From 0ab4a6d7f6a1667f75d5b51edb2c749bb60f7a4f Mon Sep 17 00:00:00 2001
From: Ben Krikler <bek07@ic.ac.uk>
Date: Thu, 7 May 2020 13:01:43 +0200
Subject: [PATCH 1/6] Add option to rotate xtick labels

---
 fast_plotter/__main__.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/fast_plotter/__main__.py b/fast_plotter/__main__.py
index 8bfe4c1..5a3c9cd 100644
--- a/fast_plotter/__main__.py
+++ b/fast_plotter/__main__.py
@@ -6,6 +6,7 @@
 import logging
 import matplotlib
 matplotlib.use('Agg')
+matplotlib.rcParams.update({'figure.autolayout': True})
 from .version import __version__ # noqa
 from .utils import read_binned_df, weighting_vars # noqa
 from .utils import decipher_filename, mask_rows  # noqa
@@ -134,8 +135,9 @@ def process_one_file(infile, args):
     return ran_ok
 
 
-def dress_main_plots(plots, annotations=[], yscale=None, ylabel=None, legend={}, limits={}, **kwargs):
-    for main_ax, _ in plots.values():
+def dress_main_plots(plots, annotations=[], yscale=None, ylabel=None, legend={},
+                     limits={}, xtickrotation=None, **kwargs):
+    for main_ax, summary_ax in plots.values():
         add_annotations(annotations, main_ax)
         if yscale:
             main_ax.set_yscale(yscale)
@@ -151,6 +153,8 @@ def dress_main_plots(plots, annotations=[], yscale=None, ylabel=None, legend={},
                     getattr(main_ax, "set_%slim" % axis)(*lims)
             elif lims.endswith("%"):
                 main_ax.margins(**{axis: float(lims[:-1])})
+        if xtickrotation:
+            matplotlib.pyplot.xticks(rotation=xtickrotation)
 
 
 def save_plots(infile, weight, plots, outdir, extensions):

From ea27a69fdb2455224fb671f7f10912b598c940ab Mon Sep 17 00:00:00 2001
From: Ben Krikler <bek07@ic.ac.uk>
Date: Thu, 7 May 2020 13:02:50 +0200
Subject: [PATCH 2/6] Add option to set output of assign_cols to be an index

---
 fast_plotter/postproc/functions.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fast_plotter/postproc/functions.py b/fast_plotter/postproc/functions.py
index 53de7ae..a0a9e9d 100644
--- a/fast_plotter/postproc/functions.py
+++ b/fast_plotter/postproc/functions.py
@@ -152,7 +152,7 @@ def keep_specific_bins(df, axis, keep, expansions={}):
     return out_df
 
 
-def combine_cols(df, format_strings):
+def combine_cols(df, format_strings, as_index=[]):
     """Combine columns together using format strings"""
     logger.info("Combining columns based on: %s", str(format_strings))
     result_names = list(format_strings.keys())
@@ -166,6 +166,8 @@ def apply_fmt(row):
     results.columns = result_names
     new_df = new_df.assign(**results)
     new_df.set_index(index, inplace=True, drop=True)
+    if as_index:
+        new_df.set_index(as_index, inplace=True, append=True)
     return new_df
 
 

From 9e279bc56ad950cef358b144840f1fa436f964f7 Mon Sep 17 00:00:00 2001
From: Ben Krikler <bek07@ic.ac.uk>
Date: Thu, 7 May 2020 13:03:15 +0200
Subject: [PATCH 3/6] Fix up to_datacard_inputs

---
 fast_plotter/postproc/functions.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fast_plotter/postproc/functions.py b/fast_plotter/postproc/functions.py
index a0a9e9d..66f0129 100644
--- a/fast_plotter/postproc/functions.py
+++ b/fast_plotter/postproc/functions.py
@@ -290,6 +290,8 @@ def to_datacard_inputs(df, select_data, rename_syst_vars=False):
     df["content"] = df.n
     df["content"][~data_mask] = df.sumw
     df["error"] = df.content / np.sqrt(df.n)
+    df.drop(["n", "sumw", "sumw2"], inplace=True, axis="columns")
+    return df
 
 
 def generic_pandas(df, func, *args, **kwargs):

From 3816f28081f9a1dee7439d2752ab48c6c7a7765e Mon Sep 17 00:00:00 2001
From: Ben Krikler <bek07@ic.ac.uk>
Date: Thu, 7 May 2020 13:03:33 +0200
Subject: [PATCH 4/6] Add new stage to remove columns in a more convenient way

---
 fast_plotter/postproc/functions.py | 24 ++++++++++++++++++++++++
 fast_plotter/postproc/stages.py    |  5 +++++
 tests/postproc/test_functions.py   | 24 ++++++++++++++++++++++++
 3 files changed, 53 insertions(+)

diff --git a/fast_plotter/postproc/functions.py b/fast_plotter/postproc/functions.py
index 66f0129..cc8e1ce 100644
--- a/fast_plotter/postproc/functions.py
+++ b/fast_plotter/postproc/functions.py
@@ -152,6 +152,30 @@ def keep_specific_bins(df, axis, keep, expansions={}):
     return out_df
 
