curo1305
02bcbb9143
- openai_provider.py: singleton self._client=AsyncOpenAI(...) in __init__ (D-07),
_truncate() 60/40 smart truncation (D-13), removed MAX_AI_CHARS constant,
changed __init__ signature to include context_chars, removed _client() method
- generic_openai_provider.py: new class, subclasses OpenAIProvider, conditional
response_format={"type":"json_object"} on supports_json_mode flag (D-01/D-02),
imports parse_classification + parse_suggestions from ai.utils (D-02 contract)
- ollama_provider.py: added context_chars kwarg with default 8000, passes through
- lmstudio_provider.py: added context_chars kwarg with default 8000, passes through
- classifier.py: removed MAX_AI_CHARS constant and text[:MAX_AI_CHARS] slices;
truncation now handled inside each provider via _truncate()
- requirements.txt: bumped anthropic floor to >=0.95.0 (D-03 output_config support)
91 lines
3.2 KiB
Python
91 lines
3.2 KiB
Python
from __future__ import annotations
|
|
|
|
from openai import AsyncOpenAI
|
|
from ai.base import AIProvider, ClassificationResult
|
|
from ai.utils import parse_classification, parse_suggestions
|
|
|
|
|
|
class OpenAIProvider(AIProvider):
|
|
def __init__(
|
|
self,
|
|
api_key: str,
|
|
model: str = "gpt-4o",
|
|
base_url: str | None = None,
|
|
context_chars: int = 8000,
|
|
):
|
|
self._api_key = api_key or "not-needed"
|
|
self._model = model
|
|
self._base_url = base_url
|
|
self._context_chars = context_chars
|
|
# Singleton: created once in __init__, reused for all calls on this instance.
|
|
# Do NOT recreate per API call — AsyncOpenAI wraps an httpx.AsyncClient
|
|
# that maintains a connection pool; recreating per call destroys pool reuse
|
|
# and forces a new TLS handshake per request (D-07 / RESEARCH.md).
|
|
self._client = AsyncOpenAI(api_key=self._api_key, base_url=self._base_url)
|
|
|
|
def _truncate(self, text: str) -> str:
|
|
"""D-13 smart truncation: first 60% + last 40% of context window.
|
|
|
|
Captures both document introduction and conclusion, which carry the
|
|
most topic signal for long documents.
|
|
"""
|
|
if len(text) <= self._context_chars:
|
|
return text
|
|
head_len = int(self._context_chars * 0.6)
|
|
tail_len = self._context_chars - head_len
|
|
return text[:head_len] + "\n[...truncated...]\n" + text[-tail_len:]
|
|
|
|
async def classify(
|
|
self,
|
|
document_text: str,
|
|
existing_topics: list[str],
|
|
system_prompt: str,
|
|
) -> ClassificationResult:
|
|
topics_str = ", ".join(existing_topics) if existing_topics else "(none yet)"
|
|
user_msg = (
|
|
f"Existing topics: [{topics_str}]\n\n"
|
|
f"Document text:\n{self._truncate(document_text)}"
|
|
)
|
|
response = await self._client.chat.completions.create(
|
|
model=self._model,
|
|
max_tokens=1024,
|
|
messages=[
|
|
{"role": "system", "content": system_prompt},
|
|
{"role": "user", "content": user_msg},
|
|
],
|
|
)
|
|
raw = response.choices[0].message.content or ""
|
|
return parse_classification(raw)
|
|
|
|
async def suggest_topics(
|
|
self,
|
|
document_text: str,
|
|
system_prompt: str,
|
|
) -> list[str]:
|
|
user_msg = (
|
|
"Suggest 3-5 topic names for this document. "
|
|
"Return ONLY valid JSON: {\"suggested_topics\": [\"topic1\", \"topic2\"]}\n\n"
|
|
f"Document text:\n{self._truncate(document_text)}"
|
|
)
|
|
response = await self._client.chat.completions.create(
|
|
model=self._model,
|
|
max_tokens=256,
|
|
messages=[
|
|
{"role": "system", "content": system_prompt},
|
|
{"role": "user", "content": user_msg},
|
|
],
|
|
)
|
|
raw = response.choices[0].message.content or ""
|
|
return parse_suggestions(raw)
|
|
|
|
async def health_check(self) -> bool:
|
|
try:
|
|
await self._client.chat.completions.create(
|
|
model=self._model,
|
|
max_tokens=5,
|
|
messages=[{"role": "user", "content": "ping"}],
|
|
)
|
|
return True
|
|
except Exception:
|
|
return False
|