nice to have..

processing/post/sortTable.py

@@ -2,10 +2,9 @@
 
 import os
 import sys
+from io import StringIO
 from optparse import OptionParser
 
-import numpy as np
-
 import damask
 
 
@@ -26,7 +25,7 @@ With coordinates in columns "x", "y", and "z"; sorting with x slowest and z fast
 
 
 parser.add_option('-l','--label',
-                  dest   = 'keys',
+                  dest   = 'labels',
                   action = 'extend', metavar = '<string LIST>',
                   help   = 'list of column labels (a,b,c,...)')
 parser.add_option('-r','--reverse',
@@ -38,42 +37,14 @@ parser.set_defaults(reverse = False,
                    )
 
 (options,filenames) = parser.parse_args()
-
-
-# --- loop over input files -------------------------------------------------------------------------
-
 if filenames == []: filenames = [None]
 
+if options.labels is None:
+    parser.error('no labels specified.')
 for name in filenames:
-  try:    table = damask.ASCIItable(name = name,
-                                    buffered = False)
-  except: continue
     damask.util.report(scriptName,name)
 
-# ------------------------------------------ assemble header ---------------------------------------  
+    table = damask.Table.from_ASCII(StringIO(''.join(sys.stdin.read())) if name is None else name)
+    table.sort_by(options.labels,not options.reverse)
 
-  table.head_read()
-  table.info_append(scriptID + '\t' + ' '.join(sys.argv[1:]))
-  table.head_write()
-
-# ------------------------------------------ process data ---------------------------------------  
-
-  table.data_readArray()
-
-  keys = table.labels(raw = True)[::-1] if options.keys is None else options.keys[::-1]             # numpy sorts with most significant column as last
-
-  cols    = []
-  remarks = []
-  for i,column in enumerate(table.label_index(keys)):
-    if column < 0: remarks.append('label "{}" not present...'.format(keys[i]))
-    else:          cols += [table.data[:,column]]
-  if remarks != []: damask.util.croak(remarks)
-  
-  ind = np.lexsort(cols) if cols != [] else np.arange(table.data.shape[0])
-  if options.reverse: ind = ind[::-1]
-
-# ------------------------------------------ output result ---------------------------------------  
-
-  table.data = table.data[ind]
-  table.data_writeArray()
-  table.close()                                                                                     # close ASCII table
+    table.to_ASCII(sys.stdout if name is None else name)

python/damask/table.py

@@ -1,3 +1,4 @@
+import random
 import re
 
 import pandas as pd
@@ -104,7 +105,7 @@ class Table():
         """
         if re.match(r'[0-9]*?_',label):
             idx,key = label.split('_',1)
-            return self.data[key].to_numpy()[:,int(idx)-1]
+            return self.data[key].to_numpy()[:,int(idx)-1].reshape((-1,1))
         else: 
             return self.data[label].to_numpy().reshape((-1,)+self.shapes[label])
 
@@ -196,6 +197,32 @@ class Table():
         self.shapes[label_new] = self.shapes.pop(label_old)
 
 
+    def sort_by(self,labels,ascending=True):
+        """
+        Get column data.
+
+        Parameters
+        ----------
+        label : list of str or str
+            Column labels.
+        ascending : bool, optional
+            Set sort order.
+
+        """
+        _temp = []
+        _labels = []
+        for label in labels if isinstance(labels,list) else [labels]:
+            if re.match(r'[0-9]*?_',label):
+                _temp.append(str(random.getrandbits(128)))
+                self.add(_temp[-1],self.get(label))
+                _labels.append(_temp[-1])
+            else: 
+                _labels.append(label)
+
+        self.data.sort_values(_labels,axis=0,inplace=True,ascending=ascending)
+        for t in _temp: self.delete(t)
+        self.comments.append('sorted by [{}]'.format(', '.join(labels)))
+
     def to_ASCII(self,fname):
         """
         Store as plain text file.

python/tests/test_Table.py

@@ -19,14 +19,22 @@ def reference_dir(reference_dir_base):
 
 class TestTable:
     
-    def test_get_tensor(self,default):
-        d = default.get('F')
-        assert np.allclose(d,1.0) and d.shape[1:] == (3,3) 
+    def test_get_scalar(self,default):
+        d = default.get('s')
+        assert np.allclose(d,1.0) and d.shape[1:] == (1,)
 
     def test_get_vector(self,default):
         d = default.get('v')
         assert np.allclose(d,1.0) and d.shape[1:] == (3,)
 
+    def test_get_tensor(self,default):
+        d = default.get('F')
+        assert np.allclose(d,1.0) and d.shape[1:] == (3,3) 
+
+    def test_get_component(self,default):
+        d = default.get('5_F')
+        assert np.allclose(d,1.0) and d.shape[1:] == (1,)
+  
     def test_write_read_str(self,default,tmpdir):
         default.to_ASCII(str(tmpdir.join('default.txt')))
         new = Table.from_ASCII(str(tmpdir.join('default.txt')))
@@ -87,3 +95,27 @@ class TestTable:
     def test_invalid_get(self,default):
         with pytest.raises(KeyError):
             default.get('n')
+
+    def test_sort_scalar(self):
+        x = np.random.random((5,13))
+        t = Table(x,{'F':(3,3),'v':(3,),'s':(1,)},['random test data'])
+        unsort = t.get('s')
+        t.sort_by('s')
+        sort   = t.get('s')
+        assert np.all(np.sort(unsort,0)==sort)
+
+    def test_sort_component(self):
+        x = np.random.random((5,12))
+        t = Table(x,{'F':(3,3),'v':(3,)},['random test data'])
+        unsort = t.get('4_F')
+        t.sort_by('4_F')
+        sort = t.get('4_F')
+        assert np.all(np.sort(unsort,0)==sort)
+
+    def test_sort_revert(self):
+        x = np.random.random((5,12))
+        t = Table(x,{'F':(3,3),'v':(3,)},['random test data'])
+        t.sort_by('4_F',False)
+        sort = t.get('4_F')
+        assert np.all(np.sort(sort,0)==sort[::-1,:])
+