add human-readable __str__ and __repr__ methods to dataframeschema (#111)

* add human-readable __str__ and __repr__ methods to dataframeschema

* remove gt/lt sign in __str__ method

Author: cosmicBboy

pandera/schemas.py

@@ -1,5 +1,7 @@
 """Core pandera schema class definitions."""
 
+import json
+
 import pandas as pd
 
 from typing import Optional
@@ -8,6 +10,9 @@
 from .checks import Check
 
 
+N_INDENT_SPACES = 4
+
+
 class DataFrameSchema(object):
     """A light-weight pandas DataFrame validator."""
 
@@ -52,6 +57,7 @@ def __init__(
         self.coerce = coerce
         self.strict = strict
         self._validate_schema()
+        self._set_column_names()
 
     def __call__(
             self,
@@ -87,6 +93,12 @@ def _validate_schema(self):
                         "specified in the DataFrameSchema." %
                         (nonexistent_groupby_columns, column_name))
 
+    def _set_column_names(self):
+        self.columns = {
+            column_name: column.set_name(column_name)
+            for column_name, column in self.columns.items()
+        }
+
     @staticmethod
     def _dataframe_to_validate(
             dataframe: pd.DataFrame,
@@ -168,6 +180,40 @@ def validate(
             dataframe = self.transformer(dataframe)
         return dataframe
 
+    def __repr__(self):
+        return "%s(columns=%s, index=%s, transformer=%s, coerce=%s)" % \
+              (self.__class__.__name__,
+               self.columns,
+               self.index,
+               self.transformer,
+               self.coerce)
+
+    def __str__(self):
+        columns = {k: str(v) for k, v in self.columns.items()}
+        columns = json.dumps(columns, indent=N_INDENT_SPACES)
+        _indent = " " * N_INDENT_SPACES
+        columns = "\n".join(
+            "{}{}".format(_indent, line) if i != 0
+            else "{}columns={}".format(_indent, line)
+            for i, line in enumerate(columns.split("\n")))
+        return (
+            "{class_name}(\n"
+            "{columns},\n"
+            "{indent}index={index},\n"
+            "{indent}transformer={transformer},\n"
+            "{indent}coerce={coerce},\n"
+            "{indent}strict={strict}\n"
+            ")"
+        ).format(
+            class_name=self.__class__.__name__,
+            columns=columns,
+            index=str(self.index),
+            transformer=str(self.transformer),
+            coerce=self.coerce,
+            strict=self.strict,
+            indent=_indent,
+        )
+
 
 class SeriesSchemaBase(object):
     """Base series validator object."""

tests/test_pandera.py

@@ -1022,3 +1022,21 @@ def test_dataframe_hypothesis_checks():
     )
     with pytest.raises(errors.SchemaDefinitionError):
         hypothesis_check_schema_groupby.validate(df)
+
+
+def test_dataframe_schema_str_repr():
+    schema = DataFrameSchema(
+        columns={
+            "col1": Column(Int),
+            "col2": Column(String),
+            "col3": Column(DateTime),
+        },
+        index=Index(Int, name="my_index"),
+
+    )
+
+    for x in [schema.__str__(), schema.__repr__()]:
+        assert isinstance(x, str)
+        assert schema.__class__.__name__ in x
+        for name in ["col1", "col2", "col3", "my_index"]:
+            assert name in x