Bazel client, JNI, Windows: impl. CreateJunction

Implement a CreateJunction function in the Windows
JNI library. Also move a bit of code from
file_windows to the JNI library, where it is
(also) needed.

This implementation is an improved version of
`blaze_util::SymlinkDirectories` in
blaze_util_windows: this version handles Windows
paths as `name` and `target`, and performs more
validation (e.g. on the length of `target`), plus
has more comments explaining the logic. In a
subsequent change I'll start using this new
function in blaze_util_windows.

This method will also be helpful in tests: we will
no longer have to shell out to mklink.

See https://github.com/bazelbuild/bazel/issues/2107

--
Change-Id: I7e9b085fdc2ba47be83da5319bded02bd323e71b
Reviewed-on: https://cr.bazel.build/8892
PiperOrigin-RevId: 147585207
MOS_MIGRATED_REVID=147585207

2 files changed, 139 insertions, 0 deletions

diff --git a/src/main/native/windows_file_operations.cc b/src/main/native/windows_file_operations.cc
index b5780b2c86..0b02fb9238 100644
--- a/src/main/native/windows_file_operations.cc
+++ b/src/main/native/windows_file_operations.cc
@@ -15,13 +15,17 @@
 #include <windows.h>
 
 #include <memory>
+#include <sstream>
+#include <string>
 
 #include "src/main/native/windows_file_operations.h"
 #include "src/main/native/windows_util.h"
 
 namespace windows_util {
 
+using std::string;
 using std::unique_ptr;
+using std::wstring;
 
 int IsJunctionOrDirectorySymlink(const WCHAR* path) {
   DWORD attrs = ::GetFileAttributesW(path);
@@ -60,4 +64,121 @@ HANDLE OpenDirectory(const WCHAR* path, bool read_write) {
       /* hTemplateFile */ NULL);
 }
 
