feat: add Azure DevOps integration support (#11243)

Co-authored-by: Graham Neubig <neubig@gmail.com>

openhands/integrations/azure_devops/service/resolver.py

@@ -0,0 +1,166 @@
+from datetime import datetime
+
+from openhands.core.logger import openhands_logger as logger
+from openhands.integrations.azure_devops.service.base import AzureDevOpsMixinBase
+from openhands.integrations.service_types import Comment
+
+
+class AzureDevOpsResolverMixin(AzureDevOpsMixinBase):
+    """Helper methods used for the Azure DevOps Resolver."""
+
+    async def get_issue_or_pr_title_and_body(
+        self, repository: str, issue_number: int
+    ) -> tuple[str, str]:
+        """Get the title and body of a pull request or work item.
+
+        First attempts to get as a PR, then falls back to work item if not found.
+
+        Args:
+            repository: Repository name in format 'organization/project/repo'
+            issue_number: The PR number or work item ID
+
+        Returns:
+            A tuple of (title, body)
+        """
+        org, project, repo = self._parse_repository(repository)
+
+        # URL-encode components to handle spaces and special characters
+        org_enc = self._encode_url_component(org)
+        project_enc = self._encode_url_component(project)
+        repo_enc = self._encode_url_component(repo)
+
+        # Try to get as a pull request first
+        try:
+            pr_url = f'{self.base_url}/{org_enc}/{project_enc}/_apis/git/repositories/{repo_enc}/pullrequests/{issue_number}?api-version=7.1'
+            response, _ = await self._make_request(pr_url)
+            title = response.get('title') or ''
+            body = response.get('description') or ''
+            return title, body
+        except Exception as pr_error:
+            logger.debug(f'Failed to get as PR: {pr_error}, trying as work item')
+
+        # Fall back to work item
+        try:
+            wi_url = f'{self.base_url}/{org_enc}/{project_enc}/_apis/wit/workitems/{issue_number}?api-version=7.1'
+            response, _ = await self._make_request(wi_url)
+            fields = response.get('fields', {})
+            title = fields.get('System.Title') or ''
+            body = fields.get('System.Description') or ''
+            return title, body
+        except Exception as wi_error:
+            logger.error(f'Failed to get as work item: {wi_error}')
+            return '', ''
+
+    async def get_issue_or_pr_comments(
+        self, repository: str, issue_number: int, max_comments: int = 10
+    ) -> list[Comment]:
+        """Get comments for a pull request or work item.
+
+        First attempts to get PR comments, then falls back to work item comments if not found.
+
+        Args:
+            repository: Repository name in format 'organization/project/repo'
+            issue_number: The PR number or work item ID
+            max_comments: Maximum number of comments to return
+
+        Returns:
+            List of Comment objects ordered by creation date
+        """
+        # Try to get PR comments first
+        try:
+            comments = await self.get_pr_comments(  # type: ignore[attr-defined]
+                repository, issue_number, max_comments
+            )
+            if comments:
+                return comments
+        except Exception as pr_error:
+            logger.debug(f'Failed to get PR comments: {pr_error}, trying work item')
+
+        # Fall back to work item comments
+        try:
+            return await self.get_work_item_comments(  # type: ignore[attr-defined]
+                repository, issue_number, max_comments
+            )
+        except Exception as wi_error:
+            logger.error(f'Failed to get work item comments: {wi_error}')
+            return []
+
+    async def get_review_thread_comments(
+        self,
+        thread_id: int,
+        repository: str,
+        pr_number: int,
+        max_comments: int = 10,
+    ) -> list[Comment]:
+        """Get all comments in a specific PR review thread.
+
+        Azure DevOps organizes PR comments into threads. This method retrieves
+        all comments from a specific thread.
+
+        Args:
+            thread_id: The thread ID
+            repository: Repository name in format 'organization/project/repo'
+            pr_number: Pull request number
+            max_comments: Maximum number of comments to return
+
+        Returns:
+            List of Comment objects representing the thread
+        """
+        org, project, repo = self._parse_repository(repository)
+
+        # URL-encode components to handle spaces and special characters
+        org_enc = self._encode_url_component(org)
+        project_enc = self._encode_url_component(project)
+        repo_enc = self._encode_url_component(repo)
+
+        url = f'{self.base_url}/{org_enc}/{project_enc}/_apis/git/repositories/{repo_enc}/pullrequests/{pr_number}/threads/{thread_id}?api-version=7.1'
+
+        try:
+            response, _ = await self._make_request(url)
+            comments_data = response.get('comments', [])
+
+            all_comments: list[Comment] = []
+
+            for comment_data in comments_data:
+                # Extract author information
+                author_info = comment_data.get('author', {})
+                author = author_info.get('displayName', 'unknown')
+
+                # Parse dates
+                created_at = (
+                    datetime.fromisoformat(
+                        comment_data.get('publishedDate', '').replace('Z', '+00:00')
+                    )
+                    if comment_data.get('publishedDate')
+                    else datetime.fromtimestamp(0)
+                )
+
+                updated_at = (
+                    datetime.fromisoformat(
+                        comment_data.get('lastUpdatedDate', '').replace('Z', '+00:00')
+                    )
+                    if comment_data.get('lastUpdatedDate')
+                    else created_at
+                )
+
+                # Check if it's a system comment
+                is_system = comment_data.get('commentType', 1) != 1  # 1 = text comment
+
+                comment = Comment(
+                    id=str(comment_data.get('id', 0)),
+                    body=self._truncate_comment(comment_data.get('content', '')),
+                    author=author,
+                    created_at=created_at,
+                    updated_at=updated_at,
+                    system=is_system,
+                )
+
+                all_comments.append(comment)
+
+            # Sort by creation date and limit
+            all_comments.sort(key=lambda c: c.created_at)
+            return all_comments[:max_comments]
+
+        except Exception as error:
+            logger.error(f'Failed to get thread {thread_id} comments: {error}')
+            return []