langgenius/dify · error · ValueError

an environment variable cannot be upserted and deleted in…

Error message

an environment variable cannot be upserted and deleted in the same patch

What it means

Raised by SyncEnvironmentVariablePatchPayload.validate_patch when the same id appears in both the upsert list and the delete list (api/controllers/console/app/workflow.py:135-136). Upserting and deleting the same variable in one patch is contradictory and would produce an undefined merge result, so the validator rejects it. Surfaced as a Pydantic ValidationError (400).

Solutions

1. Reconcile client state: if a variable is marked for deletion, remove it from the upsert list (and vice versa).
2. Compute the patch deterministically from a single source of truth before submit.
3. Add a client-side intersection check: `upsertIds.filter(id => !deleteIds.includes(id))`.

Example fix

// before: id 'a' both upserted and deleted
{environment_variables:[{id:'a',...}], deleted_environment_variable_ids:['a']}
// after: pick one — here, keep the upsert, drop from deletes
{environment_variables:[{id:'a',...}], deleted_environment_variable_ids:[]}

// Reconcile: a variable cannot be both upserted and deleted
const upsertIds = new Set(payload.environment_variables.map(v => v.id))
payload.deleted_environment_variable_ids = payload.deleted_environment_variable_ids.filter(id => !upsertIds.has(id))

function disjoint(upserts, deletes) {
  const u = new Set(upserts.map(v=>v.id)); return deletes.every(id => !u.has(id))
}

When it happens

Trigger: A variable id is present in environment_variables (upsert) AND in deleted_environment_variable_ids in the same request. Happens when UI state allows a variable to be both edited and marked for deletion.

Common situations: User edits a variable then (via a stale action) marks it deleted; front-end keeps both a 'pending edit' and a 'pending delete' entry; merging two partial patches without reconciling conflicts; buggy undo flow.

Related errors

• deleted environment variable ids must be unique
• deleted environment variable ids must not be empty
• patched environment variable ids must be unique
• patched environment variables require an id
• Unsupported environment variable value type

AI-assisted analysis of langgenius/dify@ef8544b173 (2026-08-12). Data as JSON: /api/errors/5478703879994964. Report an issue: GitHub.

Appendix: source

Thrown at api/controllers/console/app/workflow.py:136

class SyncEnvironmentVariablePatchPayload(BaseModel):
    environment_variables: list[dict[str, Any]] = Field(default_factory=list)
    deleted_environment_variable_ids: list[str] = Field(default_factory=list)

    @model_validator(mode="after")
    def validate_patch(self) -> Self:
        """Require stable, disjoint IDs so the service can merge the patch deterministically."""
        upsert_ids = [variable.get("id") for variable in self.environment_variables]
        if any(not isinstance(variable_id, str) or not variable_id for variable_id in upsert_ids):
            raise ValueError("patched environment variables require an id")
        if len(set(upsert_ids)) != len(upsert_ids):
            raise ValueError("patched environment variable ids must be unique")
        if any(not variable_id for variable_id in self.deleted_environment_variable_ids):
            raise ValueError("deleted environment variable ids must not be empty")
        if len(set(self.deleted_environment_variable_ids)) != len(self.deleted_environment_variable_ids):
            raise ValueError("deleted environment variable ids must be unique")
        if set(upsert_ids).intersection(self.deleted_environment_variable_ids):
            raise ValueError("an environment variable cannot be upserted and deleted in the same patch")
        return self


class SyncDraftWorkflowPayload(BaseModel):
    model_config = ConfigDict(extra="forbid")

    graph: dict[str, Any]
    features: dict[str, Any]
    hash: str | None = None
    is_collaborative: bool = Field(default=False, alias="_is_collaborative")
    environment_variable_patch: SyncEnvironmentVariablePatchPayload | None = None
    conversation_variables: list[dict[str, Any]] = Field(
        default_factory=list,
    )


class BaseWorkflowRunPayload(BaseModel):
    files: list[dict[str, Any]] | None = Field(default=None)

View on GitHub (pinned to ef8544b173)