+#pragma pack(push, 4)
+typedef struct _JunctionDescription {
+  typedef struct _Header {
+    DWORD ReparseTag;
+    WORD ReparseDataLength;
+    WORD Reserved;
+  } Header;
+
+  typedef struct _WriteDesc {
+    WORD SubstituteNameOffset;
+    WORD SubstituteNameLength;
+    WORD PrintNameOffset;
+    WORD PrintNameLength;
+  } WriteDesc;
+
+  Header header;
+  WriteDesc write;
+  WCHAR PathBuffer[ANYSIZE_ARRAY];
+} JunctionDescription;
+#pragma pack(pop)
+
+string CreateJunction(const wstring& junction_name,
+                      const wstring& junction_target) {
+  const wstring target = HasUncPrefix(junction_target.c_str())
+                             ? junction_target.substr(4)
+                             : junction_target;
+  // The entire JunctionDescription cannot be larger than
+  // MAXIMUM_REPARSE_DATA_BUFFER_SIZE bytes.
+  //
+  // The structure's layout is:
+  //   [JunctionDescription::Header]
+  //   [JunctionDescription::WriteDesc]
+  //   ---- start of JunctionDescription::PathBuffer ----
+  //   [4 WCHARs]             : "\??\" prefix
+  //   [target.size() WCHARs] : junction target name
+  //   [1 WCHAR]              : null-terminator
+  //   [target.size() WCHARs] : junction target displayed name
+  //   [1 WCHAR]              : null-terminator
+  // The sum of these must not exceed MAXIMUM_REPARSE_DATA_BUFFER_SIZE.
+  // We can rearrange this to get the limit for target.size().
+  static const size_t kMaxJunctionTargetLen =
+      ((MAXIMUM_REPARSE_DATA_BUFFER_SIZE - sizeof(JunctionDescription::Header) -
+        sizeof(JunctionDescription::WriteDesc) -
+        /* one "\??\" prefix */ sizeof(WCHAR) * 4 -
+        /* two null terminators */ sizeof(WCHAR) * 2) /
+       /* two copies of the string are stored */ 2) /
+      sizeof(WCHAR);
+  if (target.size() > kMaxJunctionTargetLen) {
+    std::stringstream error;
+    error << "junction target is too long (" << target.size()
+          << " characters, limit: " << kMaxJunctionTargetLen << ")";
+    return error.str();
+  }
+  const wstring name = HasUncPrefix(junction_name.c_str())
+                           ? junction_name
+                           : (wstring(L"\\\\?\\") + junction_name);
+
+  // Junctions are directories, so create one
+  if (!::CreateDirectoryW(name.c_str(), NULL)) {
+    return string("CreateDirectoryW failed");
+  }
+
+  AutoHandle handle(OpenDirectory(name.c_str(), true));
+  if (!handle.IsValid()) {
+    return string("OpenDirectory failed");
+  }
+
+  char reparse_buffer_bytes[MAXIMUM_REPARSE_DATA_BUFFER_SIZE];
+  JunctionDescription* reparse_buffer =
+      reinterpret_cast<JunctionDescription*>(reparse_buffer_bytes);
+  memset(reparse_buffer_bytes, 0, MAXIMUM_REPARSE_DATA_BUFFER_SIZE);
+
+  // "\??\" is meaningful to the kernel, it's a synomym for the "\DosDevices\"
+  // object path. (NOT to be confused with "\\?\" which is meaningful for the
+  // Win32 API.) We need to use this prefix to tell the kernel where the reparse
+  // point is pointing to.
+  memcpy(reparse_buffer->PathBuffer, L"\\??\\", 4 * sizeof(WCHAR));
+  memcpy(reparse_buffer->PathBuffer + 4, target.c_str(),
+         target.size() * sizeof(WCHAR));
+
+  // In addition to their target, junctions also have another string which is a
+  // user-visible name of where the junction points, as listed by "dir". This
+  // can be any string and won't affect the usability of the junction.
+  // MKLINK uses the target path without the "\??\" prefix as the display name,
+  // so let's do that here too. This is also in line with how UNIX behaves.
+  // Using a dummy or fake display name would be pure evil, it would make the
+  // output of `dir` look like:
+  //   2017-01-18  01:37 PM    <JUNCTION>     juncname [dummy string]
+  memcpy(reparse_buffer->PathBuffer + 4 + target.size() + 1, target.c_str(),
+         target.size() * sizeof(WCHAR));
+
+  reparse_buffer->write.SubstituteNameOffset = 0;
+  reparse_buffer->write.SubstituteNameLength =
+      (4 + target.size()) * sizeof(WCHAR);
+  reparse_buffer->write.PrintNameOffset =
+      reparse_buffer->write.SubstituteNameLength +
+      /* null-terminator */ sizeof(WCHAR);
+  reparse_buffer->write.PrintNameLength = target.size() * sizeof(WCHAR);
+
+  reparse_buffer->header.ReparseTag = IO_REPARSE_TAG_MOUNT_POINT;
+  reparse_buffer->header.ReparseDataLength =
+      sizeof(JunctionDescription::WriteDesc) +
+      reparse_buffer->write.SubstituteNameLength +
+      reparse_buffer->write.PrintNameLength +
+      /* 2 null-terminators */ (2 * sizeof(WCHAR));
+  reparse_buffer->header.Reserved = 0;
+
+  DWORD bytes_returned;
+  if (!::DeviceIoControl(handle, FSCTL_SET_REPARSE_POINT, reparse_buffer,
+                         reparse_buffer->header.ReparseDataLength +
+                             sizeof(JunctionDescription::Header),
+                         NULL, 0, &bytes_returned, NULL)) {
+    return string("DeviceIoControl(FSCTL_SET_REPARSE_POINT) failed");
+  }
+  return "";
+}
+
 }  // namespace windows_util
diff --git a/src/main/native/windows_file_operations.h b/src/main/native/windows_file_operations.h
index ded07b0a1f..9b2c76837a 100644
--- a/src/main/native/windows_file_operations.h
+++ b/src/main/native/windows_file_operations.h
@@ -17,10 +17,19 @@
 #include <windows.h>
 
 #include <memory>
+#include <string>
 
 namespace windows_util {
 
+using std::string;
 using std::unique_ptr;
+using std::wstring;
+
+template <typename char_type>
+bool HasUncPrefix(const char_type* path) {
+  return path[0] == '\\' && (path[1] == '\\' || path[1] == '?') &&
+         (path[2] == '.' || path[2] == '?') && path[3] == '\\';
+}
 
 // Keep in sync with j.c.g.devtools.build.lib.windows.WindowsFileOperations
 enum {
@@ -61,6 +70,15 @@ bool GetLongPath(const WCHAR* path, unique_ptr<WCHAR[]>* result);
 // otherwise only for reading.
 HANDLE OpenDirectory(const WCHAR* path, bool read_write);
 
+// Creates a junction at `name`, pointing to `target`.
+// Returns the empty string upon success, or a human-readable error message upon
+// failure.
+// Neither `junction_name` nor `junction_target` needs to have a "\\?\" prefix,
+// not even if they are longer than MAX_PATH, though it's okay if they do. This
+// function will add the right prefixes as necessary.
+string CreateJunction(const wstring& junction_name,
+                      const wstring& junction_target);
+
 }  // namespace windows_util
 
 #endif  // BAZEL_SRC_MAIN_NATIVE_WINDOWS_FILE_OPERATIONS_H_