0.0.1

.pre-commit-config.yaml

@@ -6,3 +6,40 @@ repos:
     -   id: trailing-whitespace
     -   id: end-of-file-fixer
     -   id: check-yaml
+-   repo: local
+    hooks:
+    -   id: masterpylintrc
+        name: Overwrite local .pylintrc by master one
+        entry: cp ../.pylintrc ./.pylintrc
+        pass_filenames: false
+        always_run: true
+        language: system
+-   repo: https://github.com/pre-commit/mirrors-pylint
+    rev: v2.4.4
+    hooks:
+    -   id: pylint
+        files: ^datar/.+$
+        pass_filenames: false
+        types: [python]
+        args: [datar]
+-   repo: local
+    hooks:
+    -   id: poetry2setuppy
+        name: Convert pyproject.toml to setup.py
+        entry: dephell deps convert --from=poetry --to=setup.py
+        language: system
+        files: pyproject.toml
+        pass_filenames: false
+    -   id: poetry2requirements
+        name: Convert pyproject.toml to requirements.txt
+        entry: dephell deps convert --from=poetry --to=requirements.txt
+        language: system
+        files: pyproject.toml
+        pass_filenames: false
+    -   id: pytest
+        name: Run pytest
+        entry: pytest
+        language: system
+        args: [tests/]
+        pass_filenames: false
+        files: ^tests/.+$|^pipda/.+$

datar/__init__.py

@@ -6,4 +6,4 @@
 
 f = Symbolic() # pylint: disable=invalid-name
 
-__version__ = '0.0.0'
+__version__ = '0.0.1'

datar/tidyr/verbs.py

@@ -2,13 +2,13 @@
 from typing import Any, Callable, Mapping, Optional, Type, Union
 
 import numpy
-from numpy.core.fromnumeric import shape
 import pandas
-from pandas import DataFrame, Series
+from pandas import DataFrame
+from pandas.core.groupby.generic import DataFrameGroupBy
 from pipda import register_verb, Context
 
-from ..core.utils import select_columns, list_diff
-from ..core.types import IntOrIter, StringOrIter, is_scalar
+from ..core.utils import objectize, select_columns, list_diff
+from ..core.types import DataFrameType, IntOrIter, StringOrIter, is_scalar
 
 @register_verb(DataFrame, context=Context.SELECT)
 def pivot_longer(
@@ -207,13 +207,31 @@ def get_new_colname(cols, names):
 
     return ret
 
-@register_verb(DataFrame, context=Context.EVAL)
+@register_verb((DataFrame, DataFrameGroupBy), context=Context.EVAL)
 def uncount(
-        _data: DataFrame,
+        _data: DataFrameType,
         weights: IntOrIter,
         _remove: bool = True,
         _id: Optional[str] = None,
-) -> DataFrame:
+) -> DataFrameType:
+    """Duplicating rows according to a weighting variable
+
+    Args:
+        _data: A data frame
+        weights: A vector of weights. Evaluated in the context of data
+        _remove: If TRUE, and weights is the name of a column in data,
+            then this column is removed.
+        _id: Supply a string to create a new variable which gives a
+            unique identifier for each created row (0-based).
+
+    Returns:
+        dataframe with rows repeated.
+    """
+    gnames = (
+        _data.grouper.names
+        if isinstance(_data, DataFrameGroupBy) else None
+    )
+    _data = objectize(_data)
     if is_scalar(weights):
         weights = [weights] * _data.shape[0]
 
@@ -231,7 +249,9 @@ def uncount(
 
     ret = _data.loc[indexes, rest_columns] if _remove else _data.loc[indexes, :]
     if _id:
-        return ret.groupby(rest_columns).apply(
+        ret = ret.groupby(rest_columns).apply(
             lambda df: df.assign(**{_id: range(df.shape[0])})
         ).reset_index(drop=True, level=0)
+    if gnames:
+        return ret.groupby(gnames, dropna=False)
     return ret

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "datar"
-version = "0.0.0"
+version = "0.0.1"
 description = "Probably the closest port of tidyr, dplyr and tibble in python"
 authors = ["pwwang <[email protected]>"]
 license = "MIT"

requirements.txt

@@ -0,0 +1,5 @@
+modkit
+pandas==1.*,>=1.2.0
+pipda
+pytest
+pytest-cov

setup.py

@@ -15,7 +15,7 @@
 setup(
     long_description=readme,
     name='datar',
-    version='0.0.0',
+    version='0.0.1',
     description='Probably the closest port of tidyr, dplyr and tibble in python',
     python_requires='==3.*,>=3.7.1',
     author='pwwang',