[HWASan] Clean up hwasan_symbolize.

The globals are better expressed as members of the Symbolizer, and all
functions operating on it should be methods instead.

Also using the standard idiom of wrapping the main code in
`if __name__ == '__main__'`.

Reviewed By: eugenis

Differential Revision: https://reviews.llvm.org/D125032

compiler-rt/lib/hwasan/scripts/hwasan_symbolize

@@ -31,9 +31,6 @@ if sys.version_info.major < 3:
   import codecs
   sys.stdout = codecs.getwriter("utf-8")(sys.stdout)
 
-last_access_address = None
-last_access_tag = None
-
 # Below, a parser for a subset of ELF. It only supports 64 bit, little-endian,
 # and only parses what is necessary to find the build ids. It uses a memoryview
 # into an mmap to avoid copying.
@@ -110,6 +107,8 @@ class Symbolizer:
     self.__index = {}
     self.__link_prefixes = []
     self.__html = False
+    self.__last_access_address = None
+    self.__last_access_tag = None
 
   def enable_html(self, enable):
     self.__html = enable
@@ -268,7 +267,7 @@ class Symbolizer:
           if bid is not None:
             self.__index[bid] = filename
 
-def symbolize_line(line, symbolizer_path):
+  def symbolize_line(self, line):
     #0 0x7f6e35cf2e45  (/blah/foo.so+0x11fe45) (BuildId: 4abce4cd41ea5c2f34753297b7e774d9)
     match = re.match(r'^(.*?)#([0-9]+)( *)(0x[0-9a-f]*) *\((.*)\+(0x[0-9a-f]+)\)'
                     r'(?:\s*\(BuildId: ([0-9a-f]+)\))?', line, re.UNICODE)
@@ -278,37 +277,36 @@ def symbolize_line(line, symbolizer_path):
       addr = int(match.group(6), 16)
       buildid = match.group(7)
 
-    frames = list(symbolizer.iter_call_stack(binary, buildid, addr))
+      frames = list(self.iter_call_stack(binary, buildid, addr))
 
       if len(frames) > 0:
-      symbolizer.print(
-        symbolizer.maybe_escape(
+        self.print(
+          self.maybe_escape(
             "%s#%s%s%s in " % (match.group(1), match.group(2), match.group(3),
                               frames[0][0])
-        ) + symbolizer.maybe_linkify(frames[0][1]),
+          ) + self.maybe_linkify(frames[0][1]),
           escape=False)
         for i in range(1, len(frames)):
           space1 = ' ' * match.end(1)
           space2 = ' ' * (match.start(4) - match.end(1) - 2)
-        symbolizer.print(
-          symbolizer.maybe_escape("%s->%s%s in " % (space1, space2, frames[i][0]))
-            + symbolizer.maybe_linkify(frames[i][1]), escape=False)
+          self.print(
+            self.maybe_escape("%s->%s%s in " % (space1, space2, frames[i][0]))
+              + self.maybe_linkify(frames[i][1]), escape=False)
       else:
-      symbolizer.print(line.rstrip())
+        self.print(line.rstrip())
     else:
-    symbolizer.print(line.rstrip())
+      self.print(line.rstrip())
 
-def save_access_address(line):
-  global last_access_address, last_access_tag
+  def save_access_address(self, line):
     match = re.match(r'^(.*?)HWAddressSanitizer: tag-mismatch on address (0x[0-9a-f]+) ', line, re.UNICODE)
     if match:
-    last_access_address = int(match.group(2), 16)
+      self.__last_access_address = int(match.group(2), 16)
     match = re.match(r'^(.*?) of size [0-9]+ at 0x[0-9a-f]* tags: ([0-9a-f]+)/[0-9a-f]+ \(ptr/mem\)', line, re.UNICODE)
     if match:
-    last_access_tag = int(match.group(2), 16)
+      self.__last_access_tag = int(match.group(2), 16)
 
-def process_stack_history(line, symbolizer, ignore_tags=False):
-  if last_access_address is None or last_access_tag is None:
+  def process_stack_history(self, line, ignore_tags=False):
+    if self.__last_access_address is None or self.__last_access_tag is None:
       return
     if re.match(r'Previously allocated frames:', line, re.UNICODE):
       return True
@@ -327,24 +325,32 @@ def process_stack_history(line, symbolizer, ignore_tags=False):
       fp = (record >> 48) << 4
       pc = record & pc_mask
 
-    for local in symbolizer.iter_locals(binary, addr, buildid):
+      for local in self.iter_locals(binary, addr, buildid):
         frame_offset = local[3]
         size = local[4]
         if frame_offset is None or size is None:
           continue
-      obj_offset = (last_access_address - fp - frame_offset) & fp_mask
+        obj_offset = (self.__last_access_address - fp - frame_offset) & fp_mask
         if obj_offset >= size:
           continue
         tag_offset = local[5]
-      if not ignore_tags and (tag_offset is None or base_tag ^ tag_offset != last_access_tag):
+        if not ignore_tags and (tag_offset is None or base_tag ^ tag_offset != self.__last_access_tag):
           continue
-      symbolizer.print('')
-      symbolizer.print('Potentially referenced stack object:')
-      symbolizer.print('  %d bytes inside variable "%s" in stack frame of function "%s"' % (obj_offset, local[2], local[0]))
-      symbolizer.print('  at %s' % (local[1],))
+        self.print('')
+        self.print('Potentially referenced stack object:')
+        self.print('  %d bytes inside variable "%s" in stack frame of function "%s"' % (obj_offset, local[2], local[0]))
+        self.print('  at %s' % (local[1],))
       return True
     return False
 
+def extract_version(s):
+  idx = s.rfind('-')
+  if idx == -1:
+    return 0
+  x = float(s[idx + 1:])
+  return x
+
+def main():
   parser = argparse.ArgumentParser()
   parser.add_argument('-d', action='store_true')
   parser.add_argument('-v', action='store_true')
@@ -410,13 +416,6 @@ if not symbolizer_path:
         symbolizer_path = p
         break
 
-def extract_version(s):
-  idx = s.rfind('-')
-  if idx == -1:
-    return 0
-  x = float(s[idx + 1:])
-  return x
-
   if not symbolizer_path:
     for path in os.environ["PATH"].split(os.pathsep):
       candidates = glob.glob(os.path.join(path, 'llvm-symbolizer-*'))
@@ -454,7 +453,11 @@ if args.linkify:
   for line in sys.stdin:
     if sys.version_info.major < 3:
       line = line.decode('utf-8')
-  save_access_address(line)
-  if process_stack_history(line, symbolizer, ignore_tags=args.ignore_tags):
+    symbolizer.save_access_address(line)
+    if symbolizer.process_stack_history(line, ignore_tags=args.ignore_tags):
       continue
-  symbolize_line(line, symbolizer_path)
+    symbolizer.symbolize_line(line)
+
+
+if __name__ == '__main__':
+  main()