Workflow Object ID Filtering

Developer guide for the Enezaframework workflow engine. Part of the Workflow series.

Table of Contents

1. Overview
2. SavedQuery.get_queryset(object_id=None)
3. object_id Type
4. Methods That Accept object_id
5. Non-Workflow Callers
6. Related Guides

Overview

Object ID filtering restricts a saved query's results to a single record. It exists because condition steps were previously evaluated against the whole table: when the same logical data appeared across documents, a condition could match a record from a different document and produce a false positive. Passing the document's primary key scopes the query to the record being evaluated.

The behavior is additive — when object_id is None, no filtering is applied.

flowchart LR
    Q["SavedQuery"] --> A["annotations + rules"]
    A --> B["filtered queryset"]
    B --> C{"object_id provided?"}
    C -->|Yes| D["filter(pk=object_id)"]
    C -->|No| E["return as-is"]

SavedQuery.get_queryset(object_id=None)

Defined in enezaworkflowengine/models/querybuilder.py:31-60:

def get_queryset(self, object_id=None):
    model_class = self.content_type.model_class()
    if not model_class:
        raise ValueError("Model not found")

    annotations = self.filters_json.get('annotations', []) if isinstance(self.filters_json, dict) else []
    if annotations:
        ann = parse_annotations(model_class, annotations)
        qs = model_class.objects.annotate(**ann)
    else:
        qs = model_class.objects.all()

    # ... build Q from rules or groups ...
    result_qs = qs.filter(q)

    if object_id is not None:
        result_qs = result_qs.filter(pk=object_id)

    return result_qs

The object filter is applied after the rule filter, so it narrows the already-matching set.

object_id Type

object_id is a string. Framework models use UUID primary keys, and Django's ORM accepts a UUID string directly in pk=..., so no manual coercion is needed. The workflow engine converts the document PK with str(document.pk) before passing it (enezaworkflowengine/services/condition_step.py:36).

Methods That Accept object_id

All three apply the same if object_id is not None: .filter(pk=object_id) guard.

Non-Workflow Callers

Callers outside the workflow engine — reports, the admin query builder, tests — can omit object_id; it defaults to None and no filtering occurs.

# All matching records
results = saved_query.get_queryset()

# Only the record with that primary key
results = saved_query.get_queryset(object_id="550e8400-e29b-41d4-a716-446655440000")

This is the backward-compatible behavior the enhancement preserves.

Related Guides

• Workflow Query Architecture
• Workflow Engine Auto-Extraction
• Workflow Aggregate Annotations
• Workflow Query Reference
• Workflow — Developer Index