pull-request-fixer 0.1.0__py3-none-any.whl

pull_request_fixer/exceptions.py

@@ -0,0 +1,66 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: 2025 The Linux Foundation
+
+"""Custom exceptions for pr-title-fixer."""
+
+from __future__ import annotations
+
+
+class PRTitleFixerError(Exception):
+    """Base exception for pr-title-fixer."""
+
+    pass
+
+
+class FileAccessError(PRTitleFixerError):
+    """Error accessing or reading a file."""
+
+    pass
+
+
+class GitHubAPIError(PRTitleFixerError):
+    """Error communicating with GitHub API."""
+
+    pass
+
+
+class AuthenticationError(GitHubAPIError):
+    """GitHub authentication failed."""
+
+    pass
+
+
+class RateLimitError(GitHubAPIError):
+    """GitHub API rate limit exceeded."""
+
+    def __init__(
+        self,
+        message: str = "GitHub API rate limit exceeded",
+        reset_time: int | None = None,
+    ):
+        """Initialize rate limit error.
+
+        Args:
+            message: Error message
+            reset_time: Unix timestamp when rate limit resets
+        """
+        super().__init__(message)
+        self.reset_time = reset_time
+
+
+class NetworkError(PRTitleFixerError):
+    """Network communication error."""
+
+    pass
+
+
+class GitOperationError(PRTitleFixerError):
+    """Error performing git operation."""
+
+    pass
+
+
+class ConfigurationError(PRTitleFixerError):
+    """Configuration error."""
+
+    pass

pull_request_fixer/github_client.py

@@ -0,0 +1,330 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: 2025 The Linux Foundation
+
+"""GitHub API client for repository and PR operations."""
+
+from __future__ import annotations
+
+import base64
+from typing import Any
+
+import httpx
+from tenacity import retry, stop_after_attempt, wait_exponential
+
+from .exceptions import FileAccessError
+
+
+class GitHubClient:
+    """Client for GitHub API operations."""
+
+    def __init__(self, token: str, base_url: str = "https://api.github.com"):
+        """Initialize GitHub client.
+
+        Args:
+            token: GitHub personal access token
+            base_url: GitHub API base URL
+        """
+        self.token = token
+        self.base_url = base_url
+        self.headers = {
+            "Authorization": f"Bearer {token}",
+            "Accept": "application/vnd.github+json",
+            "X-GitHub-Api-Version": "2022-11-28",
+        }
+
+    async def __aenter__(self) -> GitHubClient:
+        """Async context manager entry."""
+        return self
+
+    async def __aexit__(self, exc_type, exc_val, exc_tb) -> None:  # type: ignore[no-untyped-def]
+        """Async context manager exit."""
+        pass
+
+    async def validate_token(self) -> tuple[bool, str, list[str]]:
+        """Validate GitHub token and check permissions.
+
+        Returns:
+            Tuple of (is_valid, username, scopes)
+
+        Raises:
+            FileAccessError: If token validation fails
+        """
+        query = """
+        query {
+          viewer {
+            login
+          }
+        }
+        """
+
+        try:
+            result = await self._graphql_request(query)
+            viewer = result.get("viewer", {})
+            username = viewer.get("login", "")
+
+            # Get token scopes from REST API
+            # Note: GitHub Actions tokens may not have access to /user endpoint
+            scopes: list[str] = []
+            try:
+                url = f"{self.base_url}/user"
+                async with httpx.AsyncClient(timeout=30.0) as client:
+                    response = await client.get(url, headers=self.headers)
+                    response.raise_for_status()
+
+                    # Scopes are in the X-OAuth-Scopes header
+                    scopes_header = response.headers.get("X-OAuth-Scopes", "")
+                    scopes = [
+                        s.strip() for s in scopes_header.split(",") if s.strip()
+                    ]
+            except httpx.HTTPStatusError as scope_error:
+                # GitHub Actions tokens may not have user access (403)
+                # This is expected and we can continue without scope info
+                if scope_error.response.status_code != 403:
+                    raise
+
+            return True, username, scopes
+
+        except Exception as e:
+            msg = f"Token validation failed: {e}"
+            raise FileAccessError(msg) from e
+
+    @retry(  # type: ignore[misc]
+        stop=stop_after_attempt(3),
+        wait=wait_exponential(multiplier=1, min=2, max=10),
+    )
+    async def _request(
+        self,
+        method: str,
+        endpoint: str,
+        **kwargs: Any,
+    ) -> dict[str, Any] | list[dict[str, Any]]:
+        """Make an API request with retry logic.
