Merge pull request #42 from histogrammar/popmon_hgr_migration

Minor extra updates for popmon migration of histogrammar

Author: mbaak

README.rst

@@ -19,7 +19,7 @@ Histograms and other aggregators may also be converted into CUDA code for inclus
 PyCUDA is available, they can also be filled from Numpy arrays by JIT-compiling the CUDA code.
 This Python implementation of histogrammar been tested to guarantee compatibility with its Scala implementation.
 
-Latest Python release: v1.0.21 (Mar 2021).
+Latest Python release: v1.0.22 (Mar 2021).
 
 Announcements
 =============

histogrammar/dfinterface/histogram_filler_base.py

@@ -129,6 +129,13 @@ def __init__(
         # these get filled during execution
         self._hists = {}
 
+    def set_logger(self, logger):
+        """Set logger of module
+
+        :param logger: input logger
+        """
+        self.logger = logger
+
     def assert_dataframe(self, df):
         """assert dataframe datatype"""
         raise NotImplementedError("assert_dataframe not implemented!")

histogrammar/dfinterface/make_histograms.py

@@ -20,6 +20,7 @@
 
 import copy
 import logging
+import warnings
 
 import numpy as np
 import pandas as pd
@@ -159,16 +160,17 @@ def make_histograms(
         if not isinstance(bin_specs, (type(None), dict)):
             raise RuntimeError("bin_specs object is not a dictionary")
         bin_specs = copy.copy(bin_specs) if isinstance(bin_specs, dict) else {}
-        if time_axis in bin_specs:
-            raise RuntimeError(
-                f'time-axis "{time_axis}" already found in binning specifications.'
+        if time_axis not in bin_specs:
+            # convert time width and offset to nanoseconds
+            time_specs = {
+                "binWidth": float(pd.Timedelta(time_width).value),
+                "origin": float(pd.Timestamp(time_offset).value),
+            }
+            bin_specs[time_axis] = time_specs
+        else:
+            warnings.warn(
+                f'time-axis "{time_axis}" already found in binning specifications. not overwriting.'
             )
-        # convert time width and offset to nanoseconds
-        time_specs = {
-            "binWidth": float(pd.Timedelta(time_width).value),
-            "origin": float(pd.Timestamp(time_offset).value),
-        }
-        bin_specs[time_axis] = time_specs
 
     cls = PandasHistogrammar if isinstance(df, pd.DataFrame) else SparkHistogrammar
     hist_filler = cls(

histogrammar/version.py

@@ -3,9 +3,9 @@
 import re
 
 name = "histogrammar"
-__version__ = "1.0.21"
-version = "1.0.21"
-full_version = "1.0.21"
+__version__ = "1.0.22"
+version = "1.0.22"
+full_version = "1.0.22"
 release = True
 
 version_info = tuple(re.split(r"[-\.]", __version__))

setup.py

@@ -22,7 +22,7 @@
 
 MAJOR = 1
 REVISION = 0
-PATCH = 21
+PATCH = 22
 DEV = False
 # NOTE: also update version at: README.rst
 

tests/test_pandas_histogrammar.py

@@ -96,7 +96,7 @@ def test_make_histograms_no_time_axis():
     assert time_axis == ""
     assert "date" in hists
     h = hists["date"]
-    assert h.binWidth == 751582381944448.0
+    assert h.binWidth == 751582381944440.9
     for cols in features:
         cols = cols.split(":")
         assert len(cols) == 1
@@ -121,23 +121,23 @@ def test_make_histograms_with_time_axis():
     assert time_axis == "date"
     assert "date:age" in hists
     h = hists["date:age"]
-    assert h.binWidth == 751582381944448.0
+    assert h.binWidth == 751582381944440.9
     for cols in features:
         cols = cols.split(":")
         assert len(cols) == 2 and cols[0] == "date"
     for f, bs in bin_specs.items():
         assert len(bs) == 2
     assert "date:age" in bin_specs
     dateage = bin_specs["date:age"]
-    assert dateage[0]["binWidth"] == 751582381944448.0
+    assert dateage[0]["binWidth"] == 751582381944440.9
     assert dateage[1]["binWidth"] == 2.0
     assert dateage[1]["origin"] == 9.5
 
     # test get_bin_specs 1
     bin_specs = get_bin_specs(hists)
     assert "date:age" in bin_specs
     dateage = bin_specs["date:age"]
-    assert dateage[0]["binWidth"] == 751582381944448.0
+    assert dateage[0]["binWidth"] == 751582381944440.9
     assert dateage[1]["binWidth"] == 2.0
     assert dateage[1]["origin"] == 9.5
 
@@ -150,7 +150,7 @@ def test_make_histograms_with_time_axis():
 
     # test get_bin_specs 3
     bin_specs = get_bin_specs(hists["date:age"])
-    assert bin_specs[0]["binWidth"] == 751582381944448.0
+    assert bin_specs[0]["binWidth"] == 751582381944440.9
     assert bin_specs[1]["binWidth"] == 2.0
     assert bin_specs[1]["origin"] == 9.5