refactor(podcasts): propose brief with min_seconds and max_seconds

This commit is contained in:
CREDO23 2026-06-16 23:38:28 +02:00
parent d0ed5b94d9
commit af08e2f033

View file

@ -4,11 +4,12 @@ from __future__ import annotations
from sqlalchemy.ext.asyncio import AsyncSession
from app.podcasts.duration_limits import DEFAULT_MAX_SECONDS, DEFAULT_MIN_SECONDS
from app.podcasts.persistence import PodcastRepository
from app.podcasts.schemas import PodcastSpec
from app.podcasts.service import preferences_from
from .config import DEFAULT_MAX_MINUTES, DEFAULT_MIN_MINUTES, DEFAULT_SPEAKER_COUNT
from .config import DEFAULT_SPEAKER_COUNT
from .graph import graph as brief_graph
from .state import BriefState
@ -18,8 +19,8 @@ async def propose_brief(
*,
search_space_id: int,
speaker_count: int = DEFAULT_SPEAKER_COUNT,
min_minutes: int = DEFAULT_MIN_MINUTES,
max_minutes: int = DEFAULT_MAX_MINUTES,
min_seconds: int = DEFAULT_MIN_SECONDS,
max_seconds: int = DEFAULT_MAX_SECONDS,
focus: str | None = None,
) -> PodcastSpec:
"""Reuse the last-used language and voices, else English; return the spec."""
@ -29,8 +30,8 @@ async def propose_brief(
config = {
"configurable": {
"speaker_count": speaker_count,
"min_minutes": min_minutes,
"max_minutes": max_minutes,
"min_seconds": min_seconds,
"max_seconds": max_seconds,
"focus": focus,
"last_used_language": last_language,
"last_used_voices": last_voices,