Windows: Add support for unicode command lines

Summary:
The MSVCRT deliberately sends main() code-page specific characters.
This isn't too useful to LLVM as we end up converting the arguments to
UTF-16 and subsequently attempt to use the result as, for example, a
file name.  Instead, we need to have the ability to access the Unicode
command line and transform it to UTF-8.

This has the distinct advantage over using the MSVC-specific wmain()
function as our entry point because:
 - It doesn't work on cygwin.
 - It only work on MinGW with caveats and only then on certain versions.
 - We get to keep our entry point as main(). :)

N.B.  This patch includes fixes to other parts of lib/Support/Windows
s.t. we would be able to take advantage of getting the Unicode paths.
E.G.  clang spawning clang -cc1 would want to give it Unicode arguments.

Reviewers: aaron.ballman, Bigcheese, rnk, ruiu

Reviewed By: rnk

CC: llvm-commits, ygao

Differential Revision: http://llvm-reviews.chandlerc.com/D1834

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@192069 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 72 insertions, 54 deletions

diff --git a/lib/Support/Windows/Program.inc b/lib/Support/Windows/Program.inc
index 2869085..e464e2f 100644
--- a/lib/Support/Windows/Program.inc
+++ b/lib/Support/Windows/Program.inc
@@ -42,42 +42,39 @@ std::string sys::FindProgramByName(const std::string &progName) {
 
   // At this point, the file name is valid and does not contain slashes.
   // Let Windows search for it.
-  std::string buffer;
-  buffer.resize(MAX_PATH);
-  char *dummy = NULL;
-  DWORD len = SearchPath(NULL, progName.c_str(), ".exe", MAX_PATH,
-                         &buffer[0], &dummy);
-
-  // See if it wasn't found.
-  if (len == 0)
+  SmallVector<wchar_t, MAX_PATH> progNameUnicode;
+  if (windows::UTF8ToUTF16(progName, progNameUnicode))
     return "";
 
-  // See if we got the entire path.
-  if (len < MAX_PATH)
-    return buffer;
+  SmallVector<wchar_t, MAX_PATH> buffer;
+  DWORD len = MAX_PATH;
+  do {
+    buffer.reserve(len);
+    len = ::SearchPathW(NULL, progNameUnicode.data(), L".exe",
+                        buffer.capacity(), buffer.data(), NULL);
 
-  // Buffer was too small; grow and retry.
-  while (true) {
-    buffer.resize(len+1);
-    DWORD len2 = SearchPath(NULL, progName.c_str(), ".exe", len+1, &buffer[0], &dummy);
-
-    // It is unlikely the search failed, but it's always possible some file
-    // was added or removed since the last search, so be paranoid...
-    if (len2 == 0)
+    // See if it wasn't found.
+    if (len == 0)
       return "";
-    else if (len2 <= len)
-      return buffer;
 
-    len = len2;
-  }
+    // Buffer was too small; grow and retry.
+  } while (len > buffer.capacity());
+
+  buffer.set_size(len);
+  SmallVector<char, MAX_PATH> result;
+  if (windows::UTF16ToUTF8(buffer.begin(), buffer.size(), result))
+    return "";
+
+  return std::string(result.data(), result.size());
 }
 
 static HANDLE RedirectIO(const StringRef *path, int fd, std::string* ErrMsg) {
   HANDLE h;
   if (path == 0) {
-    DuplicateHandle(GetCurrentProcess(), (HANDLE)_get_osfhandle(fd),
-                    GetCurrentProcess(), &h,
-                    0, TRUE, DUPLICATE_SAME_ACCESS);
+    if (!DuplicateHandle(GetCurrentProcess(), (HANDLE)_get_osfhandle(fd),
+                         GetCurrentProcess(), &h,
+                         0, TRUE, DUPLICATE_SAME_ACCESS))
+      return INVALID_HANDLE_VALUE;
     return h;
   }
 
@@ -92,9 +89,13 @@ static HANDLE RedirectIO(const StringRef *path, int fd, std::string* ErrMsg) {
   sa.lpSecurityDescriptor = 0;
   sa.bInheritHandle = TRUE;
 
-  h = CreateFile(fname.c_str(), fd ? GENERIC_WRITE : GENERIC_READ,
-                 FILE_SHARE_READ, &sa, fd == 0 ? OPEN_EXISTING : CREATE_ALWAYS,
-                 FILE_ATTRIBUTE_NORMAL, NULL);
+  SmallVector<wchar_t, 128> fnameUnicode;
+  if (windows::UTF8ToUTF16(fname, fnameUnicode))
+    return INVALID_HANDLE_VALUE;
+
+  h = CreateFileW(fnameUnicode.data(), fd ? GENERIC_WRITE : GENERIC_READ,
+                  FILE_SHARE_READ, &sa, fd == 0 ? OPEN_EXISTING : CREATE_ALWAYS,
+                  FILE_ATTRIBUTE_NORMAL, NULL);
   if (h == INVALID_HANDLE_VALUE) {
     MakeErrMsg(ErrMsg, std::string(fname) + ": Can't open file for " +
         (fd ? "input: " : "output: "));
@@ -218,34 +219,28 @@ static bool Execute(ProcessInfo &PI, StringRef Program, const char **args,
   *p = 0;
 
