[ENH] Implements sklearn style pretty printing directly in skbase

docs/source/api_reference.rst

@@ -28,6 +28,27 @@ Base Classes
     BaseObject
     BaseEstimator
 
+.. _global_config:
+
+Configure ``skbase``
+====================
+
+.. automodule:: skbase.config
+    :no-members:
+    :no-inherited-members:
+
+.. currentmodule:: skbase.config
+
+.. autosummary::
+    :toctree: api_reference/auto_generated/
+    :template: function.rst
+
+    get_config
+    get_default_config
+    set_config
+    reset_config
+    config_context
+
 .. _obj_retrieval:
 
 Object Retrieval

docs/source/user_documentation/user_guide.rst

@@ -29,6 +29,7 @@ that ``skbase`` provides, see the :ref:`api_ref`.
    user_guide/lookup
    user_guide/validate
    user_guide/testing
+   user_guide/configuration
 
 
 .. grid:: 1 2 2 2
@@ -103,3 +104,17 @@ that ``skbase`` provides, see the :ref:`api_ref`.
             :expand:
 
             Testing
+
+    .. grid-item-card:: Configuration
+        :text-align: center
+
+        Configure ``skbase``.
+
+        +++
+
+        .. button-ref:: user_guide/configuration
+            :color: primary
+            :click-parent:
+            :expand:
+
+            Configuration

docs/source/user_documentation/user_guide/configuration.rst

@@ -0,0 +1,11 @@
+.. _user_guide_global_config:
+
+====================
+Configure ``skbase``
+====================
+
+.. note::
+
+    The user guide is under development. We have created a basic
+    structure and are looking for contributions to develop the user guide
+    further.

skbase/base/_base.py

@@ -53,6 +53,7 @@ class name: BaseEstimator
     fitted state check      - check_is_fitted (raises error if not is_fitted)
 """
 import inspect
+import re
 import warnings
 from collections import defaultdict
 from copy import deepcopy
@@ -62,7 +63,9 @@ class name: BaseEstimator
 from sklearn.base import BaseEstimator as _BaseEstimator
 
 from skbase._exceptions import NotFittedError
+from skbase.base._pretty_printing._object_html_repr import _object_html_repr
 from skbase.base._tagmanager import _FlagManager
+from skbase.config import get_config
 
 __author__: List[str] = ["mloning", "RNKuhns", "fkiraly"]
 __all__: List[str] = ["BaseEstimator", "BaseObject"]
@@ -446,7 +449,28 @@ def get_config(self):
             class attribute via nested inheritance and then any overrides
             and new tags from _onfig_dynamic object attribute.
         """
-        return self._get_flags(flag_attr_name="_config")
+        config = get_config().copy()
+
+        # Get any extension configuration interface defined in the class
+        # for example if downstream package wants to extend skbase to retrieve
+        # their own config
+        if hasattr(self, "__skbase_get_config__") and callable(
+            self.__skbase_get_config__
+        ):
+            skbase_get_config_extension_dict = self.__skbase_get_config__()
+        else:
+            skbase_get_config_extension_dict = {}
+        if isinstance(skbase_get_config_extension_dict, dict):
+            config.update(skbase_get_config_extension_dict)
+        else:
+            msg = "Use of `__skbase_get_config__` to extend the interface for local "
+            msg += "overrides of the global configuration must return a dictionary.\n"
+            msg += f"But a {type(skbase_get_config_extension_dict)} was found."
+            warnings.warn(msg, UserWarning, stacklevel=2)
+        local_config = self._get_flags(flag_attr_name="_config")
+        config.update(local_config)
+
+        return config
 
     def set_config(self, **config_dict):
         """Set config flags to given values.
@@ -682,6 +706,98 @@ def _components(self, base_class=None):
 
