mcp-ticketer 0.4.11__py3-none-any.whl → 2.0.1__py3-none-any.whl

mcp_ticketer/core/url_parser.py

@@ -0,0 +1,425 @@
+"""URL parsing utilities for extracting project/issue IDs from adapter URLs.
+
+This module provides functionality to detect and parse URLs from various ticket
+management platforms (Linear, JIRA, GitHub) and extract the relevant project or
+issue identifiers.
+
+Supported URL patterns:
+- Linear: https://linear.app/team-key/project/project-key-123
+- Linear: https://linear.app/team-key/issue/ISS-123
+- Linear: https://linear.app/team-key/view/view-name-uuid (view detection)
+- JIRA: https://company.atlassian.net/browse/PROJ
+- JIRA: https://company.atlassian.net/browse/PROJ-123
+- GitHub: https://github.com/owner/repo/projects/1
+- GitHub: https://github.com/owner/repo/issues/123
+- Asana: https://app.asana.com/0/{workspace_gid}/{task_gid}
+- Asana: https://app.asana.com/0/{workspace_gid}/project/{project_gid}
+"""
+
+import logging
+import re
+
+logger = logging.getLogger(__name__)
+
+
+class URLParserError(Exception):
+    """Raised when URL parsing fails."""
+
+    pass
+
+
+def is_url(value: str) -> bool:
+    """Detect if a string is a URL.
+
+    Args:
+        value: String to check
+
+    Returns:
+        True if the string appears to be a URL, False otherwise
+
+    Examples:
+        >>> is_url("https://linear.app/team/project/abc-123")
+        True
+        >>> is_url("PROJ-123")
+        False
+        >>> is_url("http://example.com")
+        True
+
+    """
+    if not value or not isinstance(value, str):
+        return False
+
+    # Check for URL scheme
+    return bool(
+        value.startswith(("http://", "https://")) or re.match(r"^[\w.-]+://", value)
+    )
+
+
+def extract_linear_id(url: str) -> tuple[str | None, str | None]:
+    """Extract project, issue, view, or team ID from Linear URL.
+
+    Supported formats:
+    - https://linear.app/workspace/project/project-slug-abc123/overview → "project-slug-abc123"
+    - https://linear.app/workspace/issue/ISS-123 → "ISS-123"
+    - https://linear.app/workspace/view/view-name-uuid → "view-name-uuid"
+    - https://linear.app/workspace/team/TEAM → "TEAM" (team key)
+
+    Args:
+        url: Linear URL string
+
+    Returns:
+        Tuple of (extracted_id, error_message). If successful, error_message is None.
+
+    Examples:
+        >>> extract_linear_id("https://linear.app/travel-bta/project/crm-system-f59a41/overview")
+        ('crm-system-f59a41', None)
+        >>> extract_linear_id("https://linear.app/myteam/issue/BTA-123")
+        ('BTA-123', None)
+        >>> extract_linear_id("https://linear.app/myteam/view/my-view-abc123")
+        ('my-view-abc123', None)
+
+    """
+    if not url:
+        return None, "Empty URL provided"
+
+    # Pattern 1: Project URLs - extract slug-id
+    # https://linear.app/workspace/project/project-slug-shortid/...
+    project_pattern = r"https?://linear\.app/[\w-]+/project/([\w-]+)"
+    match = re.search(project_pattern, url, re.IGNORECASE)
+    if match:
+        project_id = match.group(1)
+        logger.debug(f"Extracted Linear project ID '{project_id}' from URL")
+        return project_id, None
+
+    # Pattern 2: Issue URLs - extract issue key
+    # https://linear.app/workspace/issue/ISS-123
+    issue_pattern = r"https?://linear\.app/[\w-]+/issue/([\w]+-\d+)"
+    match = re.search(issue_pattern, url, re.IGNORECASE)
+    if match:
+        issue_id = match.group(1)
+        logger.debug(f"Extracted Linear issue ID '{issue_id}' from URL")
+        return issue_id, None
+
+    # Pattern 3: View URLs - extract view identifier (slug-uuid format)
+    # https://linear.app/workspace/view/view-name-uuid
+    view_pattern = r"https?://linear\.app/[\w-]+/view/([\w-]+)"
+    match = re.search(view_pattern, url, re.IGNORECASE)
+    if match:
+        view_id = match.group(1)
+        logger.debug(f"Extracted Linear view ID '{view_id}' from URL")
+        return view_id, None
+
+    # Pattern 4: Team URLs - extract team key
+    # https://linear.app/workspace/team/TEAM
+    team_pattern = r"https?://linear\.app/[\w-]+/team/([\w-]+)"
+    match = re.search(team_pattern, url, re.IGNORECASE)
+    if match:
+        team_key = match.group(1)
+        logger.debug(f"Extracted Linear team key '{team_key}' from URL")
+        return team_key, None
+
+    return None, f"Could not extract Linear ID from URL: {url}"
+
+
+def extract_jira_id(url: str) -> tuple[str | None, str | None]:
+    """Extract project or issue key from JIRA URL.
