Download inventories only once

.gitignore

@@ -87,9 +87,9 @@ instance/
 
 # Sphinx documentation
 docs/_build/
-docs/_api/
+docs/_doctrees/
+docs/_inventory_cache/
 docs/*/_api/
-docs/_doctrees
 
 # PyBuilder
 target/

docs/build_docs.py

@@ -33,6 +33,7 @@
     DocBuildError,
     display_errors_summary,
 )
+from docs.exts.docs_build.fetch_inventories import fetch_inventories
 from docs.exts.docs_build.github_action_utils import with_group  # pylint: disable=no-name-in-module
 from docs.exts.docs_build.spelling_checks import (  # pylint: disable=no-name-in-module
     SpellingError,
@@ -83,7 +84,7 @@ def _get_parser():
         '--for-production',
         dest='for_production',
         action='store_true',
-        help=('Builds documentation for official release i.e. all links point to stable version'),
+        help='Builds documentation for official release i.e. all links point to stable version',
     )
 
     return parser
@@ -173,6 +174,8 @@ def main():
     with with_group(f"Documentation will be built for {len(current_packages)} package(s)"):
         for pkg in current_packages:
             print(f" - {pkg}")
+    with with_group("Fetching inventories"):
+        fetch_inventories()
 
     all_build_errors: Dict[Optional[str], List[DocBuildError]] = {}
     all_spelling_errors: Dict[Optional[str], List[SpellingError]] = {}

docs/exts/airflow_intersphinx.py

@@ -26,7 +26,6 @@
 ROOT_DIR = os.path.abspath(os.path.join(CURRENT_DIR, os.pardir, os.pardir))
 DOCS_DIR = os.path.join(ROOT_DIR, 'docs')
 DOCS_PROVIDER_DIR = os.path.join(ROOT_DIR, 'docs')
-S3_DOC_URL = "http://apache-airflow-docs.s3-website.eu-central-1.amazonaws.com"
 
 
 def _create_init_py(app, config):
@@ -51,54 +50,32 @@ def _generate_provider_intersphinx_mapping():
             continue
 
         provider_base_url = f'/docs/{package_name}/{current_version}/'
+        doc_inventory = f'{DOCS_DIR}/_build/docs/{package_name}/{current_version}/objects.inv'
+        cache_inventory = f'{DOCS_DIR}/_inventory_cache/{package_name}/objects.inv'
 
         airflow_mapping[package_name] = (
             # base URI
             provider_base_url,
-            # Index locations list
-            # If passed None, this will try to fetch the index from `[base_url]/objects.inv`
-            # If we pass a path containing `://` then we will try to index from the given address.
-            # Otherwise, it will try to read the local file
-            #
-            # In this case, the local index will be read. If unsuccessful, the remote index
-            # will be fetched.
-            (
-                f'{DOCS_DIR}/_build/docs/{package_name}/{current_version}/objects.inv',
-                f'{S3_DOC_URL}/docs/{package_name}/latest/objects.inv',
-            ),
+            (doc_inventory if os.path.exists(doc_inventory) else cache_inventory,),
         )
     if os.environ.get('AIRFLOW_PACKAGE_NAME') != 'apache-airflow':
+        doc_inventory = f'{DOCS_DIR}/_build/docs/apache-airflow/{current_version}/objects.inv'
+        cache_inventory = f'{DOCS_DIR}/_inventory_cache/apache-airflow/objects.inv'
+
         airflow_mapping['apache-airflow'] = (
             # base URI
             f'/docs/apache-airflow/{current_version}/',
-            # Index locations list
-            # If passed None, this will try to fetch the index from `[base_url]/objects.inv`
-            # If we pass a path containing `://` then we will try to index from the given address.
-            # Otherwise, it will try to read the local file
-            #
-            # In this case, the local index will be read. If unsuccessful, the remote index
-            # will be fetched.
-            (
-                f'{DOCS_DIR}/_build/docs/apache-airflow/{current_version}/objects.inv',
-                f'{S3_DOC_URL}/docs/apache-airflow/latest/objects.inv',
-            ),
+            (doc_inventory if os.path.exists(doc_inventory) else cache_inventory,),
         )
 
