From 1f13fbd0c5bac045aa7d6ae3e998758e9dfa3ce4 Mon Sep 17 00:00:00 2001
From: Jordan Rose <jordan_rose@apple.com>
Date: Fri, 7 Jun 2013 17:16:01 +0000
Subject: [PATCH] Handle Unicode characters in fix-it replacement strings.

Patch by Sukolsak Sakshuwong!

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@183535 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Frontend/TextDiagnostic.cpp | 36 ++++++++++++++++-----------------
 test/FixIt/fixit-unicode.c      | 20 ++++++++++++++++++
 2 files changed, 37 insertions(+), 19 deletions(-)

diff --git a/lib/Frontend/TextDiagnostic.cpp b/lib/Frontend/TextDiagnostic.cpp
index 1572d0f1d0..2da66d3b32 100644
--- a/lib/Frontend/TextDiagnostic.cpp
+++ b/lib/Frontend/TextDiagnostic.cpp
@@ -336,13 +336,10 @@ static void selectInterestingSourceRegion(std::string &SourceLine,
   if (MaxColumns <= Columns)
     return;
 
-  // no special characters allowed in CaretLine or FixItInsertionLine
+  // No special characters are allowed in CaretLine.
   assert(CaretLine.end() ==
          std::find_if(CaretLine.begin(), CaretLine.end(),
          char_out_of_range(' ','~')));
-  assert(FixItInsertionLine.end() ==
-         std::find_if(FixItInsertionLine.begin(), FixItInsertionLine.end(),
-         char_out_of_range(' ','~')));
 
   // Find the slice that we need to display the full caret line
   // correctly.
@@ -370,8 +367,15 @@ static void selectInterestingSourceRegion(std::string &SourceLine,
       if (!isWhitespace(FixItInsertionLine[FixItEnd - 1]))
         break;
 
-    CaretStart = std::min(FixItStart, CaretStart);
-    CaretEnd = std::max(FixItEnd, CaretEnd);
+    // We can safely use the byte offset FixItStart as the column offset
+    // because the characters up until FixItStart are all ASCII whitespace
+    // characters.
+    unsigned FixItStartCol = FixItStart;
+    unsigned FixItEndCol
+      = llvm::sys::locale::columnWidth(FixItInsertionLine.substr(0, FixItEnd));
+
+    CaretStart = std::min(FixItStartCol, CaretStart);
+    CaretEnd = std::max(FixItEndCol, CaretEnd);
   }
 
   // CaretEnd may have been set at the middle of a character
@@ -1023,24 +1027,18 @@ static std::string buildFixItInsertionLine(unsigned LineNo,
         if (HintCol < PrevHintEndCol)
           HintCol = PrevHintEndCol + 1;
 
-        // FIXME: This function handles multibyte characters in the source, but
-        // not in the fixits. This assertion is intended to catch unintended
-        // use of multibyte characters in fixits. If we decide to do this, we'll
-        // have to track separate byte widths for the source and fixit lines.
-        assert((size_t)llvm::sys::locale::columnWidth(I->CodeToInsert) ==
-               I->CodeToInsert.size());
-
-        // This relies on one byte per column in our fixit hints.
         // This should NOT use HintByteOffset, because the source might have
         // Unicode characters in earlier columns.
-        unsigned LastColumnModified = HintCol + I->CodeToInsert.size();
-        if (LastColumnModified > FixItInsertionLine.size())
-          FixItInsertionLine.resize(LastColumnModified, ' ');
+        unsigned NewFixItLineSize = FixItInsertionLine.size() +
+          (HintCol - PrevHintEndCol) + I->CodeToInsert.size();
+        if (NewFixItLineSize > FixItInsertionLine.size())
+          FixItInsertionLine.resize(NewFixItLineSize, ' ');
 
         std::copy(I->CodeToInsert.begin(), I->CodeToInsert.end(),
-                  FixItInsertionLine.begin() + HintCol);
+                  FixItInsertionLine.end() - I->CodeToInsert.size());
 
-        PrevHintEndCol = LastColumnModified;
+        PrevHintEndCol =
+          HintCol + llvm::sys::locale::columnWidth(I->CodeToInsert);
       } else {
         FixItInsertionLine.clear();
         break;
diff --git a/test/FixIt/fixit-unicode.c b/test/FixIt/fixit-unicode.c
index 9c0242e92e..216b73dbfc 100644
--- a/test/FixIt/fixit-unicode.c
+++ b/test/FixIt/fixit-unicode.c
@@ -34,3 +34,23 @@ void test2() {
 
 // CHECK-MACHINE: fix-it:"{{.*}}fixit-unicode.c":{[[@LINE-9]]:16-[[@LINE-9]]:18}:"%ld"
 }
+
+void test3() {
+  int กssss = 42;
+  int a = กsss; // expected-error{{use of undeclared identifier 'กsss'; did you mean 'กssss'?}}
+// CHECK: {{^          \^}}
+// CHECK: {{^          [^ ]+ssss}}
+// CHECK-MACHINE: fix-it:"{{.*}}":{[[@LINE-3]]:11-[[@LINE-3]]:17}:"\340\270\201ssss"
+
+  int ssกss = 42;
+  int b = ssกs; // expected-error{{use of undeclared identifier 'ssกs'; did you mean 'ssกss'?}}
+// CHECK: {{^          \^}}
+// CHECK: {{^          ss.+ss}}
+// CHECK-MACHINE: fix-it:"{{.*}}":{[[@LINE-3]]:11-[[@LINE-3]]:17}:"ss\340\270\201ss"
+
+  int sssssssssก = 42;
+  int c = sssssssss; // expected-error{{use of undeclared identifier 'sssssssss'; did you mean 'sssssssssก'?}}
+// CHECK: {{^          \^}}
+// CHECK: {{^          sssssssss.+}}
+// CHECK-MACHINE: fix-it:"{{.*}}":{[[@LINE-3]]:11-[[@LINE-3]]:20}:"sssssssss\340\270\201"
+}
-- 
2.50.1