Enhance generator to update changelog only if generated code differs from existing

CHANGELOG.md

@@ -13,6 +13,7 @@ Inspired from [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 - Added GHA release ([#614](https://github.com/opensearch-project/opensearch-py/pull/614))
 - Incorporated API generation into CI workflow and fixed 'generate' nox session ([#660](https://github.com/opensearch-project/opensearch-py/pull/660))
 - Added an automated api update bot for opensearch-py ([#664](https://github.com/opensearch-project/opensearch-py/pull/664)) 
+- Enhance generator to update changelog only if generated code differs from existing ([#684](https://github.com/opensearch-project/opensearch-py/pull/684))
 ### Changed
 - Updated the `get_policy` API in the index_management plugin to allow the policy_id argument as optional ([#633](https://github.com/opensearch-project/opensearch-py/pull/633))
 - Updated the `point_in_time.md` guide with examples demonstrating the usage of the new APIs as alternatives to the deprecated ones. ([#661](https://github.com/opensearch-project/opensearch-py/pull/661))

noxfile.py

@@ -146,4 +146,5 @@ def generate(session: Any) -> None:
     """
     session.install("-rdev-requirements.txt")
     session.run("python", "utils/generate_api.py")
-    session.notify("format")
+    session.run("nox", "-s", "format", external=True)
+    session.run("python", "utils/changelog_updater.py")

utils/changelog_updater.py

@@ -0,0 +1,100 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# The OpenSearch Contributors require contributions made to
+# this file be licensed under the Apache-2.0 license or a
+# compatible open source license.
+#
+# Modifications Copyright OpenSearch Contributors. See
+# GitHub history for details.
+
+import filecmp
+import os
+import shutil
+
+import requests
+
+
+def main() -> None:
+    """
+    Update CHANGELOG.md when API generator produces new code differing from existing.
+    """
+    root_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+    after_paths = [
+        os.path.join(root_dir, f"opensearchpy/{folder}")
+        for folder in ["client", "_async/client"]
+    ]
+
+    before_paths = [
+        os.path.join(root_dir, f"before_generate/{folder}")
+        for folder in ["client", "async_client"]
+    ]
+
+    # Compare only .py files and take their union for client and async_client directories
+    before_files_client = set(
+        file for file in os.listdir(before_paths[0]) if file.endswith(".py")
+    )
+    after_files_client = set(
+        file for file in os.listdir(after_paths[0]) if file.endswith(".py")
+    )
+
+    before_files_async_client = set(
+        file for file in os.listdir(before_paths[1]) if file.endswith(".py")
+    )
+    after_files_async_client = set(
+        file for file in os.listdir(after_paths[1]) if file.endswith(".py")
+    )
+
+    all_files_union_client = before_files_client.union(after_files_client)
+    all_files_union_async_client = before_files_async_client.union(
+        after_files_async_client
+    )
+
+    # Compare files and check for mismatches or errors for client and async_client directories
+    mismatch_client, errors_client = filecmp.cmpfiles(
+        before_paths[0], after_paths[0], all_files_union_client, shallow=True
+    )[1:]
+    mismatch_async_client, errors_async_client = filecmp.cmpfiles(
+        before_paths[1], after_paths[1], all_files_union_async_client, shallow=True
+    )[1:]
+
+    if mismatch_client or errors_client or mismatch_async_client or errors_async_client:
+        print("Changes detected")
+        response = requests.get(
+            "https://api.github.com/repos/opensearch-project/opensearch-api-specification/commits"
+        )
+        if response.ok:
+            commit_info = response.json()[0]
+            commit_url = commit_info["html_url"]
+            latest_commit_sha = commit_info.get("sha")
+        else:
+            raise Exception(
+                f"Failed to fetch opensearch-api-specification commit information. Status code: {response.status_code}"
+            )
+
+        with open("CHANGELOG.md", "r+", encoding="utf-8") as file:
+            content = file.read()
+            if commit_url not in content:
+                if "### Updated APIs" in content:
+                    file_content = content.replace(
+                        "### Updated APIs",
+                        f"### Updated APIs\n- Updated opensearch-py APIs to reflect [opensearch-api-specification@{latest_commit_sha[:7]}]({commit_url})",
+                        1,
+                    )
+                    file.seek(0)
+                    file.write(file_content)
+                    file.truncate()
+                else:
+                    raise Exception(
+                        "'Updated APIs' section is not present in CHANGELOG.md"
+                    )
+    else:
+        print("No changes detected")
+
+    # Clean up
+    for path in before_paths:
+        shutil.rmtree(path)
+
+
+if __name__ == "__main__":
+    main()

utils/generate_api.py

@@ -33,6 +33,7 @@
 import json
 import os
 import re
+import shutil
 from functools import lru_cache
 from itertools import chain, groupby
 from operator import itemgetter
@@ -764,34 +765,22 @@ def dump_modules(modules: Any) -> None:
     unasync.unasync_files(filepaths, rules)
     blacken(CODE_ROOT / "opensearchpy")
 
-    # Updating the CHANGELOG.md
-    response = requests.get(
-        "https://api.github.com/repos/opensearch-project/opensearch-api-specification/commits"
-    )
-    if response.ok:
-        commit_info = response.json()[0]
-        commit_url = commit_info["html_url"]
-        latest_commit_sha = commit_info.get("sha")
-    else:
-        raise Exception(
-            f"Failed to fetch opensearch-api-specification commit information. Status code: {response.status_code}"
-        )
 
-    with open("CHANGELOG.md", "r+", encoding="utf-8") as file:
-        content = file.read()
-        if commit_url not in content:
-            if "### Updated APIs" in content:
-                file_content = content.replace(
-                    "### Updated APIs",
-                    f"### Updated APIs\n- Updated opensearch-py APIs to reflect [opensearch-api-specification@{latest_commit_sha[:7]}]({commit_url})",
-                    1,
-                )
-                file.seek(0)
-                file.write(file_content)
-                file.truncate()
-            else:
-                raise Exception("'Updated APIs' section is not present in CHANGELOG.md")
+if __name__ == "__main__":
+    # Store directories for comparison pre-generation vs post-generation.
+    root_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+    before_paths = [
+        os.path.join(root_dir, f"before_generate/{folder}")
+        for folder in ["client", "async_client"]
+    ]
 
+    for path in before_paths:
+        if os.path.exists(path):
+            shutil.rmtree(path)
+
+    shutil.copytree(os.path.join(root_dir, "opensearchpy/client"), before_paths[0])
+    shutil.copytree(
+        os.path.join(root_dir, "opensearchpy/_async/client"), before_paths[1]
+    )
 
-if __name__ == "__main__":
     dump_modules(read_modules())