Added option to output parsed data as a python-parseable map.

4 files changed, 62 insertions, 18 deletions

diff --git a/parse/point.py b/parse/point.py
index 8fdd115..d5f4a5e 100644
--- a/parse/point.py
+++ b/parse/point.py
@@ -37,12 +37,13 @@ class Measurement(object):
         self[Type.Max] = array.max()
         self[Type.Avg] = array.mean()
         self[Type.Var] = array.var()
+        self[Type.Min] = array.min()
         return self
 
     def __check_type(self, type):
         if not type in Type:
             raise AttributeError("Not a valid type '%s'" % type)
-        
+
     def __getitem__(self, type):
         self.__check_type(type)
         return self.stats[type]
@@ -60,11 +61,11 @@ class Measurement(object):
 
     def __str__(self):
         return "%s" % dict_str(self.stats, " ")
-            
+
 
 class Summary(Measurement):
     def __init__(self, id, measures, typemap = default_typemap):
-        super(Summary, self).__init__("Summary-%s" % id)
+        super(Summary, self).__init__(id)
 
         self.__check_types(measures, typemap)
         self.__summarize(measures, typemap)
@@ -108,7 +109,7 @@ class ExpPoint(object):
     def __check_val(self, obj):
         if not isinstance(obj, Measurement):
             raise AttributeError("Not a valid measurement '%s'" % obj)
-        
+
     def __getitem__(self, type):
         return self.stats[type]
 
diff --git a/parse/sched.py b/parse/sched.py
index cbb051e..80764b6 100644
--- a/parse/sched.py
+++ b/parse/sched.py
@@ -1,6 +1,9 @@
 """
 TODO: No longer very pythonic, lot of duplicate code
-print out task execution times
+print out task execution times or something
+get miss ratio and tardiness directly from schedule OR
+email list about turning on optional summary statistics OR
+set up run exps to only get release and completions to get these stats
 """
 
 import config.config as conf
diff --git a/parse/tuple_table.py b/parse/tuple_table.py
index 434eb22..465abb3 100644
--- a/parse/tuple_table.py
+++ b/parse/tuple_table.py
@@ -1,7 +1,9 @@
 from collections import defaultdict
-from point import SummaryPoint
+from point import SummaryPoint,Type
 from dir_map import DirMap
 
+from pprint import pprint
+
 class ColMap(object):
     def __init__(self):
         self.rev_map = {}
@@ -31,13 +33,16 @@ class ColMap(object):
             map[self.col_list[i]] = tuple[i]
         return map
 
+    def force_add(self, column):
+        self.rev_map[column] = len(self.col_list)
+        self.col_list += [column]
+
     def try_add(self, column, value):
         if column not in self.rev_map:
             if column not in self.value_map:
                 self.value_map[column] = value
             elif value != self.value_map[column]:
-                self.rev_map[column] = len(self.col_list)
-                self.col_list += [column]
+                self.force_add(column)
 
     def __str__(self):
         return "<ColMap>%s" % (self.rev_map)
@@ -53,6 +58,9 @@ class TupleTable(object):
         key = self.col_map.get_key(kv)
         self.table[key] += [point]
 
+    def col_map(self):
+        return self.col_map
+
     def get_exps(self, kv):
         key = self.col_map.get_key(kv)
         return self.table[key]
@@ -66,9 +74,32 @@ class TupleTable(object):
             raise Exception("cannot reduce twice!")
         self.reduced = True
         for key, values in self.table.iteritems():
-            self.table[key] = SummaryPoint(str(key), values)
+            self.table[key] = SummaryPoint(values[0].id, values)
 
-    def write_result(self, out_dir):
+    def write_map(self, out_map):
+        if not self.reduced:
+            raise Exception("must reduce table to write map!")
+
+        rows = {}
+
+        for key, point in self.table.iteritems():
+            row = {}
+            for name,measurement in point:
+                name = name.lower().replace('_','-')
+                row[name]={}
+                for base_type in Type:
+                    type_key = str(base_type).lower()
+                    if base_type in measurement[Type.Avg]:
+                        value = measurement[Type.Avg][base_type]
+                        row[name][type_key] = value
+            rows[key] = row
+
+        result = {'columns': self.col_map.columns(), 'rows':rows}
+
+        with open(out_map, 'wc') as map_file:
+            pprint(result,stream=map_file, width=20)
+
+    def write_csvs(self, out_dir):
         dir_map = DirMap(out_dir)
 
         for key, point in self.table.iteritems():
diff --git a/parse_exps.py b/parse_exps.py
index aa203d3..2d1c370 100755
--- a/parse_exps.py
+++ b/parse_exps.py
@@ -16,10 +16,12 @@ from parse.point import ExpPoint
 from parse.tuple_table import ColMap,TupleTable
 
 def parse_args():
+    # TODO: convert data-dir to proper option
     parser = OptionParser("usage: %prog [options] [data_dir]...")
 
-    parser.add_option('-o', '--out-dir', dest='out_dir',
-                      help='directory for data output', default='parse-data')
+    parser.add_option('-o', '--out', dest='out',
+                      help='file or directory for data output', default='parse-data')
+
     # TODO: this means nothing
     parser.add_option('-c', '--clean', action='store_true', default=False,
                       dest='clean', help='do not output single-point csvs')
@@ -30,6 +32,9 @@ def parse_args():
                       dest='force', help='overwrite existing data')
     parser.add_option('-v', '--verbose', action='store_true', default=False,
                       dest='verbose', help='print out data points')
+    parser.add_option('-m', '--write-map', action='store_true', default=False,
+                      dest='write_map',
+                      help='Output map of values instead of csv tree')
 
     return parser.parse_args()
 
@@ -102,8 +107,8 @@ def main():
         raise IOError("Base column '%s' not present in any parameters!" %
                       base_conf.keys()[0])
 
-    base_table   = TupleTable(col_map)
-    result_table = TupleTable(col_map)
+    base_table = TupleTable(col_map) # For tracking 'base' experiments
+    result_table  = TupleTable(col_map) # For generating csv directories
 
     # Used to find matching scaling_base for each experiment
     for base in scaling_bases:
@@ -133,13 +138,17 @@ def main():
         if opts.verbose:
             print(result)
 
-    if opts.force and os.path.exists(opts.out_dir):
-        sh.rmtree(opts.out_dir)
+    if opts.force and os.path.exists(opts.out):
+        sh.rmtree(opts.out)
 
-    # Remove un-plottable values
     result_table.reduce()
 
-    result_table.write_result(opts.out_dir)
+    if opts.write_map:
+        # Write summarized results into map
+        result_table.write_map(opts.out)
+    else:
+        # Write out csv directories for all variable params
+        result_table.write_csvs(opts.out)
 
 if __name__ == '__main__':
     main()