From c14ae92fead149703a8943fa1d537774bbb66121 Mon Sep 17 00:00:00 2001 From: Teknium <127238744+teknium1@users.noreply.github.com> Date: Wed, 2 Sep 2026 20:44:33 -0700 Subject: [PATCH] refactor(agent/context_compressor): fold derived-budget cache resets --- agent/context_compressor.py | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) diff --git a/agent/context_compressor.py b/agent/context_compressor.py index 8116b4f526..12fb107be7 100644 --- a/agent/context_compressor.py +++ b/agent/context_compressor.py @@ -1887,9 +1887,7 @@ class ContextCompressor(MicroCompactionMixin, ContextEngine): _base = getattr(self, "_base_threshold_percent", None) if _base is not None: self.threshold_percent = self._effective_threshold_percent(value, _base) - self._threshold_tokens = None - self._tail_token_budget = None - self._max_summary_tokens = None + self._threshold_tokens = self._tail_token_budget = self._max_summary_tokens = None self._emit_init_summary_once() @property @@ -2458,12 +2456,12 @@ class ContextCompressor(MicroCompactionMixin, ContextEngine): self.abort_on_summary_failure = abort_on_summary_failure # Micro-compaction is OFF by default: each pass breaks the prompt-cache prefix every turn. - self._micro_compact_enabled: bool = False - self._micro_compact_cursor: int = 0 - self._micro_compact_rolling_summary: str = "" - self._micro_compact_consecutive_failures: int = 0 - self._micro_compact_last_failure_cursor: int = -1 - self._micro_compact_defrag_threshold_tokens: int = 2000 + self._micro_compact_enabled = False + self._micro_compact_cursor = 0 + self._micro_compact_rolling_summary = "" + self._micro_compact_consecutive_failures = 0 + self._micro_compact_last_failure_cursor = -1 + self._micro_compact_defrag_threshold_tokens = 2000 # Set when _defrag_rolling_summary pops _DB_PERSISTED_MARKER in place; finalize_turn resets the flush cursor. self._flush_scan_cursor_invalidated: bool = False self._micro_compact_passes: int = 0 @@ -2477,9 +2475,7 @@ class ContextCompressor(MicroCompactionMixin, ContextEngine): self._config_context_length = config_context_length self._configured_threshold_percent = self.threshold_percent self._resolved_context_length: int | None = None - self._threshold_tokens: int | None = None - self._tail_token_budget: int | None = None - self._max_summary_tokens: int | None = None + self._threshold_tokens = self._tail_token_budget = self._max_summary_tokens = None self.compression_count = 0 # The init log reports resolved budgets; emit it on first resolution to keep construction non-blocking.