+
+        Args:
+            method: HTTP method
+            endpoint: API endpoint path
+            **kwargs: Additional arguments for httpx request
+
+        Returns:
+            Response JSON data
+
+        Raises:
+            FileAccessError: If request fails
+        """
+        url = f"{self.base_url}{endpoint}"
+
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            try:
+                response = await client.request(
+                    method,
+                    url,
+                    headers=self.headers,
+                    **kwargs,
+                )
+                response.raise_for_status()
+                result: dict[str, Any] | list[dict[str, Any]] = response.json()
+                return result
+            except httpx.HTTPStatusError as e:
+                msg = f"GitHub API error: {e.response.status_code} - {e.response.text}"
+                raise FileAccessError(msg) from e
+            except httpx.RequestError as e:
+                msg = f"Request failed: {e}"
+                raise FileAccessError(msg) from e
+
+    @retry(  # type: ignore[misc]
+        stop=stop_after_attempt(3),
+        wait=wait_exponential(multiplier=1, min=2, max=10),
+    )
+    async def _graphql_request(
+        self,
+        query: str,
+        variables: dict[str, Any] | None = None,
+    ) -> dict[str, Any]:
+        """Make a GraphQL API request with retry logic.
+
+        Args:
+            query: GraphQL query string
+            variables: Query variables
+
+        Returns:
+            Response data
+
+        Raises:
+            FileAccessError: If request fails
+        """
+        url = "https://api.github.com/graphql"
+        payload: dict[str, Any] = {"query": query}
+        if variables:
+            payload["variables"] = variables
+
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            try:
+                response = await client.post(
+                    url,
+                    headers=self.headers,
+                    json=payload,
+                )
+                response.raise_for_status()
+                json_response: Any = response.json()
+                result: dict[str, Any] = (
+                    json_response if isinstance(json_response, dict) else {}
+                )
+
+                # Check for GraphQL errors
+                if "errors" in result:
+                    errors = result["errors"]
+                    msg = f"GraphQL errors: {errors}"
+                    raise FileAccessError(msg)
+
+                data: dict[str, Any] = result.get("data", {})
+                return data
+            except httpx.HTTPStatusError as e:
+                msg = f"GitHub API error: {e.response.status_code} - {e.response.text}"
+                raise FileAccessError(msg) from e
+            except httpx.RequestError as e:
+                msg = f"Request failed: {e}"
+                raise FileAccessError(msg) from e
+
+    async def graphql(
+        self,
+        query: str,
+        variables: dict[str, Any] | None = None,
+    ) -> dict[str, Any]:
+        """Execute a GraphQL query.
+
+        Args:
+            query: GraphQL query string
+            variables: Query variables
+
+        Returns:
+            Response data
+
+        Raises:
+            FileAccessError: If request fails
+        """
+        result: dict[str, Any] = await self._graphql_request(query, variables)
+        return result
+
+    async def get_pr_files(
+        self, owner: str, repo: str, pr_number: int
+    ) -> list[dict[str, Any]]:
+        """Get files changed in a pull request.
+
+        Args:
+            owner: Repository owner
+            repo: Repository name
+            pr_number: Pull request number
+
+        Returns:
+            List of changed files
+        """
+        files = await self._request(
+            "GET",
+            f"/repos/{owner}/{repo}/pulls/{pr_number}/files",
+        )
+        return files if isinstance(files, list) else []
+
+    async def get_file_content(
+        self, owner: str, repo: str, path: str, ref: str
+    ) -> str:
+        """Get file content from a repository.
+
+        Args:
+            owner: Repository owner
+            repo: Repository name
+            path: File path
+            ref: Git ref (branch/commit SHA)
+
+        Returns:
+            Decoded file content
+
+        Raises:
+            FileAccessError: If file cannot be retrieved
+        """
+        result = await self._request(
+            "GET",
+            f"/repos/{owner}/{repo}/contents/{path}",
+            params={"ref": ref},
+        )
+
+        if not isinstance(result, dict):
+            msg = f"Unexpected response type for file content: {type(result)}"
+            raise FileAccessError(msg)
+
+        content_b64 = result.get("content", "")
+        if not content_b64:
+            return ""
+
+        try:
+            return base64.b64decode(content_b64).decode("utf-8")
+        except (ValueError, UnicodeDecodeError) as e:
+            msg = f"Failed to decode file content: {e}"
+            raise FileAccessError(msg) from e
+
+    async def update_file(
+        self,
+        owner: str,
+        repo: str,
+        path: str,
+        content: str,
+        message: str,
+        branch: str,
+        sha: str,
+    ) -> dict[str, Any]:
+        """Update a file in a repository.