 
+def filter_cols(df, items=None, like=None, regex=None, drop_not_keep=False):
+    """Filter out columns you want to keep.
+
+    Parameters:
+      items (list-like): A list of column names to filter with
+      like (str, list[string]): A string or list of strings which will filter
+            columns where they are found in the column name
+      regex (str): A regular expression to match column names to
+      drop_not_keep (bool): Inverts the selection if true so that matched columns are dropped
+    """
+    if not like or not isinstance(like, (tuple, list)):
+        df_filtered = df.filter(items=items, like=like, regex=regex)
+    elif like:
+        if items and like:
+            raise RuntimeError("Can only use one of 'items', 'like', or 'regex'")
+        filtered = [set(col for col in df.columns if i in col) for i in like]
+        filtered = set.union(*filtered)
+        df_filtered = df.filter(items=filtered, regex=regex)
+
+    if drop_not_keep:
+        return df.drop(df_filtered.columns)
+    return df_filtered
+
+
 def combine_cols(df, format_strings, as_index=[]):
     """Combine columns together using format strings"""
     logger.info("Combining columns based on: %s", str(format_strings))
diff --git a/fast_plotter/postproc/stages.py b/fast_plotter/postproc/stages.py
index d56f6c6..d97bc7c 100644
--- a/fast_plotter/postproc/stages.py
+++ b/fast_plotter/postproc/stages.py
@@ -111,6 +111,11 @@ class AssignCol(BaseManipulator):
     func = "assign_col"
 
 
+class FilterCols(BaseManipulator):
+    cardinality = "one-to-one"
+    func = "filter_cols"
+
+
 class AssignDim(BaseManipulator):
     cardinality = "one-to-one"
     func = "assign_dim"
diff --git a/tests/postproc/test_functions.py b/tests/postproc/test_functions.py
index 89dd05a..a2fc89c 100644
--- a/tests/postproc/test_functions.py
+++ b/tests/postproc/test_functions.py
@@ -91,6 +91,30 @@ def test_split(binned_df):
     assert all([r[0].index.nlevels == 3 for r in results])
 
 
+def test_filter_cols(binned_df):
+    df = binned_df.index.to_frame()
+
+    result = funcs.filter_cols(df, items=["int"])
+    assert len(result.columns) == 1
+    assert result.columns[0] == "int"
+
+    result = funcs.filter_cols(df, items=["int", "cat"])
+    assert len(result.columns) == 2
+    assert set(result.columns) == set(("int", "cat"))
+
+    result = funcs.filter_cols(df, like="int")
+    assert len(result.columns) == 2
+    assert set(result.columns) == set(("int", "interval"))
+
+    result = funcs.filter_cols(df, like=["int", "cat"])
+    assert len(result.columns) == 3
+    assert set(result.columns) == set(("int", "cat", "interval"))
+
+    result = funcs.filter_cols(df, regex="^int.*")
+    assert len(result.columns) == 2
+    assert set(result.columns) == set(("int", "interval"))
+
+
 # def test_reorder_dimensions():
 #     #def reorder_dimensions(df, order):
 #     pass

From 76bdfb2b7e7dda412f380038686013f1b84d9ab4 Mon Sep 17 00:00:00 2001
From: Ben Krikler <bek07@ic.ac.uk>
Date: Thu, 7 May 2020 13:07:50 +0200
Subject: [PATCH 5/6] Update CHANGELOG

---
 CHANGELOG.md | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index bdab811..a9463e1 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -4,6 +4,14 @@ 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).
 
+## [0.6.4] - 2020-05-07
+### Added
+- New postprocessing stage to filter columns, PR #34 [@benkrikler](https://github.com/benkrikler)
+- New option to AssignCols stage to make assignment the index, PR #34
+
+## Fixed
+- ToDatacardInputs had been broken in a previous update, PR #34
+
 ## [0.6.3] - 2020-04-29
 ### Added
 - Add GenericPandas and UnstackWeights stages, PR #33 [@benkrikler](https://github.com/benkrikler)

From 98fda7876868e80c66acb4749227b2344764b9a4 Mon Sep 17 00:00:00 2001
From: Ben Krikler <bek07@ic.ac.uk>
Date: Thu, 7 May 2020 13:22:57 +0200
Subject: [PATCH 6/6] =?UTF-8?q?Bump=20version:=200.6.3=20=E2=86=92=200.6.4?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 fast_plotter/version.py | 2 +-
 setup.cfg               | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/fast_plotter/version.py b/fast_plotter/version.py
index f8e8cef..af940b7 100644
--- a/fast_plotter/version.py
+++ b/fast_plotter/version.py
@@ -12,5 +12,5 @@ def split_version(version):
     return tuple(result)
 
 
-__version__ = '0.6.3'
+__version__ = '0.6.4'
 version_info = split_version(__version__) # noqa
diff --git a/setup.cfg b/setup.cfg
index eb63e1d..6a4046c 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.6.3
+current_version = 0.6.4
 commit = True
 tag = False