# Runtime patch query evidence

Same temporary post-render diagnostic on both revisions; WeakMap IDs identify actual patch query atom objects. Indices map to file01.txt through file06.txt. Scope omits environment UUID and workspace path.

| Interaction | Base query IDs before → after | Head query IDs before → after |
|---|---|---|
| Edit file01 | `[1, 2, 3, 4, 5, 6]` → `[7, 8, 9, None, None, None]` | `[1, 2, 3, 4, 5, 6]` → `[1, 2, 3, 4, 5, 6]` |
| No-edit control | `[7, 8, 9, 10, 11, 12]` → `[7, 8, 9, 10, 11, 12]` | `[1, 2, 3, 4, 5, 6]` → `[1, 2, 3, 4, 5, 6]` |
| Edit file02 | `[7, 8, 9, 10, 11, 12]` → `[13, 14, 15, None, None, None]` | `[1, 2, 3, 4, 5, 6]` → `[1, 2, 3, 4, 5, 6]` |

On head, edited file01 and file02 grow from 2,001 to 2,002 parsed patch rows and receive new per-file patch hashes, while their query IDs remain stable. The rendered added line 1001 is also captured for each file.

Query retention and successful patch payloads are measured. Native paint counts and release frame times are not measured.