Core Patterns: read, delete, search

Every script that uses cascade_cms follows the same shape: open the wrapper as a context manager, queue up one or more operations on cascade.operations, then call submit_requests() once to run them all concurrently.

The three examples below use the same skeleton but highlight the three distinct response shapes you'll see across the library:

Operation	Returns	Why it's shown
read	A large, structured Asset object	Most operations that touch an existing asset return this shape
delete	A simple success object (CascadeSuccess)	Mutating operations that don't return data use this minimal shape
search	A list result, driven by a required SearchInformation payload	Shows the "operation needs a payload object" pattern

All three also demonstrate the same failure-handling rule: failures are returned as values, not raised. Always check isinstance(result, CascadeError) before using a result.

---

Pattern 1 — read: fetching a structured asset

with CascadeWrapperBase("config.json") as cascade:
    # Build a Path identifier referencing a page in the site
    identifier = Path(asset_type="page", path="index", siteName="Default")

    # Queue the read operation
    cascade.operations.read(identifier)

    # Submit requests and get back the results list
    results = cascade.submit_requests()
    result = results[0]

    # Check if the operation returned a CascadeError
    if isinstance(result, CascadeError):
        print(f"Error: {result.message}")
    else:
        # Access structured asset fields
        print(result.displayName)
        print(result.metadata)

read represents the response shape most other "fetch" operations follow — readAudits, readAccessRights, readWorkflowSettings, etc. — and they all return a structured object specific to what was requested.

---

Pattern 2 — delete: a simple success response

with CascadeWrapperBase("config.json") as cascade:
    # Build a Path identifier for the asset to delete
    identifier = Path(asset_type="page", path="old-page", siteName="Default")

    # Configure delete parameters
    payload = deleteParameters(
        doWorkflow=False,
        destinations_identifiers=[],
        unpublish=True
    )

    # Queue the delete operation
    cascade.operations.delete(identifier, payload=payload)

    # Submit requests and get the result
    results = cascade.submit_requests()
    result = results[0]

    if isinstance(result, CascadeError):
        print(f"Delete failed: {result.message}")
    else:
        print("Asset successfully deleted.")

delete and other mutating operations (copy, move, publish, checkIn, editAccessRights) return confirmation only — not the modified asset — so callers should not expect asset data back from these operations.

---

Pattern 3 — search: payload-driven, list response

with CascadeWrapperBase("config.json") as cascade:
    # Build the SearchInformation payload
    payload = SearchInformation(
        siteName="Default",
        searchTerms="news",
        searchFields=["name"],
        searchTypes=["page"]
    )

    # Queue the search operation
    cascade.operations.search(payload)

    # Submit requests and inspect the result
    results = cascade.submit_requests()
    result = results[0]

    if isinstance(result, CascadeError):
        print(f"Search failed: {result.message}")
    else:
        # Iterate over the flat list elements
        for item in result.flat:
            print(item)

search requires a typed payload object — there is no bare identifier shortcut — and naming the other operations that follow the same pattern: readAudits (auditParameters), editWorkflowSettings, etc.

---

Chaining and Batching

All three patterns above run a single operation per script. In practice you can queue multiple chains — even mixing operation types — before calling submit_requests() once:

with CascadeWrapperBase("config.json") as cascade:
    # Queue multiple independent chains
    cascade.operations.read(Path(asset_type="page", path="index", siteName="Default"))
    cascade.operations.delete(Path(asset_type="page", path="old-page", siteName="Default"), payload=deleteParameters(doWorkflow=False, destinations_identifiers=[], unpublish=True))
    cascade.operations.search(SearchInformation(siteName="Default", searchTerms="news", searchFields=["name"], searchTypes=["page"]))

    # All three run concurrently and results are returned in creation order
    results = cascade.submit_requests()

See Administrative Operations for the messages and preferences operations.