From ed119820f223401b0c64769759eafc46de30a9da Mon Sep 17 00:00:00 2001
From: Yunzhong Gao <Yunzhong_Gao@playstation.sony.com>
Date: Thu, 5 Sep 2013 19:14:26 +0000
Subject: [PATCH] Improve handling of .file, .include and .incbin directives to
 allow escaped octal character sequences.

The patch was discussed in Phabricator. See:
http://llvm-reviews.chandlerc.com/D1289



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190089 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/MC/MCParser/AsmParser.cpp         | 28 +++++++++++++++------------
 test/MC/AsmParser/directive_file.s    |  2 +-
 test/MC/AsmParser/directive_incbin.s  |  2 +-
 test/MC/AsmParser/directive_include.s |  2 +-
 4 files changed, 19 insertions(+), 15 deletions(-)

diff --git a/lib/MC/MCParser/AsmParser.cpp b/lib/MC/MCParser/AsmParser.cpp
index 31a09a967c6..ae88a3ee73c 100644
--- a/lib/MC/MCParser/AsmParser.cpp
+++ b/lib/MC/MCParser/AsmParser.cpp
@@ -2550,17 +2550,21 @@ bool AsmParser::ParseDirectiveFile(SMLoc DirectiveLoc) {
     return TokError("unexpected token in '.file' directive");
 
   // Usually the directory and filename together, otherwise just the directory.
-  StringRef Path = getTok().getString();
-  Path = Path.substr(1, Path.size()-2);
+  // Allow the strings to have escaped octal character sequence.
+  std::string Path = getTok().getString();
+  if (parseEscapedString(Path))
+    return true;
   Lex();
 
   StringRef Directory;
   StringRef Filename;
+  std::string FilenameData;
   if (getLexer().is(AsmToken::String)) {
     if (FileNumber == -1)
       return TokError("explicit path specified, but no file number");
-    Filename = getTok().getString();
-    Filename = Filename.substr(1, Filename.size()-2);
+    if (parseEscapedString(FilenameData))
+      return true;
+    Filename = FilenameData;
     Directory = Path;
     Lex();
   } else {
@@ -3496,16 +3500,16 @@ bool AsmParser::ParseDirectiveInclude() {
   if (getLexer().isNot(AsmToken::String))
     return TokError("expected string in '.include' directive");
 
-  std::string Filename = getTok().getString();
+  // Allow the strings to have escaped octal character sequence.
+  std::string Filename;
+  if (parseEscapedString(Filename))
+    return true;
   SMLoc IncludeLoc = getLexer().getLoc();
   Lex();
 
   if (getLexer().isNot(AsmToken::EndOfStatement))
     return TokError("unexpected token in '.include' directive");
 
-  // Strip the quotes.
-  Filename = Filename.substr(1, Filename.size()-2);
-
   // Attempt to switch the lexer to the included file before consuming the end
   // of statement to avoid losing it when we switch.
   if (EnterIncludeFile(Filename)) {
@@ -3522,16 +3526,16 @@ bool AsmParser::ParseDirectiveIncbin() {
   if (getLexer().isNot(AsmToken::String))
     return TokError("expected string in '.incbin' directive");
 
-  std::string Filename = getTok().getString();
+  // Allow the strings to have escaped octal character sequence.
+  std::string Filename;
+  if (parseEscapedString(Filename))
+    return true;
   SMLoc IncbinLoc = getLexer().getLoc();
   Lex();
 
   if (getLexer().isNot(AsmToken::EndOfStatement))
     return TokError("unexpected token in '.incbin' directive");
 
-  // Strip the quotes.
-  Filename = Filename.substr(1, Filename.size()-2);
-
   // Attempt to process the included file.
   if (ProcessIncbinFile(Filename)) {
     Error(IncbinLoc, "Could not find incbin file '" + Filename + "'");
diff --git a/test/MC/AsmParser/directive_file.s b/test/MC/AsmParser/directive_file.s
index 121890e69a4..9b99e0f24e9 100644
--- a/test/MC/AsmParser/directive_file.s
+++ b/test/MC/AsmParser/directive_file.s
@@ -1,7 +1,7 @@
 # RUN: llvm-mc -triple i386-unknown-unknown %s | FileCheck %s
 
         .file "hello"
-        .file 1 "world"
+        .file 1 "worl\144"   # "\144" is "d"
         .file 2 "directory" "file"
 
 # CHECK: .file "hello"
diff --git a/test/MC/AsmParser/directive_incbin.s b/test/MC/AsmParser/directive_incbin.s
index 55f9c7951ff..ed4e27a2c1a 100644
--- a/test/MC/AsmParser/directive_incbin.s
+++ b/test/MC/AsmParser/directive_incbin.s
@@ -1,6 +1,6 @@
 # RUN: llvm-mc -triple i386-unknown-unknown %s -I %p | FileCheck %s
 
 .data
-.incbin "incbin_abcd"
+.incbin "incbin\137abcd"  # "\137" is underscore "_"
 
 # CHECK: .ascii	 "abcd\n"
diff --git a/test/MC/AsmParser/directive_include.s b/test/MC/AsmParser/directive_include.s
index fabd941d999..f53bc671fc6 100644
--- a/test/MC/AsmParser/directive_include.s
+++ b/test/MC/AsmParser/directive_include.s
@@ -5,5 +5,5 @@
 # CHECK: a = 0
 # CHECK: TESTB:
 TESTA:  
-	.include       "directive_set.s"
+	.include       "directive\137set.s"   # "\137" is underscore "_"
 TESTB: