create new export() method in PageObjectRegistry

Author: BurnzZ

docs/intro/pop.rst

@@ -118,6 +118,8 @@ by simply importing them:
     page = FurnitureProductPage(response)
     item = page.to_item()
 
+.. _`pop-recommended-requirements`:
+
 Recommended Requirements
 ~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -143,13 +145,15 @@ inside of ``ecommerce-page-objects/ecommerce_page_objects/__init__.py``:
 
 .. code-block:: python
 
-    from web_poet import default_registry, consume_modules
+    from web_poet import default_registry
 
-    # This allows all of the OverrideRules declared inside the package
-    # using @handle_urls to be properly discovered and loaded.
-    consume_modules(__package__)
+    REGISTRY = default_registry.export(__package__)
 
-    REGISTRY = default_registry
+The :meth:`~.PageObjectRegistry.export` method returns a new instance of
+:class:`~.PageObjectRegistry` which contains only the :class:`~.OverrideRule`
+from the given package. This means that if there are other **POPs** using the
+recommended ``default_registry``, any :class:`~.OverrideRule` that are not part
+of the packages are not included.
 
 This allows any developer using a **POP** to easily access all of the
 :class:`~.OverrideRule` using the convention of accessing it via the
@@ -173,8 +177,35 @@ This allows any developer using a **POP** to easily access all of the
 
     However, it is **recommended** to use the instances of
     :class:`~.PageObjectRegistry` to leverage the validation logic for its
-    contents.
+    contents, as well as its other functionalities.
+
+Lastly, when trying to repackage multiple **POPs** into a single unifying **POP**
+which contains all of the :class:`~.OverrideRule`, it can easily be packaged
+as:
+
+.. code-block:: python
+
+    from web_poet import PageObjectRegistry
+
+    import base_A_package
+    import base_B_package
+
+    # If on Python 3.9+
+    combined_rules = base_A_package.REGISTRY | base_B_package.REGISTRY
+
+    # If on lower Python versions
+    combined_rules = {**base_A_package.REGISTRY, **base_B_package.REGISTRY}
+
+    REGISTRY = PageObjectRegistry(combined_rules)
+
+Note that you can also opt to use only a subset of the :class:`~.OverrideRule`
+by selecting the specific ones in ``combined_rules`` before creating a new
+:class:`~.PageObjectRegistry` instance. An **inclusion** rule is preferred than
+an **exclusion** rule (see **Tip #4** in the :ref:`conventions-and-best-practices`).
+You can use :meth:`~.PageObjectRegistry.search_overrides` when selecting the
+rules.
 
+.. _`conventions-and-best-practices`:
 
 Conventions and Best Practices
 ------------------------------

tests/test_pop.py

@@ -0,0 +1,60 @@
+"""This tests the expected behavior of importing multiple different POPs such
+that they don't leak the OverrideRules from other POPs that use the same
+``default_registry``.
+
+Packacking and exporting a given POP should be resilient from such cases.
+
+In particular, this tests the :meth:`PageObjectRegistry.export` functionality.
+"""
+
+def test_base_A():
+    from tests_pop import base_A_package
+
+    reg = base_A_package.REGISTRY
+
+    assert len(reg) == 2
+    assert base_A_package.site_1.A_Site1 in reg
+    assert base_A_package.site_2.A_Site2 in reg
+
+
+def test_base_B():
+    from tests_pop import base_B_package
+
+    reg = base_B_package.REGISTRY
+
+    assert len(reg) == 2
+    assert base_B_package.site_2.B_Site2 in reg
+    assert base_B_package.site_3.B_Site3 in reg
+
+
+def test_improved_A():
+    from tests_pop import improved_A_package, base_A_package
+
+    reg = improved_A_package.REGISTRY
+
+    assert len(reg) == 3
+    assert improved_A_package.site_1.A_Improved_Site1 in reg
+    assert improved_A_package.base_A_package.site_1.A_Site1 in reg
+    assert improved_A_package.base_A_package.site_2.A_Site2 in reg
+
+
+def test_combine_A_B():
+    from tests_pop import combine_A_B_package, base_A_package, base_B_package
+
+    reg = combine_A_B_package.REGISTRY
+
+    assert len(reg) == 4
+    assert combine_A_B_package.base_A_package.site_1.A_Site1 in reg
+    assert combine_A_B_package.base_A_package.site_2.A_Site2 in reg
+    assert combine_A_B_package.base_B_package.site_2.B_Site2 in reg
+    assert combine_A_B_package.base_B_package.site_3.B_Site3 in reg
+
+
+def test_combine_A_B_subset():
+    from tests_pop import combine_A_B_subset_package, improved_A_package, base_B_package
+
+    reg = combine_A_B_subset_package.REGISTRY
+
+    assert len(reg) == 2
+    assert combine_A_B_subset_package.improved_A_package.site_1.A_Improved_Site1 in reg
+    assert combine_A_B_subset_package.base_B_package.site_3.B_Site3 in reg

tests_pop/__init__.py

@@ -0,0 +1,3 @@
+from web_poet import default_registry
+
+REGISTRY = default_registry.export(__package__)

tests_pop/base_A_package/__init__.py

@@ -0,0 +1,2 @@
+class BasePage:
+    ...