+
+    Supported formats:
+    - https://company.atlassian.net/browse/PROJ → "PROJ"
+    - https://company.atlassian.net/browse/PROJ-123 → "PROJ-123"
+    - https://jira.company.com/browse/PROJ → "PROJ"
+
+    Args:
+        url: JIRA URL string
+
+    Returns:
+        Tuple of (extracted_id, error_message). If successful, error_message is None.
+
+    Examples:
+        >>> extract_jira_id("https://company.atlassian.net/browse/PROJ")
+        ('PROJ', None)
+        >>> extract_jira_id("https://company.atlassian.net/browse/PROJ-123")
+        ('PROJ-123', None)
+
+    """
+    if not url:
+        return None, "Empty URL provided"
+
+    # Pattern: Extract key from browse URL
+    # https://company.atlassian.net/browse/PROJ or PROJ-123
+    browse_pattern = r"https?://[\w.-]+/browse/([\w]+-?\d*)"
+    match = re.search(browse_pattern, url, re.IGNORECASE)
+    if match:
+        issue_key = match.group(1)
+        logger.debug(f"Extracted JIRA key '{issue_key}' from URL")
+        return issue_key, None
+
+    # Alternative pattern for project URLs
+    # https://company.atlassian.net/projects/PROJ
+    project_pattern = r"https?://[\w.-]+/projects/([\w]+)"
+    match = re.search(project_pattern, url, re.IGNORECASE)
+    if match:
+        project_key = match.group(1)
+        logger.debug(f"Extracted JIRA project key '{project_key}' from URL")
+        return project_key, None
+
+    return None, f"Could not extract JIRA key from URL: {url}"
+
+
+def extract_github_id(url: str) -> tuple[str | None, str | None]:
+    """Extract project, issue, milestone, or PR number from GitHub URL.
+
+    Supported formats:
+    - https://github.com/owner/repo/projects/1 → "1"
+    - https://github.com/owner/repo/issues/123 → "123"
+    - https://github.com/owner/repo/milestones/5 → "5"
+    - https://github.com/owner/repo/pull/456 → "456"
+
+    Args:
+        url: GitHub URL string
+
+    Returns:
+        Tuple of (extracted_id, error_message). If successful, error_message is None.
+
+    Examples:
+        >>> extract_github_id("https://github.com/owner/repo/projects/1")
+        ('1', None)
+        >>> extract_github_id("https://github.com/owner/repo/issues/123")
+        ('123', None)
+        >>> extract_github_id("https://github.com/owner/repo/milestones/5")
+        ('5', None)
+
+    """
+    if not url:
+        return None, "Empty URL provided"
+
+    # Pattern 1: Project URLs - extract project number
+    # https://github.com/owner/repo/projects/1
+    project_pattern = r"https?://github\.com/[\w-]+/[\w-]+/projects/(\d+)"
+    match = re.search(project_pattern, url, re.IGNORECASE)
+    if match:
+        project_id = match.group(1)
+        logger.debug(f"Extracted GitHub project ID '{project_id}' from URL")
+        return project_id, None
+
+    # Pattern 2: Issue URLs - extract issue number
+    # https://github.com/owner/repo/issues/123
+    issue_pattern = r"https?://github\.com/[\w-]+/[\w-]+/issues/(\d+)"
+    match = re.search(issue_pattern, url, re.IGNORECASE)
+    if match:
+        issue_id = match.group(1)
+        logger.debug(f"Extracted GitHub issue ID '{issue_id}' from URL")
+        return issue_id, None
+
+    # Pattern 3: Milestone URLs - extract milestone number
+    # https://github.com/owner/repo/milestones/5
+    milestone_pattern = r"https?://github\.com/[\w-]+/[\w-]+/milestones/(\d+)"
+    match = re.search(milestone_pattern, url, re.IGNORECASE)
+    if match:
+        milestone_id = match.group(1)
+        logger.debug(f"Extracted GitHub milestone ID '{milestone_id}' from URL")
+        return milestone_id, None
+
+    # Pattern 4: Pull request URLs - extract PR number
+    # https://github.com/owner/repo/pull/456
+    pr_pattern = r"https?://github\.com/[\w-]+/[\w-]+/pull/(\d+)"
+    match = re.search(pr_pattern, url, re.IGNORECASE)
+    if match:
+        pr_id = match.group(1)
+        logger.debug(f"Extracted GitHub PR ID '{pr_id}' from URL")
+        return pr_id, None
+
+    return None, f"Could not extract GitHub ID from URL: {url}"
+
+
+def parse_github_repo_url(url: str) -> tuple[str | None, str | None, str | None]:
+    """Parse GitHub repository URL to extract owner and repo name.
