You can use the operation Document.Fetch
when you want to fetch a given document by its ID, or by its path. Actually you won't probably have to use it frequently: most of the time you'll be working on the current document, or one of its children, or its parent, or on the result of a query.
Anyway, if you want to access a document by its path or its ID, just drop this Fetch > Document
operation in the operation chain editor, and either put:
- A path, like
/default-domain/workspaces/my-workspace
; You will find the path of a document in its URL: It is the part of URL that starts with/default-domain/...
(if you are in the default domain) and that finishes with the name of the document (most often its title without accents, space, ... and sometimes with some additional numbers). - A computed path using a scripted instruction like
expr:/default-domain/sections+@{Document.path.substring("/default-domain/workspaces".length -1)}
. This expression will give the path of a published document in the sections considering it has been published in the same place as in the workspaces. - An ID like
76c69a54-0230-457a-b42c-e819d5ace862
. You can get the ID of a document by clicking on the permanent link beside the title of a document in Nuxeo Platform.