feat(pypi): generate filegroup with all extracted wheel files

CHANGELOG.md

@@ -70,6 +70,10 @@ END_UNRELEASED_TEMPLATE
 * (pypi) To configure the environment for `requirements.txt` evaluation, use the newly added
   developer preview of the `pip.default` tag class. Only `rules_python` and root modules can use
   this feature. You can also configure custom `config_settings` using `pip.default`.
+* (pypi) PyPI dependencies now expose an `:extracted_whl_files` filegroup target
+  of all the files extracted from the wheel. This can be used in lieu of
+  {obj}`whl_filegroup` to avoid copying/extracting wheel multiple times to
+  get a subset of their files.
 
 {#v0-0-0-removed}
 ### Removed

docs/pypi/use.md

@@ -40,9 +40,16 @@ Note that the hub repo contains the following targets for each package:
 * `@pypi//numpy:data` - the {obj}`filegroup` for all of the extra files that are included
   as data in the `pkg` target.
 * `@pypi//numpy:dist_info` - the {obj}`filegroup` for all of the files in the `<pkg prefix with version>.distinfo` directory.
+* `@pypi//numpy:extracted_whl_files` - a {obj}`filegroup` of all the files
+  extracted from the whl file.
 * `@pypi//numpy:whl` - the {obj}`filegroup` that is the `.whl` file itself, which includes all
   transitive dependencies via the {attr}`filegroup.data` attribute.
 
+:::{versionadded} VERSION_NEXT_FEATURE
+
+The `:extracted_whl_files` target was added
+:::
+
 ## Entry points
 
 If you would like to access [entry points][whl_ep], see the `py_console_script_binary` rule documentation,

python/private/pypi/labels.bzl

@@ -14,6 +14,7 @@
 
 """Constants used by parts of pip_repository for naming libraries and wheels."""
 
+EXTRACTED_WHEEL_FILES = "extracted_whl_files"
 WHEEL_FILE_PUBLIC_LABEL = "whl"
 WHEEL_FILE_IMPL_LABEL = "_whl"
 PY_LIBRARY_PUBLIC_LABEL = "pkg"

python/private/pypi/pkg_aliases.bzl

@@ -79,6 +79,7 @@ load(
     ":labels.bzl",
     "DATA_LABEL",
     "DIST_INFO_LABEL",
+    "EXTRACTED_WHEEL_FILES",
     "PY_LIBRARY_IMPL_LABEL",
     "PY_LIBRARY_PUBLIC_LABEL",
     "WHEEL_FILE_IMPL_LABEL",
@@ -151,6 +152,7 @@ def pkg_aliases(
         WHEEL_FILE_PUBLIC_LABEL: WHEEL_FILE_IMPL_LABEL if group_name else WHEEL_FILE_PUBLIC_LABEL,
         DATA_LABEL: DATA_LABEL,
         DIST_INFO_LABEL: DIST_INFO_LABEL,
+        EXTRACTED_WHEEL_FILES: EXTRACTED_WHEEL_FILES,
     } | {
         x: x
         for x in extra_aliases or []

python/private/pypi/whl_library.bzl

@@ -248,6 +248,7 @@ def _whl_library_impl(rctx):
     environment = _create_repository_execution_environment(rctx, python_interpreter, logger = logger)
 
     whl_path = None
+    sdist_filename = None
     if rctx.attr.whl_file:
         whl_path = rctx.path(rctx.attr.whl_file)
 
@@ -276,6 +277,8 @@ def _whl_library_impl(rctx):
         if filename.endswith(".whl"):
             whl_path = rctx.path(filename)
         else:
+            sdist_filename = filename
+
             # It is an sdist and we need to tell PyPI to use a file in this directory
             # and, allow getting build dependencies from PYTHONPATH, which we
             # setup in this repository rule, but still download any necessary
@@ -381,6 +384,7 @@ def _whl_library_impl(rctx):
 
         build_file_contents = generate_whl_library_build_bazel(
             name = whl_path.basename,
+            sdist_filename = sdist_filename,
             dep_template = rctx.attr.dep_template or "@{}{{name}}//:{{target}}".format(rctx.attr.repo_prefix),
             entry_points = entry_points,
             metadata_name = metadata.name,
@@ -454,6 +458,7 @@ def _whl_library_impl(rctx):
 
