Merge pull request #31 from FAST-HEP/BK_add_postproc_module

Postproc: CLI option to control the column names used for values

CHANGELOG.md

@@ -4,6 +4,13 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.6.1] - 2020-04-20
+### Added
+- Option to specify which columns contain bin values rather than labels, PR #31 [@benkrikler](github.com/benkrikler)
+
+### Fixed
+- Report package version number properly, PR #31
+
 ## [0.6.0] - 2020-04-17
 ### Added
 - Proper support for non-constant interval widths

fast_plotter/__init__.py

@@ -1,5 +1,8 @@
 # -*- coding: utf-8 -*-
 
+from .version import __version__, version_info
+
+
+__all__ = ["__version__", "version_info"]
 __author__ = """Ben Krikler"""
 __email__ = 'fast-hep@cern.ch'
-__version__ = '0.3.0'

fast_plotter/__main__.py

@@ -6,6 +6,7 @@
 import logging
 import matplotlib
 matplotlib.use('Agg')
+from .version import __version__ # noqa
 from .utils import read_binned_df, weighting_vars # noqa
 from .utils import decipher_filename, mask_rows  # noqa
 from .plotting import plot_all, add_annotations # noqa
@@ -42,6 +43,7 @@ def arg_parser(args=None):
                         help="Scale the MC yields by this lumi")
     parser.add_argument("-y", "--yscale", default="log", choices=["log", "linear"],
                         help="Use this scale for the y-axis")
+    parser.add_argument('--version', action='version', version='%(prog)s ' + __version__)
 
     def split_equals(arg):
         return arg.split("=")

fast_plotter/postproc/__main__.py

@@ -19,6 +19,9 @@ def make_parser():
                         help="A yaml to configure the post-processing step")
     parser.add_argument("-o", "--outdir", default=".",
                         help="The name of the output directory")
+    parser.add_argument("-V", "--value-cols", default=r"(.*sumw2?|n)",
+                        help="A regular expression to control which columns are"
+                        " identified as values and not bin labels")
     parser.add_argument("files", nargs="+",
                         help="Input dataframes that need merging together")
     return parser
@@ -50,7 +53,7 @@ def main(args=None):
     if args.debug_dfs:
         logger.setLevel(logging.DEBUG)
 
-    dfs = open_many(args.files)
+    dfs = open_many(args.files, value_columns=args.value_cols)
     sequence = read_processing_cfg(args.post_process, args.outdir)
 
     for stage in sequence:

fast_plotter/postproc/functions.py

@@ -1,5 +1,6 @@
 import os
 import six
+import re
 import numpy as np
 import pandas as pd
 import logging
@@ -365,15 +366,15 @@ def normalise_group(df, groupby_dimensions, apply_if=None, use_column=None):
     return normed
 
 
-def open_many(file_list, return_meta=True):
+def open_many(file_list, value_columns=r"(.*sumw2?|n)", return_meta=True):
     """ Open a list of dataframe files
     """
     dfs = []
     for fname in file_list:
         df = pd.read_csv(fname, comment='#')
         drop_cols = [col for col in df.columns if "Unnamed" in col]
         df.drop(drop_cols, inplace=True, axis="columns")
-        index_cols = [col for col in df.columns if "sumw" not in col and col != "n"]
+        index_cols = [col for col in df.columns if not re.match(value_columns, col)]
         df.set_index(index_cols, inplace=True, drop=True)
         if return_meta:
             name = fname

fast_plotter/version.py

@@ -12,5 +12,5 @@ def split_version(version):
     return tuple(result)
 
 
-__version__ = '0.6.0'
+__version__ = '0.6.1'
 version_info = split_version(__version__) # noqa

setup.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.6.0
+current_version = 0.6.1
 commit = True
 tag = False