mirror of
https://github.com/angr/angr
synced 2026-08-17 12:23:11 -04:00
* Move serialized functions directly into LMDB on load. * Serialize the callgraph instead of rebuilding it on load. * Move serialized CFG nodes directly into LMDB on angrdb load. * Copy clean spilled functions directly from LMDB on dump. * Do not serialize empty variable managers. * Serialize CFG edges at the key level to avoid node materialization. * Bulk-insert function and variable rows via SQLAlchemy Core. * Skip the CFGNode.function_address fill when node records carry it. * struct-pack spilled edge data instead of protobuf. * Avoid O(n^2) full-LRU-order copy in spilling eviction. * Byte-copy clean spilled CFG nodes when serializing a CFG model. * Lint code. |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| test_db.py | ||
| test_pickle.py | ||
| test_serialization.py | ||
| test_vault.py | ||