Fixed jobs parsing. pbsstatus now prints status

README.rst

@@ -9,3 +9,13 @@ pypbs
 
 .. image:: https://coveralls.io/repos/VDBWRAIR/pypbs/badge.svg
     :target: https://coveralls.io/r/VDBWRAIR/pypbs
+
+Small api for pbs stuff
+
+Some useful commands too
+
+.. code-block:: bash
+
+    pbsstatus 
+    NP Utilization  Cluster Load    Avail CPU   Used CPU    Total CPU   Running Jobs
+            90.62%        43.03%            3         29           32             17

pypbs/nodesxml.py

@@ -12,6 +12,10 @@ def parse_xml(nodes_xml):
         tdict = {}
         for attrib in node:
             tdict[attrib.tag] = parse_list_string(attrib.text)
+        if 'jobs' not in tdict:
+            tdict['jobs'] = []
+        elif isinstance(tdict['jobs'],str):
+            tdict['jobs'] = [tdict['jobs']]
         nodes[tdict['name']] = tdict
     return nodes
 
@@ -42,11 +46,11 @@ def parse_list_string(string):
 
     # Now do another search for subdict patterns
     subdict = {}
-    p = '(\w+)=((?:\w\(\S+\)\s{0,1})+)'
+    p = '(\w+)=([\w\d\.]+\(.*?\)(?: \S+\(.*?\)){0,})'
     stuff = re.findall(p, string)
     for key, values in stuff:
         subdict[key] = {}
-        defs = re.findall('(?:(\w)\((.*?)\))', values)
+        defs = re.findall('(?:(\S+)\((.*?)\))', values)
         for _key, d in defs:
             subdict[key][_key] = parse_list_string(d)
     _dict.update(subdict)

pypbs/pbsstatus.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import print_function
+
 import xml.etree.ElementTree as ET
 import re
 
@@ -103,8 +105,22 @@ def parse_job_string(job_str):
 
     return job
 
+def cluster_status(cluster_info):
+    '''
+    Return a formatted string that should display cluster status nicely
+
+    :param dict cluster_info: dict of cluster information
+    '''
+    template = 'NP Utilization\tCluster Load\tAvail CPU\tUsed CPU\tTotal CPU\tRunning Jobs\n'
+    template += '{np_utilization:14.2%}\t{load_utilization:12.2%}\t{avail_np:9}\t{used_np:8}\t{total_np:9}\t{running_jobs:12}'
+    return template.format(
+        **cluster_info
+    )
+
 def main():
     xml = get_pbsnodes_xml()
     nodes = nodesxml.parse_xml(xml)
-    import pprint
-    pprint.pprint(nodes)
+    #import pprint
+    #pprint.pprint(nodes)
+    cinfo = cluster_info(nodes)
+    print(cluster_status(cinfo))

tests/output.xml

@@ -49,4 +49,29 @@
   <mom_manager_port>15003</mom_manager_port>
   <gpus>1</gpus>
  </Node>
+ <Node>
+  <name>host5.example.com</name>
+  <state>free</state>
+  <power_state>Running</power_state>
+  <np>12</np>
+  <properties>bigmem,tesla</properties>
+  <ntype>cluster</ntype>
+  <status>rectime=1430334888,macaddr=00:00:00:00:00:03,cpuclock=OnDemand:2301MHz,varattr=,jobs=,state=free,netload=622455994750,gres=,loadave=17.77,ncpus=24,physmem=264455068kb,availmem=260210680kb,totmem=272647060kb,idletime=102260,nusers=1,nsessions=11,sessions=25749 5514 2372 1830 1494 1421 1350 18106 17897 31317 2600,uname=Linux host5.example.com 2.6.32-431.20.3.el6.x86_64 #1 SMP Fri Jun 6 18:30:54 EDT 2014 x86_64,opsys=linux</status>
+  <mom_service_port>15002</mom_service_port>
+  <mom_manager_port>15003</mom_manager_port>
+  <gpus>2</gpus>
+ </Node>
+ <Node>
+  <name>host6.example.com</name>
+  <state>free</state>
+  <power_state>Running</power_state>
+  <np>12</np>
+  <properties>bigmem,tesla</properties>
+  <ntype>cluster</ntype>
+  <jobs>0/1.host1.example.com</jobs>
+  <status>rectime=1430334888,macaddr=00:00:00:00:00:03,cpuclock=OnDemand:2301MHz,varattr=,jobs=,state=free,netload=622455994750,gres=,loadave=17.77,ncpus=24,physmem=264455068kb,availmem=260210680kb,totmem=272647060kb,idletime=102260,nusers=1,nsessions=11,sessions=25749 5514 2372 1830 1494 1421 1350 18106 17897 31317 2600,uname=Linux host6.example.com 2.6.32-431.20.3.el6.x86_64 #1 SMP Fri Jun 6 18:30:54 EDT 2014 x86_64,opsys=linux</status>
+  <mom_service_port>15002</mom_service_port>
+  <mom_manager_port>15003</mom_manager_port>
+  <gpus>2</gpus>
+ </Node>
 </Data>

