Consolidate file handle usage in Editline.cpp

Summary:
To implement wide character reading, editline was mixing FILE*-based access with
a Connection-based one (plus it did some selects on the raw FD), which is very
fragile. Here, I replace it with one which uses only a Connection-based reads.
The code is somewhat longer as I had to read characters one by one to detect the
end of the multibyte sequence.

I've verified that international characters still work in lldb command line on
OSX.

Reviewers: clayborg, zturner

Subscribers: lldb-commits

Differential Revision: https://reviews.llvm.org/D28356

llvm-svn: 291220

1 files changed, 37 insertions, 10 deletions

diff --git a/lldb/source/Host/common/Editline.cpp b/lldb/source/Host/common/Editline.cpp
index a600c61c8e6..1c5c0ffe902 100644
--- a/lldb/source/Host/common/Editline.cpp
+++ b/lldb/source/Host/common/Editline.cpp
@@ -526,17 +526,8 @@ int Editline::GetCharacter(EditLineCharType *c) {
     }
 
     if (read_count) {
-#if LLDB_EDITLINE_USE_WCHAR
-      // After the initial interruptible read, this is guaranteed not to block
-      ungetc(ch, m_input_file);
-      *c = fgetwc(m_input_file);
-      if (*c != WEOF)
-        return 1;
-#else
-      *c = ch;
-      if (ch != (char)EOF)
+      if (CompleteCharacter(ch, *c))
         return 1;
-#endif
     } else {
       switch (status) {
       case lldb::eConnectionStatusSuccess: // Success
@@ -1367,3 +1358,39 @@ void Editline::PrintAsync(Stream *stream, const char *s, size_t len) {
     MoveCursor(CursorLocation::BlockEnd, CursorLocation::EditingCursor);
   }
 }
+
+bool Editline::CompleteCharacter(char ch, EditLineCharType &out) {
+#if !LLDB_EDITLINE_USE_WCHAR
+  if (ch == (char)EOF)
+    return false;
+
+  out = ch;
+  return true;
+#else
+  std::codecvt_utf8<wchar_t> cvt;
+  llvm::SmallString<4> input;
+  for (;;) {
+    const char *from_next;
+    wchar_t *to_next;
+    std::mbstate_t state = std::mbstate_t();
+    input.push_back(ch);
+    switch (cvt.in(state, input.begin(), input.end(), from_next, &out, &out + 1,
+                   to_next)) {
+    case std::codecvt_base::ok:
+      return out != WEOF;
+
+    case std::codecvt_base::error:
+    case std::codecvt_base::noconv:
+      return false;
+
+    case std::codecvt_base::partial:
+      lldb::ConnectionStatus status;
+      size_t read_count = m_input_connection.Read(
+          &ch, 1, std::chrono::seconds(0), status, nullptr);
+      if (read_count == 0)
+        return false;
+      break;
+    }
+  }
+#endif
+}