tests_pop/base_A_package/base.py

@@ -0,0 +1,8 @@
+from web_poet import handle_urls
+
+from .base import BasePage
+
+
+@handle_urls("site_1.com", overrides=BasePage)
+class A_Site1:
+    ...

tests_pop/base_A_package/site_1.py

@@ -0,0 +1,8 @@
+from web_poet import handle_urls
+
+from .base import BasePage
+
+
+@handle_urls("site_2.com", overrides=BasePage)
+class A_Site2:
+    ...

tests_pop/base_A_package/site_2.py

@@ -0,0 +1,3 @@
+from web_poet import default_registry
+
+REGISTRY = default_registry.export(__package__)

tests_pop/base_B_package/__init__.py

@@ -0,0 +1,2 @@
+class BasePage:
+    ...

tests_pop/base_B_package/base.py

@@ -0,0 +1,8 @@
+from web_poet import handle_urls
+
+from .base import BasePage
+
+
+@handle_urls("site_2.com", overrides=BasePage)
+class B_Site2:
+    ...

tests_pop/base_B_package/site_2.py

@@ -0,0 +1,8 @@
+from web_poet import handle_urls
+
+from .base import BasePage
+
+
+@handle_urls("site_3.com", overrides=BasePage)
+class B_Site3:
+    ...

tests_pop/base_B_package/site_3.py

@@ -0,0 +1,9 @@
+"""This POP simply wants to repackage POP "A" and "B" into one unifying package."""
+
+from web_poet import PageObjectRegistry
+
+from . import base_A_package
+from . import base_B_package
+
+combined = {**base_A_package.REGISTRY, **base_B_package.REGISTRY}
+REGISTRY = PageObjectRegistry(combined)

tests_pop/combine_A_B_package/__init__.py

@@ -0,0 +1 @@
+../base_A_package

tests_pop/combine_A_B_package/base_A_package

@@ -0,0 +1 @@
+../base_B_package

tests_pop/combine_A_B_package/base_B_package

@@ -0,0 +1,16 @@
+"""This POP simply wants to repackage POP "A" and "B" into one unifying package."""
+
+from web_poet import PageObjectRegistry
+
+from . import improved_A_package
+from . import base_B_package
+
+rules_A_improved = improved_A_package.REGISTRY.search_overrides(
+    use=improved_A_package.site_1.A_Improved_Site1  # type:ignore
+)
+rules_B = base_B_package.REGISTRY.search_overrides(
+    use=base_B_package.site_3.B_Site3  # type: ignore
+)
+
+combined_rules = rules_A_improved + rules_B
+REGISTRY = PageObjectRegistry.from_override_rules(combined_rules)

tests_pop/combine_A_B_subset_package/__init__.py

@@ -0,0 +1 @@
+../base_B_package

tests_pop/combine_A_B_subset_package/base_B_package

@@ -0,0 +1 @@
+../improved_A_package

tests_pop/combine_A_B_subset_package/improved_A_package

@@ -0,0 +1,3 @@
+from web_poet import default_registry
+
+REGISTRY = default_registry.export(__package__)

tests_pop/improved_A_package/__init__.py

@@ -0,0 +1 @@
+../base_A_package

tests_pop/improved_A_package/base_A_package

@@ -0,0 +1,9 @@
+from web_poet import handle_urls
+
+from .base_A_package.base import BasePage
+from .base_A_package.site_1 import A_Site1
+
+
+@handle_urls("site_1.com", overrides=BasePage)
+class A_Improved_Site1(A_Site1):
+    ...  # some improvements here after subclassing the original one.

tests_pop/improved_A_package/site_1.py

@@ -4,6 +4,7 @@
 import importlib.util
 import warnings
 import pkgutil
+from copy import deepcopy
 from collections import deque
 from dataclasses import dataclass, field
 from operator import attrgetter
@@ -121,6 +122,27 @@ def from_override_rules(
         """
         return cls({rule.use: rule for rule in rules})
 
+    def export(self, package: str) -> PageObjectRegistry:
+        """Returns a new :class:`~.PageObjectRegistry` instance containing
+        :class:`~.OverrideRule` which are only found in the provided **package**.
+
+        This is used in cases wherein all of the :class:`~.OverrideRule` in a
+        given **Page Object Project (POP)** should be placed inside a dedicated
+        registry for packaging. See :ref:`POP Recommended Requirements
+        <pop-recommended-requirements>` for more info about this.
+
+        Note that the :func:`~.consume_modules` will be called on the said
+        **package** which adds any undiscovered :class:`~.OverrideRule` to the
+        original :class:`~.PageObjectRegistry` instance. There's no need to worry
+        about unrelated rules from being added since it wouldn't happen if the
+        given registry's ``@handle_urls`` annotation wasn't the one used.
+        """
+        backup_state = deepcopy(self)
+        self.clear()
+        rules = self.get_overrides(consume=package)
+        self = backup_state
+        return self.from_override_rules(rules)
+
     def handle_urls(
         self,
         include: Strings,