New pca workflow and datset updates

system/dataset.py

@@ -12,12 +12,16 @@ class Dataset:
     """
     def __init__(self,input_array,def_list):
         self._data = asarray(input_array)
-        self.dims = shape(self._data)
+        dims = shape(self._data)
         self.def_list = def_list
         self._ids_set = set()
         self.ids={}
         self._dim_num = {}
         self._dim_names = []
+        if len(dims)==1: # a vector is defined to be column vector!
+            self.dims = (dims[0],1)
+        else:
+            self.dims = dims
         if len(def_list)!=len(self.dims):
             raise ValueError,"array dims and identifyer mismatch"
         for axis,(dim_name,ids) in enumerate(def_list):
@@ -25,7 +29,7 @@ class Dataset:
             #if dim_name not in project.c_p.dim_names:
             #    dim_name = project.c_p.suggest_dim_name(dim_name)
             if not ids:
-                logger.log('debug','Creating identifiers along: '+dim_name)
+                logger.log('debug','Creating identifiers along: '+ str(dim_name))
                 ids = self._create_identifiers(axis)
             for num,name in enumerate(ids):
                 enum_ids[name] = num
@@ -40,13 +44,21 @@ class Dataset:
                 raise ValueError,"dim size and identifyer mismatch"
             
     def names(self,axis=0):
-        """Returns identifier names of a dimension. NB: not in any order! """
+        """Returns identifier names of a dimension.
+        NB: sorted by values!
+        OK? necessary?"""
         
         if type(axis)==int:
             dim_name = self._dim_names[axis]
         elif type(axis)==str:
             dim_name = axis
-        return self.ids[dim_name].keys()
+        if dim_name not in self._dim_names:
+            raise ValueError, dim_name + " not a dimension in dataset" 
+        items = self.ids[dim_name].items()
+        backitems=[ [v[1],v[0]] for v in items]
+        backitems.sort()
+        sorted_ids=[ backitems[i][1] for i in range(0,len(backitems))]
+        return sorted_ids
     
     def extract_data(self,ids,dim_name):
         """Extracts data along a dimension by identifiers"""