+
+    Supported formats:
+    - https://github.com/owner/repo → ("owner", "repo")
+    - https://github.com/owner/repo/ → ("owner", "repo")
+    - https://github.com/owner/repo/issues → ("owner", "repo")
+    - https://github.com/owner/repo/projects/1 → ("owner", "repo")
+    - http://github.com/owner/repo → ("owner", "repo")
+
+    Args:
+        url: GitHub repository URL string
+
+    Returns:
+        Tuple of (owner, repo, error_message). If successful, error_message is None.
+
+    Examples:
+        >>> parse_github_repo_url("https://github.com/owner/repo")
+        ('owner', 'repo', None)
+        >>> parse_github_repo_url("https://github.com/owner/repo/")
+        ('owner', 'repo', None)
+        >>> parse_github_repo_url("https://github.com/owner/repo/issues/123")
+        ('owner', 'repo', None)
+
+    """
+    if not url:
+        return None, None, "Empty URL provided"
+
+    # Pattern: Extract owner and repo from any GitHub URL
+    # https://github.com/{owner}/{repo}[/anything/else]
+    github_pattern = r"https?://github\.com/([\w-]+)/([\w.-]+)(?:/|$)"
+    match = re.search(github_pattern, url, re.IGNORECASE)
+
+    if match:
+        owner = match.group(1)
+        repo = match.group(2)
+        logger.debug(f"Extracted GitHub owner '{owner}' and repo '{repo}' from URL")
+        return owner, repo, None
+
+    return None, None, f"Could not parse GitHub repository URL: {url}"
+
+
+def extract_asana_id(url: str) -> tuple[str | None, str | None]:
+    """Extract task or project GID from Asana URL.
+
+    Supported formats:
+    - https://app.asana.com/0/{workspace_gid}/{task_gid} → "{task_gid}"
+    - https://app.asana.com/0/{workspace_gid}/{task_gid}/f → "{task_gid}"
+    - https://app.asana.com/0/{workspace_gid}/list/{project_gid} → "{project_gid}"
+
+    Args:
+        url: Asana URL string
+
+    Returns:
+        Tuple of (extracted_id, error_message). If successful, error_message is None.
+
+    Examples:
+        >>> extract_asana_id("https://app.asana.com/0/1234567890/9876543210")
+        ('9876543210', None)
+        >>> extract_asana_id("https://app.asana.com/0/1234567890/list/5555555555")
+        ('5555555555', None)
+
+    """
+    if not url:
+        return None, "Empty URL provided"
+
+    # Pattern 1: Task URLs - extract task GID
+    # https://app.asana.com/0/{workspace_gid}/{task_gid}
+    # https://app.asana.com/0/{workspace_gid}/{task_gid}/f (with focus mode)
+    task_pattern = r"https?://app\.asana\.com/0/\d+/(\d+)"
+    match = re.search(task_pattern, url, re.IGNORECASE)
+    if match:
+        task_gid = match.group(1)
+        logger.debug(f"Extracted Asana task GID '{task_gid}' from URL")
+        return task_gid, None
+
+    # Pattern 2: Project/List URLs - extract project GID
+    # https://app.asana.com/0/{workspace_gid}/list/{project_gid}
+    project_pattern = r"https?://app\.asana\.com/0/\d+/list/(\d+)"
+    match = re.search(project_pattern, url, re.IGNORECASE)
+    if match:
+        project_gid = match.group(1)
+        logger.debug(f"Extracted Asana project GID '{project_gid}' from URL")
+        return project_gid, None
+
+    return None, f"Could not extract Asana ID from URL: {url}"
+
+
+def extract_id_from_url(
+    url: str, adapter_type: str | None = None
+) -> tuple[str | None, str | None]:
+    """Extract project/issue ID from URL for any supported adapter.
+
+    This is the main entry point for URL parsing. It auto-detects the adapter type
+    from the URL if not explicitly provided.
+
+    Args:
+        url: URL string to parse
+        adapter_type: Optional adapter type hint ("linear", "jira", "github").
+                     If not provided, adapter is auto-detected from URL domain.
+
+    Returns:
+        Tuple of (extracted_id, error_message). If successful, error_message is None.
+
+    Raises:
+        URLParserError: If URL parsing fails
+
+    Examples:
+        >>> extract_id_from_url("https://linear.app/team/project/abc-123")
+        ('abc-123', None)
+        >>> extract_id_from_url("https://company.atlassian.net/browse/PROJ-123")
+        ('PROJ-123', None)
+        >>> extract_id_from_url("https://github.com/owner/repo/issues/123")
+        ('123', None)
+
+    """
+    if not url:
+        return None, "Empty URL provided"
+
+    if not is_url(url):
+        # Not a URL - return as-is (could be a plain ID)
+        return url, None
+
+    # Auto-detect adapter type from URL if not provided
+    if not adapter_type:
+        # Check for specific domains first (more reliable than path patterns)
+        if "linear.app" in url.lower():
+            adapter_type = "linear"
+        elif "github.com" in url.lower():
+            adapter_type = "github"
+        elif "atlassian.net" in url.lower():
+            adapter_type = "jira"
+        elif "app.asana.com" in url.lower():
+            adapter_type = "asana"
+        # Fallback to path-based detection for self-hosted instances
+        elif "/browse/" in url:
+            adapter_type = "jira"
+        else:
+            return None, f"Unknown URL format - cannot auto-detect adapter: {url}"
+
+    # Route to appropriate parser
+    if adapter_type.lower() == "linear":
+        return extract_linear_id(url)
+    elif adapter_type.lower() == "jira":
+        return extract_jira_id(url)
+    elif adapter_type.lower() == "github":
+        return extract_github_id(url)
+    elif adapter_type.lower() == "asana":
+        return extract_asana_id(url)
+    else:
+        return None, f"Unsupported adapter type: {adapter_type}"
+
+
+def normalize_project_id(value: str, adapter_type: str | None = None) -> str:
+    """Normalize a project ID by extracting from URL if necessary.
+
+    This is a convenience function that handles both URLs and plain IDs.
+    If the value is a URL, it extracts the ID. If it's already a plain ID,
+    it returns it unchanged.
+
+    Args:
+        value: Project ID or URL
+        adapter_type: Optional adapter type hint
+
+    Returns:
+        Normalized project ID (extracted from URL if applicable)
+
+    Raises:
+        URLParserError: If URL parsing fails
+
+    Examples:
+        >>> normalize_project_id("PROJ-123")
+        'PROJ-123'
+        >>> normalize_project_id("https://linear.app/team/project/abc-123")
+        'abc-123'
+
+    """
+    if not value:
+        return value
+
+    # If not a URL, return as-is
+    if not is_url(value):
+        return value
+
+    # Extract ID from URL
+    extracted_id, error = extract_id_from_url(value, adapter_type)
+
+    if error:
+        raise URLParserError(error)
+
+    return extracted_id or value

mcp_ticketer/core/validators.py

@@ -0,0 +1,69 @@
+"""Field validation utilities for adapter data."""
+
+
+class ValidationError(Exception):
+    """Raised when field validation fails."""
+
+    pass
+
+
+class FieldValidator:
+    """Validates field lengths and formats across adapters."""
+
+    # Field length limits per adapter
+    LIMITS = {
+        "linear": {
+            "epic_description": 255,
+            "epic_name": 255,
+            "issue_description": 100000,  # Issues have much higher limit
+            "issue_title": 255,
+        },
+        "jira": {
+            "summary": 255,
+            "description": 32767,
+        },
+        "github": {
+            "title": 256,
+            "body": 65536,
+        },
+    }
+
+    @classmethod
+    def validate_field(
+        cls,
+        adapter_name: str,
+        field_name: str,
+        value: str | None,
+        truncate: bool = False,
+    ) -> str:
+        """Validate and optionally truncate a field value.
+
+        Args:
+            adapter_name: Name of adapter (linear, jira, github)
+            field_name: Name of field being validated
+            value: Field value to validate
+            truncate: If True, truncate instead of raising error
+
+        Returns:
+            Validated (possibly truncated) value
+
+        Raises:
+            ValidationError: If value exceeds limit and truncate=False
+
+        """
+        if value is None:
+            return ""
+
+        adapter_limits = cls.LIMITS.get(adapter_name.lower(), {})
+        limit = adapter_limits.get(field_name)
+
+        if limit and len(value) > limit:
+            if truncate:
+                return value[:limit]
+            else:
+                raise ValidationError(
+                    f"{field_name} exceeds {adapter_name} limit of {limit} characters "
+                    f"(got {len(value)}). Use truncate=True to auto-truncate."
+                )
+
+        return value