         build_file_contents = generate_whl_library_build_bazel(
             name = whl_path.basename,
+            sdist_filename = sdist_filename,
             dep_template = rctx.attr.dep_template or "@{}{{name}}//:{{target}}".format(rctx.attr.repo_prefix),
             entry_points = entry_points,
             # TODO @aignas 2025-05-17: maybe have a build flag for this instead

python/private/pypi/whl_library_targets.bzl

@@ -24,6 +24,7 @@ load(
     ":labels.bzl",
     "DATA_LABEL",
     "DIST_INFO_LABEL",
+    "EXTRACTED_WHEEL_FILES",
     "PY_LIBRARY_IMPL_LABEL",
     "PY_LIBRARY_PUBLIC_LABEL",
     "WHEEL_ENTRY_POINT_PREFIX",
@@ -33,6 +34,16 @@ load(
 load(":namespace_pkgs.bzl", "create_inits")
 load(":pep508_deps.bzl", "deps")
 
+# Files that are special to the Bazel processing of things.
+_BAZEL_REPO_FILE_GLOBS = [
+    "BUILD",
+    "BUILD.bazel",
+    "REPO.bazel",
+    "WORKSPACE",
+    "WORKSPACE",
+    "WORKSPACE.bazel",
+]
+
 def whl_library_targets_from_requires(
         *,
         name,
@@ -100,11 +111,8 @@ def whl_library_targets(
         data_exclude = [],
         srcs_exclude = [],
         tags = [],
-        filegroups = {
-            DIST_INFO_LABEL: ["site-packages/*.dist-info/**"],
-            DATA_LABEL: ["data/**"],
-        },
         dependencies = [],
+        filegroups = None,
         dependencies_by_platform = {},
         dependencies_with_markers = {},
         group_deps = [],
@@ -126,6 +134,8 @@ def whl_library_targets(
     Args:
         name: {type}`str` The file to match for including it into the `whl`
             filegroup. This may be also parsed to generate extra metadata.
+        sdist_filename: {type}`str | None` If the wheel was built from an sdist,
+            the filename of the sdist.
         dep_template: {type}`str` The dep_template to use for dependency
             interpolation.
         tags: {type}`list[str]` The tags set on the `py_library`.
@@ -134,8 +144,8 @@ def whl_library_targets(
             dependencies by platform key.
         dependencies_with_markers: {type}`dict[str, str]` A marker to evaluate
             in order for the dep to be included.
-        filegroups: {type}`dict[str, list[str]]` A dictionary of the target
-            names and the glob matches.
+        filegroups: {type}`dict[str, list[str]] | None` A dictionary of the target
+            names and the glob matches. If `None`, defaults will be used.
         group_name: {type}`str` name of the dependency group (if any) which
             contains this library. If set, this library will behave as a shim
             to group implementation rules which will provide simultaneously
@@ -168,10 +178,28 @@ def whl_library_targets(
     tags = sorted(tags)
     data = [] + data
 
-    for filegroup_name, glob in filegroups.items():
+    if filegroups == None:
+        filegroups = {
+            EXTRACTED_WHEEL_FILES: dict(
+                include = ["**"],
+                exclude = (
+                    _BAZEL_REPO_FILE_GLOBS +
+                    [sdist_filename] if sdist_filename else []
+                ),
+            ),
+            DIST_INFO_LABEL: dict(
+                include = ["site-packages/*.dist-info/**"],
+            ),
+            DATA_LABEL: dict(
+                include = ["data/**"],
+            ),
+        }
+
+    for filegroup_name, glob_kwargs in filegroups.items():
+        glob_kwargs = {"allow_empty": True} | glob_kwargs
         native.filegroup(
             name = filegroup_name,
-            srcs = native.glob(glob, allow_empty = True),
+            srcs = native.glob(**glob_kwargs),
             visibility = ["//visibility:public"],
         )
 

python/private/whl_filegroup/whl_filegroup.bzl

@@ -42,7 +42,14 @@ cc_library(
     includes = ["numpy_includes/numpy/core/include"],
     deps = ["@rules_python//python/cc:current_py_cc_headers"],
 )
+
 ```
+
+:::{seealso}
+
+The `:extracted_whl_files` target, which is a filegroup of all the files
+from the already extracted whl file.
+:::
 """,
     attrs = {
         "pattern": attr.string(default = "", doc = "Only file paths matching this regex pattern will be extracted."),