   // The pointer to the environment block for the new process.
-  OwningArrayPtr<char> envblock;
+  std::vector<wchar_t> EnvBlock;
 
   if (envp) {
     // An environment block consists of a null-terminated block of
     // null-terminated strings. Convert the array of environment variables to
     // an environment block by concatenating them.
+    for (unsigned i = 0; envp[i]; ++i) {
+      SmallVector<wchar_t, MAX_PATH> EnvString;
+      if (error_code ec = windows::UTF8ToUTF16(envp[i], EnvString)) {
+        SetLastError(ec.value());
+        MakeErrMsg(ErrMsg, "Unable to convert environment variable to UTF-16");
+        return false;
+      }
 
-    // First, determine the length of the environment block.
-    len = 0;
-    for (unsigned i = 0; envp[i]; i++)
-      len += strlen(envp[i]) + 1;
-
-    // Now build the environment block.
-    envblock.reset(new char[len+1]);
-    p = envblock.get();
-
-    for (unsigned i = 0; envp[i]; i++) {
-      const char *ev = envp[i];
-      size_t len = strlen(ev) + 1;
-      memcpy(p, ev, len);
-      p += len;
+      EnvBlock.insert(EnvBlock.end(), EnvString.begin(), EnvString.end());
+      EnvBlock.push_back(0);
     }
-
-    *p = 0;
+    EnvBlock.push_back(0);
   }
 
   // Create a child process.
-  STARTUPINFO si;
+  STARTUPINFOW si;
   memset(&si, 0, sizeof(si));
   si.cb = sizeof(si);
   si.hStdInput = INVALID_HANDLE_VALUE;
@@ -269,9 +264,14 @@ static bool Execute(ProcessInfo &PI, StringRef Program, const char **args,
     if (redirects[1] && redirects[2] && *(redirects[1]) == *(redirects[2])) {
       // If stdout and stderr should go to the same place, redirect stderr
       // to the handle already open for stdout.
-      DuplicateHandle(GetCurrentProcess(), si.hStdOutput,
-                      GetCurrentProcess(), &si.hStdError,
-                      0, TRUE, DUPLICATE_SAME_ACCESS);
+      if (!DuplicateHandle(GetCurrentProcess(), si.hStdOutput,
+                           GetCurrentProcess(), &si.hStdError,
+                           0, TRUE, DUPLICATE_SAME_ACCESS)) {
+        CloseHandle(si.hStdInput);
+        CloseHandle(si.hStdOutput);
+        MakeErrMsg(ErrMsg, "can't dup stderr to stdout");
+        return false;
+      }
     } else {
       // Just redirect stderr
       si.hStdError = RedirectIO(redirects[2], 2, ErrMsg);
@@ -289,9 +289,27 @@ static bool Execute(ProcessInfo &PI, StringRef Program, const char **args,
 
   fflush(stdout);
   fflush(stderr);
-  std::string ProgramStr = Program;
-  BOOL rc = CreateProcess(ProgramStr.c_str(), command.get(), NULL, NULL, TRUE,
-                          0, envblock.get(), NULL, &si, &pi);
+
+  SmallVector<wchar_t, MAX_PATH> ProgramUtf16;
+  if (error_code ec = windows::UTF8ToUTF16(Program, ProgramUtf16)) {
+    SetLastError(ec.value());
+    MakeErrMsg(ErrMsg,
+               std::string("Unable to convert application name to UTF-16"));
+    return false;
+  }
+
+  SmallVector<wchar_t, MAX_PATH> CommandUtf16;
+  if (error_code ec = windows::UTF8ToUTF16(command.get(), CommandUtf16)) {
+    SetLastError(ec.value());
+    MakeErrMsg(ErrMsg,
+               std::string("Unable to convert command-line to UTF-16"));
+    return false;
+  }
+
+  BOOL rc = CreateProcessW(ProgramUtf16.data(), CommandUtf16.data(), 0, 0,
+                           TRUE, CREATE_UNICODE_ENVIRONMENT,
+                           EnvBlock.empty() ? 0 : EnvBlock.data(), 0, &si,
+                           &pi);
   DWORD err = GetLastError();
 
   // Regardless of whether the process got created or not, we are done with
@@ -304,7 +322,7 @@ static bool Execute(ProcessInfo &PI, StringRef Program, const char **args,
   if (!rc) {
     SetLastError(err);
     MakeErrMsg(ErrMsg, std::string("Couldn't execute program '") +
-               ProgramStr + "'");
+               Program.str() + "'");
     return false;
   }