ci: temporarily exclude sqlalchemy-bigquery from package shards

Author: chalmerlowe

ci/get_package_shards.py

@@ -48,6 +48,16 @@
     "google-crc32c",
 }
 
+# Packages temporarily excluded from CI test execution.
+# NOTE: 'sqlalchemy-bigquery' is temporarily excluded to allow testing in this PR
+# to complete due to an upstream packaging issue in sqlalchemy (duplicate normalized
+# extra name 'mssql-pymssql' under strict uv PEP 621 parsing in sqlalchemy==2.1.0rc2,
+# pulled via global UV_PRERELEASE=allow). Awaiting team feedback on a long-term
+# solution (e.g. package migration out of the monorepo or adjusting workflow settings).
+EXCLUDED_PACKAGES = {
+    "sqlalchemy-bigquery",
+}
+
 
 def get_package_directories():
     """Parses package directory roots from the PACKAGE_DIRS environment variable.
@@ -56,7 +66,7 @@ def get_package_directories():
     """
     env_dirs = os.environ.get("PACKAGE_DIRS", "")
     if env_dirs:
-        dirs = [d.strip() for d in env_dirs.replace('\n', ' ').split(' ') if d.strip()]
+        dirs = [d.strip() for d in env_dirs.replace("\n", " ").split(" ") if d.strip()]
         if dirs:
             return dirs
     return ["packages", "preview-packages"]
@@ -103,7 +113,9 @@ def get_packages(handwritten_only=False):
         if not os.path.exists(subdir):
             continue
         for d in os.listdir(subdir):
-            full_path = os.path.join(subdir, d) + '/'
+            if d in EXCLUDED_PACKAGES:
+                continue
+            full_path = os.path.join(subdir, d) + "/"
             if not os.path.isdir(full_path):
                 continue
             if handwritten_only:
@@ -112,7 +124,10 @@ def get_packages(handwritten_only=False):
                     try:
                         with open(meta_file) as f:
                             data = json.load(f)
-                            if isinstance(data, dict) and data.get("library_type") == "GAPIC_AUTO":
+                            if (
+                                isinstance(data, dict)
+                                and data.get("library_type") == "GAPIC_AUTO"
+                            ):
                                 continue
                     except Exception:
                         pass
@@ -130,24 +145,26 @@ def get_packages_to_test():
     Returns:
         dict: A dictionary mapping package_name -> list of relative directory paths to be tested.
     """
-    build_type = os.environ.get('BUILD_TYPE', 'presubmit')
-    target_branch = os.environ.get('TARGET_BRANCH', 'main')
-    test_all_packages = os.environ.get('TEST_ALL_PACKAGES', 'false').lower() == 'true'
+    build_type = os.environ.get("BUILD_TYPE", "presubmit")
+    target_branch = os.environ.get("TARGET_BRANCH", "main")
+    test_all_packages = os.environ.get("TEST_ALL_PACKAGES", "false").lower() == "true"
 
     all_packages = get_packages()
 
     if test_all_packages:
         return all_packages
 
-    if build_type == 'presubmit':
+    if build_type == "presubmit":
         git_diff_arg = f"origin/{target_branch}..."
-    elif build_type == 'continuous':
+    elif build_type == "continuous":
         git_diff_arg = "HEAD~1.."
     else:
         return all_packages
 
     try:
-        res = subprocess.check_output(['git', 'diff', '--name-only', git_diff_arg]).decode('utf-8')
+        res = subprocess.check_output(
+            ["git", "diff", "--name-only", git_diff_arg]
+        ).decode("utf-8")
         changed_files = res.splitlines()
     except subprocess.CalledProcessError:
         # If change detection fails, fall back to all packages
@@ -220,7 +237,11 @@ def group_packages(packages_map):
     for name, paths, weight in pkg_items:
         # If adding this package would exceed target weight AND we haven't reached the
         # shard limit, start a new shard. Otherwise, keep "stuffing" the current one.
-        if current_shard_items and (current_shard_weight + weight > target_weight) and len(shards_list) < max_shards - 1:
+        if (
+            current_shard_items
+            and (current_shard_weight + weight > target_weight)
+            and len(shards_list) < max_shards - 1
+        ):
             shards_list.append(current_shard_items)
             current_shard_items = [(name, paths, weight)]
             current_shard_weight = weight
@@ -250,13 +271,15 @@ def group_packages(packages_map):
         for _, paths, _ in shard_items:
             all_paths.extend(paths)
 
-        shards.append({
-            "name": name,
-            "index": index,
-            "description": desc,
-            "packages": " ".join(all_paths),
-            "is_sharded": True
-        })
+        shards.append(
+            {
+                "name": name,
+                "index": index,
+                "description": desc,
+                "packages": " ".join(all_paths),
+                "is_sharded": True,
+            }
+        )
 
     # Set is_sharded dynamically based on the total number of shards
     total_shards = len(shards)