tests/test_nodesxml.py

@@ -21,6 +21,23 @@ def test_parses_comma_lists(self):
             self.nodes['host4.example.com']['jobs'], list
         )
 
+    def test_parses_status_jobs(self):
+        hostinfo = self.nodes['host3.example.com']
+        jobs = hostinfo['status']['jobs']
+        self.assertEqual(
+            '2600', jobs['656.host1.example.com']['session_id']
+        )
+
+    def test_when_only_one_job_still_returns_list(self):
+        self.assertEqual(
+            ['0/1.host1.example.com'], self.nodes['host6.example.com']['jobs']
+        )
+
+    def test_when_no_jobs_sets_empty_list(self):
+        self.assertEqual(
+            [], self.nodes['host5.example.com']['jobs']
+        )
+
     def test_builds_sub_dict_with_equal_signs(self):
         self.assertEqual(
             '1430334904', self.nodes['host1.example.com']['status']['rectime']
@@ -44,8 +61,11 @@ def test_returns_dictionary(self):
             nodesxml.parse_list_string('foo=bar,baz=,bar=foo')
         )
 
-    def test_parses_sub_dictionary_in_paren(self):
-        self.assertEqual(
-            {'foo':'bar','baz':'foo','jobs':{'1':{'foo':'1','bar':'2'},'2':{'foo':'1','bar':'2'}}},
-            nodesxml.parse_list_string('foo=bar,jobs=1(foo=1,bar=2) 2(foo=1,bar=2),baz=foo')
-        )
+    def test_parses_actual(self):
+        status = 'rectime=1430334888,macaddr=00:00:00:00:00:03,cpuclock=OnDemand:2301MHz,varattr=,jobs=656.host1.example.com(cput=704205,energy_used=0,mem=663876kb,vmem=290383028kb,walltime=1143760,session_id=2600) 812.host1.example.com(cput=466890,energy_used=0,mem=851568kb,vmem=286089568kb,walltime=513972,session_id=31317) 851.host1.example.com(cput=171210,energy_used=0,mem=856996kb,vmem=290125588kb,walltime=171501,session_id=17897) 855.host1.example.com(cput=169197,energy_used=0,mem=530380kb,vmem=290383028kb,walltime=169460,session_id=18106) 864.host1.example.com(cput=107891,energy_used=0,mem=941696kb,vmem=290326052kb,walltime=108039,session_id=1350) 865.host1.example.com(cput=107606,energy_used=0,mem=832564kb,vmem=290193180kb,walltime=107777,session_id=1421) 866.host1.example.com(cput=107372,energy_used=0,mem=916400kb,vmem=290324252kb,walltime=107567,session_id=1494) 868.host1.example.com(cput=106830,energy_used=0,mem=830340kb,vmem=290060052kb,walltime=107005,session_id=1830) 870.host1.example.com(cput=106174,energy_used=0,mem=843748kb,vmem=290125588kb,walltime=106340,session_id=2372) 892.host1.example.com(cput=95790,energy_used=0,mem=836868kb,vmem=286180260kb,walltime=95935,session_id=5514) 978.host1.example.com(cput=7449,energy_used=0,mem=846060kb,vmem=290060052kb,walltime=7463,session_id=25749),state=free,netload=622455994750,gres=,loadave=17.77,ncpus=24,physmem=264455068kb,availmem=260210680kb,totmem=272647060kb,idletime=102260,nusers=1,nsessions=11,sessions=25749 5514 2372 1830 1494 1421 1350 18106 17897 31317 2600,uname=Linux host3.example.com 2.6.32-431.20.3.el6.x86_64 #1 SMP Fri Jun 6 18:30:54 EDT 2014 x86_64,opsys=linux'
+        r = nodesxml.parse_list_string(status)
+        self.assertEqual('1430334888', r['rectime'])
+        self.assertIn('jobs', r)
+        self.assertIn('656.host1.example.com', r['jobs'])
+        self.assertEqual('704205', r['jobs']['656.host1.example.com']['cput'])
+        self.assertEqual(11, len(r['jobs']))