     if os.environ.get('AIRFLOW_PACKAGE_NAME') != 'apache-airflow-providers':
+        doc_inventory = f'{DOCS_DIR}/_build/docs/apache-airflow-providers/objects.inv'
+        cache_inventory = f'{DOCS_DIR}/_inventory_cache/apache-airflow-providers/objects.inv'
+
         airflow_mapping['apache-airflow-providers'] = (
             # base URI
             '/docs/apache-airflow-providers/',
-            # Index locations list
-            # If passed None, this will try to fetch the index from `[base_url]/objects.inv`
-            # If we pass a path containing `://` then we will try to index from the given address.
-            # Otherwise, it will try to read the local file
-            #
-            # In this case, the local index will be read. If unsuccessful, the remote index
-            # will be fetched.
-            (
-                f'{DOCS_DIR}/_build/docs/apache-airflow-providers/objects.inv',
-                f'{S3_DOC_URL}/docs/apache-airflow-providers/objects.inv',
-            ),
+            (doc_inventory if os.path.exists(doc_inventory) else cache_inventory,),
         )
 
     return airflow_mapping

docs/exts/docs_build/docs_builder.py

@@ -200,7 +200,12 @@ def publish(self):
         print()
 
 
+def get_available_providers_packages():
+    """Get list of all available providers packages to build."""
+    return [provider['package-name'] for provider in ALL_PROVIDER_YAMLS]
+
+
 def get_available_packages():
     """Get list of all available packages to build."""
-    provider_package_names = [provider['package-name'] for provider in ALL_PROVIDER_YAMLS]
+    provider_package_names = get_available_providers_packages()
     return ["apache-airflow", *provider_package_names, "apache-airflow-providers"]

docs/exts/docs_build/fetch_inventories.py

@@ -0,0 +1,91 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+import concurrent
+import concurrent.futures
+import datetime
+import os
+import shutil
+
+import requests
+from requests.adapters import DEFAULT_POOLSIZE
+
+from docs.exts.docs_build.docs_builder import get_available_providers_packages
+
+CURRENT_DIR = os.path.dirname(__file__)
+ROOT_DIR = os.path.abspath(os.path.join(CURRENT_DIR, os.pardir, os.pardir, os.pardir))
+DOCS_DIR = os.path.join(ROOT_DIR, 'docs')
+CACHE_DIR = os.path.join(DOCS_DIR, '_inventory_cache')
+EXPIRATION_DATE_PATH = os.path.join(DOCS_DIR, '_inventory_cache', "expiration-date")
+
+S3_DOC_URL = "http://apache-airflow-docs.s3-website.eu-central-1.amazonaws.com"
+S3_DOC_URL_VERSIONED = S3_DOC_URL + "/docs/{package_name}/latest/objects.inv"
+S3_DOC_URL_NON_VERSIONED = S3_DOC_URL + "/docs/{package_name}/objects.inv"
+
+
+def _fetch_file(session: requests.Session, url: str, path: str):
+
+    response = session.get(url, allow_redirects=True, stream=True)
+    if not response.ok:
+        print(f"Failed to fetch inventory: {url}")
+        return
+
+    os.makedirs(os.path.dirname(path), exist_ok=True)
+    with open(path, 'wb') as f:
+        response.raw.decode_content = True
+        shutil.copyfileobj(response.raw, f)
+    print(f"Fetched inventory: {url}")
+
+
+def _is_outdated(path: str):
+    delta = datetime.datetime.now() - datetime.datetime.fromtimestamp(os.path.getmtime(path))
+    return delta < datetime.timedelta(hours=12)
+
+
+def fetch_inventories():
+    """Fetch all inventories for Airflow documentatio packages and store in cache."""
+    os.makedirs(os.path.dirname(CACHE_DIR), exist_ok=True)
+    to_download = []
+
+    for pkg_name in get_available_providers_packages():
+        to_download.append(
+            (
+                S3_DOC_URL_VERSIONED.format(package_name=pkg_name),
+                f'{CACHE_DIR}/{pkg_name}/objects.inv',
+            )
+        )
+    to_download.append(
+        (
+            S3_DOC_URL_VERSIONED.format(package_name='apache-airflow'),
+            f'{CACHE_DIR}/apache-airflow/objects.inv',
+        )
+    )
+    to_download.append(
+        (
+            S3_DOC_URL_NON_VERSIONED.format(package_name='apache-airflow-providers'),
+            f'{CACHE_DIR}/apache-airflow-providers/objects.inv',
+        )
+    )
+    to_download = [
+        (url, path) for url, path in to_download if not (os.path.isfile(path) and _is_outdated(path))
+    ]
+    print(f"To download {len(to_download)} inventorie(s)")
+    if not to_download:
+        return
+    with requests.Session() as session, concurrent.futures.ThreadPoolExecutor(DEFAULT_POOLSIZE) as pool:
+        for url, path in to_download:
+            pool.submit(_fetch_file, session=session, url=url, path=path)