Add Dr. Memory tool

This CL adds DrMemory-Windows-sfx.exe into tools/drmemory.
It also adds a set of scripts from chromium/src/tools/valgrind for
running Dr. Memory.  Now we can run pdfium tests with Dr. Memory:
$ .\tools\drmemory\scripts\pdfium_tests.bat -b out\Debug -t pdfium_unittests

More update on scripts is required, but this should be a good start.

Changes from chromium/src/tools/valgrind
- rename chrome_tests.py to pdfium_tests.py
- update pdfium_tests.bat
- remove chrome tests in pdfium_tests.py
- remove memcheck code in valgrind_test.py

R=thestig@chromium.org
BUG=pdfium:238

Review URL: https://codereview.chromium.org/1452293002 .

1 files changed, 82 insertions, 0 deletions

diff --git a/tools/drmemory/scripts/logging_utils.py b/tools/drmemory/scripts/logging_utils.py
new file mode 100644
index 0000000000..ef2d674950
--- /dev/null
+++ b/tools/drmemory/scripts/logging_utils.py
@@ -0,0 +1,82 @@
+# Copyright (c) 2011 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+''' Utility functions and objects for logging.
+'''
+
+import logging
+import sys
+
+class StdoutStderrHandler(logging.Handler):
+  ''' Subclass of logging.Handler which outputs to either stdout or stderr
+  based on a threshold level.
+  '''
+
+  def __init__(self, threshold=logging.WARNING, err=sys.stderr, out=sys.stdout):
+    ''' Args:
+          threshold: below this logging level messages are sent to stdout,
+            otherwise they are sent to stderr
+          err: a stream object that error messages are sent to, defaults to
+            sys.stderr
+          out: a stream object that non-error messages are sent to, defaults to
+            sys.stdout
+    '''
+    logging.Handler.__init__(self)
+    self._err = logging.StreamHandler(err)
+    self._out = logging.StreamHandler(out)
+    self._threshold = threshold
+    self._last_was_err = False
+
+  def setLevel(self, lvl):
+    logging.Handler.setLevel(self, lvl)
+    self._err.setLevel(lvl)
+    self._out.setLevel(lvl)
+
+  def setFormatter(self, formatter):
+    logging.Handler.setFormatter(self, formatter)
+    self._err.setFormatter(formatter)
+    self._out.setFormatter(formatter)
+
+  def emit(self, record):
+    if record.levelno < self._threshold:
+      self._out.emit(record)
+      self._last_was_err = False
+    else:
+      self._err.emit(record)
+      self._last_was_err = False
+
+  def flush(self):
+    # preserve order on the flushing, the stalest stream gets flushed first
+    if self._last_was_err:
+      self._out.flush()
+      self._err.flush()
+    else:
+      self._err.flush()
+      self._out.flush()
+
+
+FORMAT = "%(asctime)s %(filename)s [%(levelname)s] %(message)s"
+DATEFMT = "%H:%M:%S"
+
+def config_root(level=logging.INFO, threshold=logging.WARNING, format=FORMAT,
+         datefmt=DATEFMT):
+  ''' Configure the root logger to use a StdoutStderrHandler and some default
+  formatting.
+    Args:
+      level: messages below this level are ignored
+      threshold: below this logging level messages are sent to stdout,
+        otherwise they are sent to stderr
+      format: format for log messages, see logger.Format
+      datefmt: format for date in log messages
+
+  '''
+  # to set the handler of the root logging object, we need to do setup
+  # manually rather than using basicConfig
+  root = logging.getLogger()
+  root.setLevel(level)
+  formatter = logging.Formatter(format, datefmt)
+  handler = StdoutStderrHandler(threshold=threshold)
+  handler.setLevel(level)
+  handler.setFormatter(formatter)
+  root.addHandler(handler)