+
+        Args:
+            owner: Repository owner
+            repo: Repository name
+            path: File path
+            content: New file content
+            message: Commit message
+            branch: Branch name
+            sha: Current file SHA (for conflict detection)
+
+        Returns:
+            Commit data
+        """
+        content_b64 = base64.b64encode(content.encode("utf-8")).decode("utf-8")
+
+        result = await self._request(
+            "PUT",
+            f"/repos/{owner}/{repo}/contents/{path}",
+            json={
+                "message": message,
+                "content": content_b64,
+                "branch": branch,
+                "sha": sha,
+            },
+        )
+        return result if isinstance(result, dict) else {}
+
+    async def get_rate_limit(self) -> dict[str, Any]:
+        """Get current API rate limit status.
+
+        Returns:
+            Rate limit information
+        """
+        result = await self._request("GET", "/rate_limit")
+        return result if isinstance(result, dict) else {}
+
+    async def create_comment(
+        self, owner: str, repo: str, pr_number: int, body: str
+    ) -> dict[str, Any]:
+        """Create a comment on a pull request.
+
+        Args:
+            owner: Repository owner
+            repo: Repository name
+            pr_number: Pull request number
+            body: Comment body
+
+        Returns:
+            Comment data
+        """
+        result = await self._request(
+            "POST",
+            f"/repos/{owner}/{repo}/issues/{pr_number}/comments",
+            json={"body": body},
+        )
+        return result if isinstance(result, dict) else {}

pull_request_fixer/graphql_queries.py

@@ -0,0 +1,310 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: 2025 The Linux Foundation
+
+"""GraphQL queries for efficient organization scanning."""
+
+# Lightweight query to list repositories without PR nodes for accurate counting
+# This is the SAME query used by dependamerge for efficient repo enumeration
+ORG_REPOS_ONLY = """
+query($org: String!, $reposCursor: String) {
+  organization(login: $org) {
+    repositories(first: 100, after: $reposCursor, orderBy: { field: NAME, direction: ASC }) {
+      totalCount
+      pageInfo {
+        hasNextPage
+        endCursor
+      }
+      nodes {
+        nameWithOwner
+        isArchived
+      }
+    }
+  }
+}
+"""
+
+
+# Query to get repositories WITH open PRs in an organization (aggregated)
+# NOTE: This should NOT be used for counting - use ORG_REPOS_ONLY instead
+ORG_REPOS_WITH_PRS = """
+query($org: String!, $cursor: String, $prsPageSize: Int!, $contextsPageSize: Int!) {
+  organization(login: $org) {
+    repositories(first: 30, after: $cursor, orderBy: { field: NAME, direction: ASC }) {
+      pageInfo {
+        hasNextPage
+        endCursor
+      }
+      nodes {
+        nameWithOwner
+        isArchived
+        owner {
+          login
+        }
+        name
+        pullRequests(
+          states: OPEN
+          first: $prsPageSize
+          orderBy: { field: CREATED_AT, direction: DESC }
+        ) {
+          totalCount
+          pageInfo {
+            hasNextPage
+            endCursor
+          }
+          nodes {
+            number
+            title
+            body
+            url
+            isDraft
+            author { login }
+            mergeable
+            mergeStateStatus
+            baseRefName
+            headRefName
+            headRefOid
+            createdAt
+            updatedAt
+            headRepository {
+              nameWithOwner
+              url
+            }
+            baseRepository {
+              nameWithOwner
+              url
+            }
+            isCrossRepository
+            maintainerCanModify
+            commits(last: 1) {
+              nodes {
+                commit {
+                  oid
+                  statusCheckRollup {
+                    state
+                    contexts(first: $contextsPageSize) {
+                      nodes {
+                        __typename
+                        ... on CheckRun {
+                          name
+                          status
+                          conclusion
+                        }
+                        ... on StatusContext {
+                          context
+                          state
+                        }
+                      }
+                    }
+                  }
+                }
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+}
+"""
+
+
+# Query to get open PRs for a specific repository with status checks
