Use the correct short_name for swap

dool

@@ -1386,7 +1386,8 @@ class dool_swap(dool):
         else:
             ret = []
             for dev in self.vars:
-                short_name = dev_short_name(dev)
+                dev        = basename(dev)
+                short_name = dev_short_name(dev, 4)
                 ret.append('swap/' + short_name)
 
         return ret

plugins/dool__pid_detail.py

@@ -0,0 +1,178 @@
+### Author: Scott Baker - <https://www.perturb.org/>
+
+class dool_plugin(dool):
+	"""
+	Plugin to gather data about a specific PID
+	"""
+
+	def __init__(self):
+		self.vars  = ('cpu  read  writ  mem',)
+		self.type  = 's'
+		self.width = 20
+		self.scale = 0
+		self.prev  = { 'read': -1, 'write': -1, 'cpu': 0, 'mem': 0}
+
+		# Get the pid from the CLI options
+		try:
+			pid  = int(op.plugin_params['pid-detail'])
+		except:
+			msg = 'pid-detail: %s is not a number' % (op.plugin_params['pid-detail'])
+			raise Exception(msg)
+
+		path = '/proc/%s/stat' % pid
+		if not os.path.exists(path):
+			msg = 'pid-detail: %d is not an active pid' % (pid)
+			raise Exception(msg)
+
+		name = getnamebypid(pid, 'unknown')
+
+		# Set the plugin title at the top of the column
+		self.name = "pid info: %s" % name
+
+		# Save the pid name and number for easier access later
+		self.pid_name = name
+		self.pid      = pid
+
+	def extract(self):
+		pid = self.pid
+
+		# Make sure we have data for this pid
+		path = '/proc/%s/stat' % pid
+		if not os.path.exists(path):
+			self.output = 'Pid #%s not found' % pid
+			return
+
+		(disk_rd, disk_wr) = self.get_pid_disk(pid)
+		cpu_usage = self.get_pid_cpu(pid)
+		mem_usage = self.get_pid_mem(pid)
+
+		cpu_str   = cprint(cpu_usage, 'f', 3, 34)
+		disk_rd   = cprint(disk_rd	, 'd', 5, 1024)
+		disk_wr   = cprint(disk_wr	, 'd', 5, 1024)
+		mem_usage = cprint(mem_usage, 'd', 4, 1024)
+
+		self.output = "%s %s %s %s" % (cpu_str, disk_rd, disk_wr, mem_usage)
+
+	# Sometimes /proc/12345/stat has a space in the proc name field
+	# which messes up the split. This tries to be smart and work around that
+	def get_proc_pid_stat(self, pid):
+		path = '/proc/%s/stat' % pid
+		if not os.path.exists(path):
+			return []
+
+		# Read the first line
+		line = proc_readline(path)
+		line = line.strip()
+
+		# Find the proc name that's between the ( )
+		x     = re.search("\((.*?)\)", line)
+		paren = x.group(1)
+
+		# Replace any spaces in the parens part with underscores
+		cleaned  = paren.replace(" ", "_")
+		new_line = line.replace(paren, cleaned, 1)
+
+		# Return the split array
+		parts = new_line.split(" ")
+
+		return parts
+
+	def get_pid_cpu(self, pid):
+		path = '/proc/%s/stat' % pid
+		if not os.path.exists(path):
+			self.output = 'Pid #%s not found' % pid
+
+		l = self.get_proc_pid_stat(pid)
+		if len(l) < 35: return 0
+
+		user_time   = int(l[13])
+		kernel_time = int(l[14])
+
+		current_cpu = user_time + kernel_time
+		prev_cpu    = self.prev['cpu']
+		usage       = (current_cpu - prev_cpu) * 1.0 / elapsed / cpunr
+
+		if prev_cpu:
+			ret = usage
+		else:
+			ret = 0
+
+		# If we're at a real update (not interim) we save the data
+		if step == op.delay:
+			self.prev['cpu'] = current_cpu
+
+		return ret
+
+	def get_pid_disk(self, pid):
+		path  = '/proc/%s/io' % pid
+		count = 0
+		for p in proc_splitlines(path):
+			key = p[0]
+			val = p[1]
+
+			if key == "read_bytes:":
+				cur_read = int(val)
+			elif key == "write_bytes:":
+				cur_write = int(val)
+
+			count += 1
+
+		# This is usually a permission error reading that part of /proc/
+		if count == 0:
+			return (0,0)
+
+		if (op.bits):
+			factor = 8
+		else:
+			factor = 1
+
+		# We have previous data so we calculate the difference
+		# between the two
+		if self.prev['read'] > -1:
+			rread	= (cur_read - self.prev['read'])   * factor / elapsed
+			rwrite	= (cur_write - self.prev['write']) * factor / elapsed
+		# First round we just gather data
+		else:
+			rread  = 0
+			rwrite = 0
+
+		# If we're at a real update (not interim) we save the data
+		if step == op.delay:
+			self.prev['read']  = cur_read
+			self.prev['write'] = cur_write
+
+		ret = (rread, rwrite)
+
+		return ret
+
+	def get_pid_mem(self, pid):
+		l = proc_splitline('/proc/%s/statm' % pid)
+
+		if len(l) < 2: return 0
+
+		prev_usage = self.prev['mem']
+		cur_usage  = (int(l[1]) * pagesize)
+
+		# If we have previous numbers we return the average
+		if prev_usage:
+			usage = self.prev['mem'] / elapsed
+		# First run just gets the current number
+		else:
+			usage = cur_usage
+
+		self.prev['mem'] += cur_usage
+
+		if step == op.delay:
+			self.prev['mem'] = cur_usage
+
+		# Uncomment this to always return the absolute (not average)
+		#usage = cur_usage
+
+		return usage
+
+	def showcsv(self):
+		xstr = "%s %s %s %s" % (1,2,3,4)
+
+		return xstr
+
+# vim: tabstop=4 shiftwidth=4 noexpandtab autoindent softtabstop=4