         return comp_dict
 
+    def __repr__(self, n_char_max: int = 700):
+        """Represent class as string.
+
+        This follows the scikit-learn implementation for the string representation
+        of parameterized objects.
+
+        Parameters
+        ----------
+        n_char_max : int
+            Maximum (approximate) number of non-blank characters to render. This
+            can be useful in testing.
+        """
+        from skbase.base._pretty_printing._pprint import _BaseObjectPrettyPrinter
+
+        n_max_elements_to_show = 30  # number of elements to show in sequences
+        # use ellipsis for sequences with a lot of elements
+        pp = _BaseObjectPrettyPrinter(
+            compact=True,
+            indent=1,
+            indent_at_name=True,
+            n_max_elements_to_show=n_max_elements_to_show,
+            changed_only=get_config()["print_changed_only"],
+        )
+
+        repr_ = pp.pformat(self)
+
+        # Use bruteforce ellipsis when there are a lot of non-blank characters
+        n_nonblank = len("".join(repr_.split()))
+        if n_nonblank > n_char_max:
+            lim = n_char_max // 2  # apprx number of chars to keep on both ends
+            regex = r"^(\s*\S){%d}" % lim
+            # The regex '^(\s*\S){%d}' matches from the start of the string
+            # until the nth non-blank character:
+            # - ^ matches the start of string
+            # - (pattern){n} matches n repetitions of pattern
+            # - \s*\S matches a non-blank char following zero or more blanks
+            left_match = re.match(regex, repr_)
+            right_match = re.match(regex, repr_[::-1])
+            left_lim = left_match.end() if left_match is not None else 0
+            right_lim = right_match.end() if right_match is not None else 0
+
+            if "\n" in repr_[left_lim:-right_lim]:
+                # The left side and right side aren't on the same line.
+                # To avoid weird cuts, e.g.:
+                # categoric...ore',
+                # we need to start the right side with an appropriate newline
+                # character so that it renders properly as:
+                # categoric...
+                # handle_unknown='ignore',
+                # so we add [^\n]*\n which matches until the next \n
+                regex += r"[^\n]*\n"
+                right_match = re.match(regex, repr_[::-1])
+                right_lim = right_match.end() if right_match is not None else 0
+
+            ellipsis = "..."
+            if left_lim + len(ellipsis) < len(repr_) - right_lim:
+                # Only add ellipsis if it results in a shorter repr
+                repr_ = repr_[:left_lim] + "..." + repr_[-right_lim:]
+
+        return repr_
+
+    @property
+    def _repr_html_(self):
+        """HTML representation of BaseObject.
+
+        This is redundant with the logic of `_repr_mimebundle_`. The latter
+        should be favorted in the long term, `_repr_html_` is only
+        implemented for consumers who do not interpret `_repr_mimbundle_`.
+        """
+        if get_config()["display"] != "diagram":
+            raise AttributeError(
+                "_repr_html_ is only defined when the "
+                "`display` configuration option is set to 'diagram'."
+            )
+        return self._repr_html_inner
+
+    def _repr_html_inner(self):
+        """Return HTML representation of class.
+
+        This function is returned by the @property `_repr_html_` to make
+        `hasattr(BaseObject, "_repr_html_") return `True` or `False` depending
+        on `get_config()["display"]`.
+        """
+        return _object_html_repr(self)
+
+    def _repr_mimebundle_(self, **kwargs):
+        """Mime bundle used by jupyter kernels to display instances of BaseObject."""
+        output = {"text/plain": repr(self)}
+        if get_config()["display"] == "diagram":
+            output["text/html"] = _object_html_repr(self)
+        return output
+
 
 class TagAliaserMixin:
     """Mixin class for tag aliasing and deprecation of old tags.

skbase/base/_pretty_printing/__init__.py

@@ -0,0 +1,11 @@
+#!/usr/bin/env python3 -u
+# -*- coding: utf-8 -*-
+# copyright: skbase developers, BSD-3-Clause License (see LICENSE file)
+# Many elements of this code were developed in scikit-learn. These elements
+# are copyrighted by the scikit-learn developers, BSD-3-Clause License. For
+# conditions see https://github.com/scikit-learn/scikit-learn/blob/main/COPYING
+"""Functionality for pretty printing BaseObjects."""
+from typing import List
+
+__author__: List[str] = ["RNKuhns"]
+__all__: List[str] = []