selfhostyourtech/reflector
2
0
Fork 0
mirror of https://github.com/Monadical-SAS/reflector.git synced 2025-12-20 20:29:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

hotfix/server: fix duplication of topics

Browse Source
This commit is contained in:
Mathieu Virbel
2023-11-02 19:40:45 +01:00
parent 37f6fe6345
commit 9642d0fd1e
3 changed files with 15 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
server/reflector/db/transcripts.py
View File

@@ -106,9 +106,9 @@ class Transcript(BaseModel):
return ev return ev
def upsert_topic(self, topic: TranscriptTopic): def upsert_topic(self, topic: TranscriptTopic):
existing_topic = next((t for t in self.topics if t.id == topic.id), None) index = next((i for i, t in enumerate(self.topics) if t.id == topic.id), None)
if existing_topic: if index is not None:
existing_topic.update_from(topic) self.topics[index] = topic
else: else:
self.topics.append(topic) self.topics.append(topic)

9
server/reflector/pipelines/main_live_pipeline.py
View File

@@ -46,7 +46,9 @@ from reflector.processors import (
TranscriptTranslatorProcessor, TranscriptTranslatorProcessor,
) )
from reflector.processors.types import AudioDiarizationInput from reflector.processors.types import AudioDiarizationInput
from reflector.processors.types import TitleSummary as TitleSummaryProcessorType from reflector.processors.types import (
TitleSummaryWithId as TitleSummaryWithIdProcessorType,
)
from reflector.processors.types import Transcript as TranscriptProcessorType from reflector.processors.types import Transcript as TranscriptProcessorType
from reflector.settings import settings from reflector.settings import settings
from reflector.ws_manager import WebsocketManager, get_ws_manager from reflector.ws_manager import WebsocketManager, get_ws_manager
@@ -163,6 +165,8 @@ class PipelineMainBase(PipelineRunner):
text=data.transcript.text, text=data.transcript.text,
words=data.transcript.words, words=data.transcript.words,
) )
if isinstance(data, TitleSummaryWithIdProcessorType):
topic.id = data.id
async with self.transaction(): async with self.transaction():
transcript = await self.get_transcript() transcript = await self.get_transcript()
await transcripts_controller.upsert_topic(transcript, topic) await transcripts_controller.upsert_topic(transcript, topic)
@@ -302,7 +306,8 @@ class PipelineMainDiarization(PipelineMainBase):
# XXX translation is lost when converting our data model to the processor model # XXX translation is lost when converting our data model to the processor model
transcript = await self.get_transcript() transcript = await self.get_transcript()
topics = [ topics = [
TitleSummaryProcessorType( TitleSummaryWithIdProcessorType(
id=topic.id,
title=topic.title, title=topic.title,
summary=topic.summary, summary=topic.summary,
timestamp=topic.timestamp, timestamp=topic.timestamp,

6
server/reflector/processors/types.py
View File

@@ -167,6 +167,10 @@ class TitleSummary(BaseModel):
return f"{minutes:02d}:{seconds:02d}.{milliseconds:03d}" return f"{minutes:02d}:{seconds:02d}.{milliseconds:03d}"
class TitleSummaryWithId(TitleSummary):
id: str
class FinalLongSummary(BaseModel): class FinalLongSummary(BaseModel):
long_summary: str long_summary: str
duration: float duration: float
@@ -386,4 +390,4 @@ class TranslationLanguages(BaseModel):
class AudioDiarizationInput(BaseModel): class AudioDiarizationInput(BaseModel):
audio_url: str audio_url: str
topics: list[TitleSummary] topics: list[TitleSummaryWithId]