gstlal.lloidparts: print state vector transitions to stderr

gstlal/python/lloidparts.py

@@ -208,7 +208,8 @@ def mkcontrolsnksrc(pipeline, rate, verbose = False, suffix = None, inj_seg_list
 class get_state_vector(object):
 	# monitor state vector transitions, export via web
 	# interface
-	def __init__(self, elem):
+	def __init__(self, elem, verbose = False):
+		self.verbose = verbose
 		self.current_segment = "unknown"
 		self.segment_start = "unknown"
 		elem.connect("start", self.sighandler, "science")
@@ -217,6 +218,8 @@ class get_state_vector(object):
 	def sighandler(self, elem, timestamp, segment_type):
 		self.current_segment = segment_type
 		self.segment_start = "%.9f" % (timestamp / 1e9)
+		if self.verbose:
+			print >>sys.stderr, "state transition: %s" % self.text()
 
 	def text(self):
 		return "%s @ %s\n" % (self.current_segment, self.segment_start)
@@ -320,7 +323,9 @@ def mkLLOIDbasicsrc(pipeline, seekevent, instrument, detector, fake_data = None,
 		src = pipeparts.mkgate(pipeline, strain, threshold = 1, control = statevector)
 		# export state vector state
 		src.set_property("emit-signals", True)
-		bottle.route("/%s/current_segment.txt" % instrument)(get_state_vector(src).text)
+		# FIXME:  let the state vector messages going to stderr be
+		# controled somehow
+		bottle.route("/%s/current_segment.txt" % instrument)(get_state_vector(src, verbose = True).text)
 	else:
 		src = pipeparts.mkaudioconvert(pipeline, src)