| icon | grid-round-2-plus |
|---|
Filters are PCGEx's universal condition system. You configure them once as sub-nodes and wire them into any consuming node's filter input. The same filter definition works across dozens of operations without changes.
Every filter lives here, organized by what it evaluates:
| Section | Contents | Cover image |
|---|---|---|
| common-settings | Base filter definition — priority, error handling, result output settings | |
| point-filters | Evaluate individual points — attribute checks, spatial tests, math comparisons | |
| edge-filters | Evaluate cluster edges — segment properties, connectivity | |
| vtx-filters | Evaluate cluster Vtx — adjacency, topology | |
| data-filters | Evaluate entire datasets — entry count, tag checks, bounds |
Filter groups let you compose multiple filters into compound AND/OR logic, and groups can nest for arbitrarily complex conditions. When you need filters to drive data flow directly, the Uber Filter nodes are the primary consumers: one variant keeps or discards entire collections based on point pass rates, while another routes points to the first matching branch in priority order.
For understanding filter architecture, composition patterns, and reusability: