cwiki_get_page

Instructions

Implementation Reference

• src/incubator_cwiki_mcp/tools.py:87-112 (handler)

  The main handler for the cwiki_get_page tool. It fetches a Confluence wiki page by title or page_id, retrieves its content in the requested format (plain, view, or storage), and returns a dict with page summary, ancestors, and content.

  @mcp.tool()
  def cwiki_get_page(
      title: str | None = None,
      page_id: str | None = None,
      format: Literal["plain", "view", "storage"] = "plain",
      force_refresh: bool = False,
  ) -> dict[str, Any]:
      """Fetch a wiki page by page title or page id."""
      if not title and not page_id:
          raise ValueError("Provide either title or page_id.")
  
      page = (
          client.get_page_by_id(page_id, force_refresh=force_refresh)
          if page_id
          else client.get_page_by_title(title or "", force_refresh=force_refresh)
      )
  
      return {
          **client.page_summary(page),
          "ancestors": [
              {"id": ancestor.get("id"), "title": ancestor.get("title")}
              for ancestor in page.get("ancestors", [])
          ],
          "contentFormat": format,
          "content": page_content(page, format),
      }

• src/incubator_cwiki_mcp/tools.py:169-175 (helper)

  Helper function page_content() called by cwiki_get_page to extract body content in storage, view, or plain (HTML-to-text) format.

  def page_content(page: dict[str, Any], format: Literal["plain", "view", "storage"]) -> str:
      body = page.get("body", {})
      if format == "storage":
          return body.get("storage", {}).get("value", "")
  
      view = body.get("view", {}).get("value") or body.get("storage", {}).get("value", "")
      return html_to_text(view) if format == "plain" else view

• src/incubator_cwiki_mcp/tools.py:87-88 (registration)

  The @mcp.tool() decorator on line 87 registers cwiki_get_page as an MCP tool with the FastMCP server.

  @mcp.tool()
  def cwiki_get_page(

• src/incubator_cwiki_mcp/client.py:64-71 (helper)

  Helper client.get_page_by_id() called by the handler when page_id is provided.

  def get_page_by_id(page_id: str | None, *, force_refresh: bool = False) -> dict[str, Any]:
      if not page_id:
          raise ValueError("page_id must not be empty")
      return confluence_get(
          f"/rest/api/content/{urllib.parse.quote(page_id)}",
          {"expand": "body.storage,body.view,version,history,ancestors,space"},
          force_refresh=force_refresh,
      )

• src/incubator_cwiki_mcp/client.py:74-95 (helper)

  Helper client.get_page_by_title() called by the handler when title is provided.

  def get_page_by_title(title: str, *, force_refresh: bool = False) -> dict[str, Any]:
      if not title.strip():
          raise ValueError("title must not be empty")
  
      pages = confluence_get(
          "/rest/api/content",
          {
              "spaceKey": SPACE_KEY,
              "title": title,
              "type": "page",
              "status": "current",
              "limit": "2",
              "expand": "body.storage,body.view,version,history,ancestors,space",
          },
          force_refresh=force_refresh,
      )
      results = pages.get("results", [])
      if not results:
          raise ValueError(f'No page found with title "{title}" in space {SPACE_KEY}.')
      if len(results) > 1:
          raise ValueError(f'More than one page matched title "{title}". Use page_id instead.')
      return results[0]