'([^'])+'"
         # strip off quotes
         t.value = t.value[1:-1]
         t.lexer.lineno += t.value.count("\n")
@@ -860,19 +861,19 @@ class ISAParser(Grammar):
     # "Code literal"... like a string literal, but delimiters are
     # '{{' and '}}' so they get formatted nicely under emacs c-mode
     def t_CODELIT(self, t):
-        r"(?m)\{\{([^\}]|}(?!\}))+\}\}"
+        r"\{\{([^\}]|}(?!\}))+\}\}"
         # strip off {{ & }}
         t.value = t.value[2:-2]
         t.lexer.lineno += t.value.count("\n")
         return t
 
     def t_CPPDIRECTIVE(self, t):
-        r"^\#[^\#].*\n"
+        r"^\#[^\#][^\n]*\n"
         t.lexer.lineno += t.value.count("\n")
         return t
 
     def t_NEWFILE(self, t):
-        r'^\#\#newfile\s+"[^"]*"\n'
+        r'^\#\#newfile\s+"[^"\n]*"\n'
         self.fileNameStack.push(t.lexer.lineno)
         t.lexer.lineno = LineTracker(t.value[11:-2])
 
@@ -892,7 +893,7 @@ class ISAParser(Grammar):
 
     # Comments
     def t_comment(self, t):
-        r"//.*"
+        r"//[^\n]*\n"
 
     # Completely ignored characters
     t_ignore = " \t\x0c"