Only remember log lines we need to print later.

When testing proccessing a large logfile (430 MB, 2M lines) without
--print-all-missed or --print-all-ignored, fail2ban-regex consumes
just over 2 GB of RAM before completing.  Even though it does not need
to retain any of the missed or ignored log lines, it does anyway.

With this patch, memory use never grows beyond about 11 MB (unless
--print-all-* are enabled).

fail2ban-regex

@@ -156,22 +156,16 @@ class LineStats(object):
 	"""
 	def __init__(self):
 		self.tested = self.matched = 0
+		self.missed = 0
 		self.missed_lines = []
 		self.missed_lines_timeextracted = []
+		self.ignored = 0
 		self.ignored_lines = []
 		self.ignored_lines_timeextracted = []
 
 	def __str__(self):
 		return "%(tested)d lines, %(ignored)d ignored, %(matched)d matched, %(missed)d missed" % self
 
-	@property
-	def ignored(self):
-		return len(self.ignored_lines)
-
-	@property
-	def missed(self):
-		return self.tested - (self.ignored + self.matched)
-
 	# just for convenient str
 	def __getitem__(self, key):
 		return getattr(self, key)
@@ -267,6 +261,8 @@ class Fail2banRegex(object):
 			line_datetimestripped, ret = fail2banRegex.testRegex(line)
 
 			if is_ignored:
+				self._line_stats.ignored += 1
+				if self._print_all_ignored:
 					self._line_stats.ignored_lines.append(line)
 					self._line_stats.ignored_lines_timeextracted.append(line_datetimestripped)
 
@@ -275,6 +271,8 @@ class Fail2banRegex(object):
 				self._line_stats.matched += 1
 			else:
 				if not is_ignored:
+					self._line_stats.missed += 1
+					if self._print_all_missed:
 						self._line_stats.missed_lines.append(line)
 						self._line_stats.missed_lines_timeextracted.append(line_datetimestripped)
 			self._line_stats.tested += 1