8 enh compare two and only two dataframes by common attributes (#23)

.github/CODEOWNERS

@@ -0,0 +1 @@
+* @hadarsharon

README.md

@@ -6,4 +6,6 @@
 
 <hr>
 
-compars is a simple-yet-comprehensive, bear-agnostic :bear: :panda_face: :koala: DataFrame comparison library, powered by Rust with polars.
+DataFrame comparison done right, powered by Rust with polars
+
+(AKA the bear-agnostic :bear: :panda_face: :koala: DataFrame comparison library)

compars/dataframe/types.py

@@ -0,0 +1,30 @@
+from typing import Iterable, Any
+
+from dask.dataframe.core import DataFrame as DaskDataFrame
+from modin.pandas.dataframe import DataFrame as ModinDataFrame
+from pandas.core.frame import DataFrame as PandasDataFrame
+from polars.dataframe.frame import DataFrame as PolarsDataFrame
+from pyspark.sql.dataframe import DataFrame as SparkDataFrame
+
+
+class DataFrame:
+    """Generic (bear-agnostic) DataFrame type"""
+
+    def __init__(self, df: PolarsDataFrame | PandasDataFrame | SparkDataFrame | DaskDataFrame | ModinDataFrame):
+        self._df = df
+
+    # TODO: property
+    def shape(self) -> tuple[int, int]:
+        if isinstance(self.df, (PolarsDataFrame, PandasDataFrame, ModinDataFrame)):
+            return self.df.shape
+        if isinstance(self.df, DaskDataFrame):
+            ...  # TODO: figure out
+        if isinstance(self.df, SparkDataFrame):
+            ...  # TODO: figure out
+
+    # TODO: property
+    def columns(self) -> Iterable[Any]:
+        return self.df.columns
+
+    def n_columns(self) -> int:
+        return len(self.columns())

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "compars"
-description = "Bear-agnostic DataFrame comparison library"
-version = "0.0.1"
+description = "DataFrame comparison done right (AKA the Bear-agnostic DataFrame comparison library)"
+version = "0.0.0"
 authors = [
     { name = "Hadar Sharon", email = "hadar.sharon94@gmail.com" },
 ]
@@ -53,4 +53,4 @@ exclude = [
 [tool.ruff.format]
 quote-style = "double"
 skip-magic-trailing-comma = false
-preview = true
+preview = true

requirements.txt

@@ -1,3 +1,8 @@
+dask==2024.2.1 # support for dask DataFrames
+modin[dask]==0.27.0 # support for modin DataFrames
+pandas==2.2.1 # support for pandas DataFrames
+polars==0.20.10 # support for polars DataFrames
 pre-commit==3.6.2 # pre-commit hooks
+pyspark==3.5.0 # support for spark DataFrames
 pytest==8.0.1 # testing framework
 ruff==0.2.2 # formatter/linter