dearwolf/aider
1
0
Fork 0
mirror of https://github.com/Aider-AI/aider.git synced 2025-06-01 10:14:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

fix: Warm cache only when add_cache_headers is True

Browse source
This commit is contained in:
Paul Gauthier 2024-08-26 15:40:02 -07:00 • committed by Paul Gauthier (aider)
parent 1472fc56cb
commit 8cec00eb9c
1 changed files with 20 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

25
aider/coders/base_coder.py
View file

@ -985,19 +985,34 @@ class Coder:
return chunks return chunks
def warm_cache(self, chunks): def warm_cache(self, chunks):
if self.add_cache_headers: if not self.add_cache_headers:
chunks.add_cache_control_headers() return
if self.cache_warming_thread and self.cache_warming_thread.is_alive(): if self.cache_warming_thread and self.cache_warming_thread.is_alive():
self.cache_warming_thread.cancel() self.cache_warming_thread.cancel()
def warm_cache_worker(): def warm_cache_worker():
for _ in range(5): for _ in range(5):
time.sleep(10) # 290 == 4 minutes and 50 seconds
try: try:
self.send(chunks.cacheable_messages(), stream=False) completion = litellm.completion(
except retry_exceptions() as err: model=self.main_model.name,
messages=chunks.cacheable_messages(),
stream=False,
max_tokens=1,
extra_headers=self.main_model.extra_headers,
)
except Exception as err:
self.io.tool_error(f"Cache warming error: {str(err)}") self.io.tool_error(f"Cache warming error: {str(err)}")
time.sleep(290) # 4 minutes and 50 seconds continue
cache_hit_tokens = getattr(completion.usage, "prompt_cache_hit_tokens", 0) or getattr(
completion.usage, "cache_read_input_tokens", 0
)
self.io.tool_output(f"Warmed {cache_hit_tokens} cached tokens.")
self.io.tool_output("Stopped warming.")
self.cache_warming_thread = threading.Timer(0, warm_cache_worker) self.cache_warming_thread = threading.Timer(0, warm_cache_worker)
self.cache_warming_thread.start() self.cache_warming_thread.start()