changed name to pdrle and added get_sn

README.md

@@ -9,21 +9,20 @@ pip install git+https://github.com/darshanbaral/pyrle.git
 ## Usage
 
 ```python
-import pyrle
+import pdrle
 import pandas
 
-
 x = pandas.Series(["a", "a", "b", "b", "a", "a", "a", "c"])
 
-rle = pyrle.encode(x)
+rle = pdrle.encode(x)
 rle
 #   vals  runs
 # 0    a     2
 # 1    b     2
 # 2    a     3
 # 3    c     1
 
-y = pyrle.decode(rle.vals, rle.runs)
+y = pdrle.decode(rle.vals, rle.runs)
 y
 # 0    a
 # 1    a
@@ -35,7 +34,7 @@ y
 # 7    c
 # dtype: object
 
-pandas.concat({"x": x, "id": pyrle.get_id(x)}, axis=1)
+pandas.concat({"x": x, "id": pdrle.get_id(x)}, axis=1)
 #    x  id
 # 0  a   0
 # 1  a   0
@@ -45,4 +44,15 @@ pandas.concat({"x": x, "id": pyrle.get_id(x)}, axis=1)
 # 5  a   2
 # 6  a   2
 # 7  c   3
+
+pandas.concat({"x": x, "sn": pdrle.get_sn(x)}, axis=1)
+#    x  sn
+# 0  a   0
+# 1  a   1
+# 2  b   0
+# 3  b   1
+# 4  a   0
+# 5  a   1
+# 6  a   2
+# 7  c   0
 ```

pyrle/__init__.py → pdrle/__init__.py

@@ -45,3 +45,14 @@ def get_id(data: pandas.Series) -> pandas.Series:
 
     rle_id = check.cumsum().astype(numpy.int64)
     return rle_id - 1
+
+
+def get_sn(data: pandas.Series) -> pandas.Series:
+    """
+    Generates serial number for different elements of each consecutive runs of values in a pandas Series
+    :param data: input value, a pandas Series
+    :return: pandas Series
+    """
+    grp = get_id(data)
+    rle_sn = data.groupby(grp).cumcount()
+    return rle_sn

setup.cfg

@@ -0,0 +1,2 @@
+[metadata]
+description-file = README.md

setup.py

@@ -1,12 +1,12 @@
 from setuptools import setup
 
 
-setup(name="pyrle",
-      version="0.2",
+setup(name="pdrle",
+      version="0.3",
       description="python package for run length encoding on pandas Series",
       url="https://github.com/darshanbaral/pyrle",
       author="Darshan Baral",
       license="MIT",
-      packages=["pyrle"],
+      packages=["pdrle"],
       install_requires=["pandas"]
       )

tests/test.py

@@ -2,7 +2,7 @@
 from parameterized import parameterized
 import pandas
 import numpy
-import pyrle
+import pdrle
 
 
 class TestPyrle(unittest.TestCase):
@@ -25,7 +25,7 @@ class TestPyrle(unittest.TestCase):
                            "runs": [2, 3, 1]})]
     ])
     def test_encode(self, input_data, expected_output):
-        actual_output = pyrle.encode(input_data)
+        actual_output = pdrle.encode(input_data)
         pandas.testing.assert_frame_equal(actual_output, expected_output)
 
     # test decode
@@ -41,7 +41,7 @@ def test_encode(self, input_data, expected_output):
                            "runs": [1]})]
     ])
     def test_decode(self, expected_output, input_data):
-        actual_output = pyrle.decode(input_data.vals, input_data.runs)
+        actual_output = pdrle.decode(input_data.vals, input_data.runs)
         pandas.testing.assert_series_equal(actual_output, expected_output)
 
     # test get_id
@@ -60,5 +60,24 @@ def test_decode(self, expected_output, input_data):
          pandas.Series({"a": 0, "b": 0, "c": 1, "d": 1, "e": 1, "f": 2})]
     ])
     def test_get_id(self, input_data, expected_output):
-        actual_output = pyrle.get_id(input_data)
+        actual_output = pdrle.get_id(input_data)
         pandas.testing.assert_series_equal(actual_output, expected_output)
+
+    # test get_sn
+    @parameterized.expand([
+        [pandas.Series(["a", "a", "b", "b", "b", "a", "a", "c"]),
+         pandas.Series([0, 1, 0, 1, 2, 0, 1, 0])],
+        [pandas.Series([1, 1, 1, 1, 1, 1, 1]),
+         pandas.Series([0, 1, 2, 3, 4, 5, 6])],
+        [pandas.Series([2]),
+         pandas.Series([0])],
+        [pandas.Series({"a": 1, "b": 2, "c": 3, "d": 3, "e": 3, "f": 1}),
+         pandas.Series({"a": 0, "b": 0, "c": 0, "d": 1, "e": 2, "f": 0})],
+        [pandas.Series({"a": 1, "b": 1, "c": numpy.nan, "d": 1, "e": 2, "f": 2}),
+         pandas.Series({"a": 0, "b": 1, "c": 0, "d": 0, "e": 0, "f": 1})],
+        [pandas.Series({"a": 1, "b": 1, "c": numpy.nan, "d": numpy.nan, "e": numpy.nan, "f": 2}),
+         pandas.Series({"a": 0, "b": 1, "c": 0, "d": 1, "e": 2, "f": 0})]
+    ])
+    def test_get_sn(self, input_data, expected_output):
+        actual_output = pdrle.get_sn(input_data)
+        pandas.testing.assert_series_equal(actual_output, expected_output)