+# This matches dependamerge's REPO_OPEN_PRS_PAGE with parameterized page sizes
+REPO_OPEN_PRS_PAGE = """
+query($owner: String!, $name: String!, $prsCursor: String, $prsPageSize: Int!, $filesPageSize: Int!, $commentsPageSize: Int!, $contextsPageSize: Int!) {
+  repository(owner: $owner, name: $name) {
+    nameWithOwner
+    pullRequests(
+      states: OPEN
+      first: $prsPageSize
+      after: $prsCursor
+      orderBy: { field: CREATED_AT, direction: DESC }
+    ) {
+      pageInfo {
+        hasNextPage
+        endCursor
+      }
+      nodes {
+        number
+        title
+        body
+        url
+        isDraft
+        author { login }
+        mergeable
+        mergeStateStatus
+        baseRefName
+        headRefName
+        headRefOid
+        createdAt
+        updatedAt
+        files(first: $filesPageSize) {
+          nodes {
+            path
+            additions
+            deletions
+          }
+        }
+        comments(first: $commentsPageSize, orderBy: { field: UPDATED_AT, direction: DESC }) {
+          nodes {
+            author { login }
+            body
+            createdAt
+          }
+        }
+        reviews(first: 20, states: [PENDING, COMMENTED, APPROVED, CHANGES_REQUESTED]) {
+          nodes {
+            id
+            author { login }
+            state
+            body
+            createdAt
+            updatedAt
+          }
+        }
+        headRepository {
+          nameWithOwner
+          url
+        }
+        baseRepository {
+          nameWithOwner
+          url
+        }
+        isCrossRepository
+        maintainerCanModify
+        commits(last: 1) {
+          nodes {
+            commit {
+              oid
+              statusCheckRollup {
+                state
+                contexts(first: $contextsPageSize) {
+                  nodes {
+                    __typename
+                    ... on CheckRun {
+                      name
+                      status
+                      conclusion
+                    }
+                    ... on StatusContext {
+                      context
+                      state
+                    }
+                  }
+                }
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+}
+"""
+
+# Query to get a single PR with status checks
+PR_WITH_STATUS = """
+query($owner: String!, $name: String!, $number: Int!, $contextsPageSize: Int!) {
+  repository(owner: $owner, name: $name) {
+    pullRequest(number: $number) {
+      number
+      title
+      body
+      url
+      isDraft
+      author { login }
+      mergeable
+      mergeStateStatus
+      baseRefName
+      headRefName
+      headRefOid
+      createdAt
+      updatedAt
+      headRepository {
+        nameWithOwner
+        url
+      }
+      baseRepository {
+        nameWithOwner
+        url
+      }
+      isCrossRepository
+      maintainerCanModify
+      commits(last: 1) {
+        nodes {
+          commit {
+            oid
+            statusCheckRollup {
+              state
+              contexts(first: $contextsPageSize) {
+                nodes {
+                  __typename
+                  ... on CheckRun {
+                    name
+                    status
+                    conclusion
+                  }
+                  ... on StatusContext {
+                    context
+                    state
+                  }
+                }
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+}
+"""
+
+# Query to get first commit message for a PR (lightweight)
+PR_FIRST_COMMIT = """
+query($owner: String!, $name: String!, $number: Int!) {
+  repository(owner: $owner, name: $name) {
+    pullRequest(number: $number) {
+      number
+      title
+      commits(first: 1) {
+        nodes {
+          commit {
+            oid
+            message
+            messageHeadline
+            messageBody
+          }
+        }
+      }
+    }
+  }
+}
+"""
+
+# Bulk query to get first commit messages for multiple PRs (performance optimized)
+# This uses GraphQL aliases to fetch multiple PRs in a single request
+BULK_PR_COMMITS_TEMPLATE = """
+query($owner: String!, $name: String!) {
+  repository(owner: $owner, name: $name) {
+    {pr_queries}
+  }
+}
+"""
+
+# Template for individual PR in bulk query
+PR_COMMIT_FRAGMENT = """
+    pr{number}: pullRequest(number: {number}) {
+      number
+      title
+      commits(first: 1) {
+        nodes {
+          commit {
+            oid
+            message
+            messageHeadline
+            messageBody
+          }
+        }
+      }
+    }
+"""