Format, add code formatting tooling

Introduce riot command for formatting code using
[`black`](https://black.readthedocs.io/en/stable/) and
[`isort`](https://github.com/PyCQA/isort).

Instructions are added to the README as well.

Author: Kyle-Verhoog

.github/workflows/test.yml

@@ -5,6 +5,17 @@ on:
       - master
   pull_request:
 jobs:
+  check:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/setup-python@v2
+        with:
+          python-version: '3.9'
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 0
+      - run: pip install riot==0.8
+      - run: riot -v run check_fmt
   test:
     strategy:
       matrix:

README.md

@@ -87,5 +87,12 @@ Edit the generated file to include notes on the changes made in the commit/PR
 and add commit it.
 
 
+### Formatting
+
+Format code with
+
+    riot run fmt
+
+
 ## References
 [1] Charles Masson and Jee E Rim and Homin K. Lee. DDSketch: A fast and fully-mergeable quantile sketch with relative-error guarantees. PVLDB, 12(12): 2195-2205, 2019. (The code referenced in the paper, including our implementation of the the Greenwald-Khanna (GK) algorithm, can be found at: https://github.com/DataDog/sketches-py/releases/tag/v0.1 )

ddsketch/__init__.py

@@ -1,3 +1,4 @@
 from .ddsketch import DDSketch
 
+
 __all__ = ["DDSketch"]

ddsketch/ddsketch.py

@@ -35,9 +35,12 @@
 <a href="https://github.com/DataDog/sketches-js/">JavaScript</a>
 """
 
-from .exception import IllegalArgumentException, UnequalSketchParametersException
+from .exception import IllegalArgumentException
+from .exception import UnequalSketchParametersException
 from .mapping import LogarithmicMapping
-from .store import CollapsingHighestDenseStore, CollapsingLowestDenseStore, DenseStore
+from .store import CollapsingHighestDenseStore
+from .store import CollapsingLowestDenseStore
+from .store import DenseStore
 
 
 DEFAULT_REL_ACC = 0.01  # "alpha" in the paper

ddsketch/mapping.py

@@ -16,7 +16,8 @@
 when computing the index. Other mappings can approximate the logarithmic
 mapping, while being less computationally costly.
 """
-from abc import ABC, abstractmethod
+from abc import ABC
+from abc import abstractmethod
 import math
 import sys
 
@@ -52,17 +53,17 @@ def __init__(self, relative_accuracy, offset=0.0):
 
     @classmethod
     def from_gamma_offset(cls, gamma, offset):
-        """ Constructor used by pb.proto """
+        """Constructor used by pb.proto"""
         relative_accuracy = (gamma - 1.0) / (gamma + 1.0)
         return cls(relative_accuracy, offset=offset)
 
     @abstractmethod
     def _log_gamma(self, value):
-        """ Return (an approximation of) the logarithm of the value base gamma """
+        """Return (an approximation of) the logarithm of the value base gamma"""
 
     @abstractmethod
     def _pow_gamma(self, value):
-        """ Return (an approximation of) gamma to the power value """
+        """Return (an approximation of) gamma to the power value"""
 
     def key(self, value):
         """
@@ -102,7 +103,7 @@ def _pow_gamma(self, value):
 
 def _cbrt(x):
     # type: (float) -> float
-    y = abs(x)**(1./3.)
+    y = abs(x) ** (1.0 / 3.0)
     if x < 0:
         return -y
     return y
@@ -127,7 +128,7 @@ def _log2_approx(self, value):
         return significand + (exponent - 1)
 
     def _exp2_approx(self, value):
-        """ inverse of _log2_approx """
+        """inverse of _log2_approx"""
         exponent = math.floor(value) + 1
         mantissa = (value - exponent + 2) / 2.0
         return math.ldexp(mantissa, exponent)
@@ -158,15 +159,15 @@ def __init__(self, relative_accuracy, offset=0.0):
         self._multiplier /= self.C
 
     def _cubic_log2_approx(self, value):
-        """ approximates log2 using a cubic polynomial """
+        """approximates log2 using a cubic polynomial"""
         mantissa, exponent = math.frexp(value)
         significand = 2 * mantissa - 1
         return (
             (self.A * significand + self.B) * significand + self.C
         ) * significand + (exponent - 1)
 
     def _cubic_exp2_approx(self, value):
-        """ Derived from Cardano's formula """
+        """Derived from Cardano's formula"""
 
         exponent = math.floor(value)
         delta_0 = self.B * self.B - 3 * self.A * self.C
@@ -175,7 +176,10 @@ def _cubic_exp2_approx(self, value):
             - 9 * self.A * self.B * self.C
             - 27 * self.A * self.A * (value - exponent)
         )
-        cardano = _cbrt((delta_1 - ((delta_1 * delta_1 - 4 * delta_0 * delta_0 * delta_0)**.5)) / 2)
+        cardano = _cbrt(
+            (delta_1 - ((delta_1 * delta_1 - 4 * delta_0 * delta_0 * delta_0) ** 0.5))
+            / 2
+        )
         significand_plus_one = (
             -(self.B + cardano + delta_0 / cardano) / (3 * self.A) + 1
         )

ddsketch/store.py

@@ -7,7 +7,8 @@
 We start with 128 bins and grow the store in chunks of 128 unless specified
 otherwise."""
 
-from abc import ABC, abstractmethod
+from abc import ABC
+from abc import abstractmethod
 import math
 
 
@@ -167,7 +168,7 @@ def _shift_bins(self, shift):
         self.offset -= shift
 
     def _center_bins(self, new_min_key, new_max_key):
-        """ center the bins; this changes the offset"""
+        """center the bins; this changes the offset"""
         middle_key = new_min_key + (new_max_key - new_min_key + 1) // 2
         self._shift_bins(self.offset + self.length() // 2 - middle_key)
 

pyproject.toml

@@ -0,0 +1,20 @@
+[tool.isort]
+force_single_line = true
+lines_after_imports = 2
+force_sort_within_sections = true
+known_first_party = "ddsketch"
+default_section = "THIRDPARTY"
+skip = [".riot/", ".venv/", "ddsketch/pb"]
+line_length = 120
+
+[tool.black]
+exclude = '''
+(
+  /(
+    \.riot
+  | ddsketch/pb.*
+  | \.venv.*
+  | \.eggs
+  )/
+)
+'''

riotfile.py

@@ -22,6 +22,27 @@
                 Venv(
                     name="reno",
                     command="reno {cmdargs}",
+                )
+            ],
+        ),
+        Venv(
+            pkgs={
+                "black": "==21.7b0",
+                "isort": latest,
+                "toml": latest,
+            },
+            venvs=[
+                Venv(
+                    name="black",
+                    command="black {cmdargs}",
+                ),
+                Venv(
+                    name="fmt",
+                    command="isort . && black .",
+                ),
+                Venv(
+                    name="check_fmt",
+                    command="isort --check . && black --check .",
                 ),
             ],
         ),

setup.py

@@ -1,5 +1,6 @@
 import setuptools
 
+
 with open("README.md", "r", encoding="utf-8") as fh:
     long_description = fh.read()
 

tests/datasets.py

@@ -3,7 +3,9 @@
 # This product includes software developed at Datadog (https://www.datadoghq.com/).
 # Copyright 2020 Datadog, Inc.
 
-from abc import ABC, abstractmethod, abstractproperty
+from abc import ABC
+from abc import abstractmethod
+from abc import abstractproperty
 
 import numpy as np