CI: modify release-notes.py to auto-categorize pull requests based on…

… labels (#4850)

## Which problem is this PR solving?
- fixes: #4799

## Description of the changes
- The modifications in release-notes.py would now make sure that the
final results printed to the console are in a categorized manner which
could be just copy-pasted to release-notes.
- The `categories` object could be modified to accommodate more
categories along with some meaningful titles like `Changes related to
CI: changelog:ci`

## How was this change tested?
- Locally, on the jaegertracing/jaeger repo.

![image](https://github.com/jaegertracing/jaeger/assets/94157520/ba1cb4a8-dcf9-4b25-bad4-ba812d7772fa)


## Checklist
- [x] I have read
https://github.com/jaegertracing/jaeger/blob/master/CONTRIBUTING_GUIDELINES.md
- [x] I have signed all commits
- [x] I have added unit tests for the new functionality
- [x] I have run lint and test steps successfully
  - for `jaeger`: `make lint test`
  - for `jaeger-ui`: `yarn lint` and `yarn test`

---------

Signed-off-by: Ansh Goyal <anshgoyal1704@gmail.com>

scripts/release-notes.py

@@ -43,6 +43,23 @@ def num_commits_since_prev_tag(token, base_url):
     print(f"There are {num_commits} new commits since {prev_release_tag}")
     return num_commits
 
+UNCATTEGORIZED = 'Uncategorized'
+categories = [
+    {'title': '#### ⛔ Breaking Changes', 'label': 'changelog:breaking-change'},
+    {'title': '#### ✨ New Features', 'label': 'changelog:new-feature'},
+    {'title': '#### 🐞 Bug fixes, Minor Improvements', 'label': 'changelog:bugfix-or-minor-feature'},
+    {'title': '#### 🚧 Experimental Features', 'label': 'changelog:exprimental'},
+    {'title': '#### 👷 CI Improvements', 'label': 'changelog:ci'},
+    {'title': None, 'label': 'changelog:test'},
+    {'title': None, 'label': 'changelog:skip'}
+]
+
+def categorize_pull_request(label):
+    for category, prefix in categories.items():
+        if label.startswith(prefix):
+            return category
+    return UNCATTEGORIZED  # Default category if no matching prefix is found
+
 
 def main(token, repo, num_commits, exclude_dependabot):
     accept_header = "application/vnd.github.groot-preview+json"
@@ -66,6 +83,10 @@ def main(token, repo, num_commits, exclude_dependabot):
     print('Retrieved', len(commits), 'commits')
 
     # Load PR for each commit and print summary
+    category_results = {category['title']: [] for category in categories}
+    other_results = []
+    commits_with_multiple_labels = []
+
     for commit in commits:
         sha = commit['sha']
         author = commit['author']['login']
@@ -88,17 +109,73 @@ def main(token, repo, num_commits, exclude_dependabot):
         if not pulls:
             short_sha = sha[:7]
             commit_url = commit['html_url']
-            print(f'* {msg} ([@{author}]({author_url}) in [{short_sha}]({commit_url}))')
+
+            result = f'* {msg} ([@{author}]({author_url}) in [{short_sha}]({commit_url}))'
+            other_results.append(result)
             continue
 
         pull = pulls[0]
         pull_id = pull['number']
         pull_url = pull['html_url']
         msg = msg.replace(f'(#{pull_id})', '').strip()
-        print(f'* {msg} ([@{author}]({author_url}) in [#{pull_id}]({pull_url}))')
+
+        # Check if the pull request has changelog label
+        pull_labels = get_pull_request_labels(token, args.repo, pull_id)
+        changelog_labels = [label for label in pull_labels if label.startswith('changelog:')]
+
+        # Handle multiple changelog labels
+        if len(changelog_labels) > 1:
+            commits_with_multiple_labels.append((sha, pull_id, changelog_labels))
+            continue
+
+        category = UNCATTEGORIZED
+        if changelog_labels:
+            for cat in categories:
+                if changelog_labels[0].startswith(cat['label']):
+                    category = cat['title']
+                    break
+
+        result = f'* {msg} ([@{author}]({author_url}) in [#{pull_id}]({pull_url}))'
+        if category == UNCATTEGORIZED:
+            other_results.append(result)
+        else:
+            category_results[category].append(result)
+
+    # Print categorized pull requests
+    print()
+    for category, results in category_results.items():
+        if results and category:
+            print(f'{category}:\n')
+            for result in results:
+                print(result)
+            print()
+
+    # Print pull requests in the 'UNCATTEGORIZED' category
+    if other_results:
+        print(f'#### {UNCATTEGORIZED}:')
+        for result in other_results:
+            print(result)
+        print()
+
+    # Print warnings for commits with more than one changelog label
+    if commits_with_multiple_labels:
+        print("Warnings: Commits with more than one changelog label found. Please fix them:\n")
+        for sha, pull_id, labels in commits_with_multiple_labels:
+            pr_url = f"https://github.com/jaegertracing/{repo}/pull/{pull_id}"
+            print(f"Commit {sha} associated with multiple changelog labels: {', '.join(labels)}")
+            print(f"Pull Request URL: {pr_url}\n")
+        print()
 
     if skipped_dependabot:
-        print(f"\n(Skipped {skipped_dependabot} dependabot commit{'' if skipped_dependabot == 1 else 's'})")
+        print(f"(Skipped {skipped_dependabot} dependabot commit{'' if skipped_dependabot == 1 else 's'})")
+
+
+def get_pull_request_labels(token, repo, pull_number):
+    labels_url = f"https://api.github.com/repos/jaegertracing/{repo}/issues/{pull_number}/labels"
+    req = Request(labels_url)
+    req.add_header('Authorization', f'token {token}')
+    labels = json.loads(urlopen(req).read())
+    return [label['name'] for label in labels]
 
 
 if __name__ == "__main__":