Losita/mai-bot
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

ref：解耦表情包识别和图片识别

Browse Source
This commit is contained in:
SengokuCola
2025-11-26 13:11:45 +08:00
parent ac51f0c41d
commit 98d6775e62
3 changed files with 139 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files

115
src/chat/utils/utils_image.py
View File

@@ -12,7 +12,7 @@ from rich.traceback import install
from src.common.logger import get_logger
from src.common.database.database import db
from src.common.database.database_model import Images, ImageDescriptions
from src.common.database.database_model import Images, ImageDescriptions, EmojiDescriptionCache
from src.config.config import global_config, model_config
from src.llm_models.utils_model import LLMRequest
@@ -40,7 +40,7 @@ class ImageManager:
try:
db.connect(reuse_if_open=True)
db.create_tables([Images, ImageDescriptions], safe=True)
db.create_tables([Images, ImageDescriptions, EmojiDescriptionCache], safe=True)
except Exception as e:
logger.error(f"数据库连接或表创建失败: {e}")
@@ -49,6 +49,11 @@ class ImageManager:
except Exception as e:
logger.warning(f"数据库清理失败: {e}")
try:
self._cleanup_emoji_from_image_descriptions()
except Exception as e:
logger.warning(f"清理ImageDescriptions中的emoji记录失败: {e}")
self._initialized = True
def _ensure_image_dir(self):
@@ -119,6 +124,31 @@ class ImageManager:
else:
logger.info("[清理完成] 未发现无效描述记录")
@staticmethod
def _cleanup_emoji_from_image_descriptions():
"""清理Images和ImageDescriptions表中type为emoji的记录（已迁移到EmojiDescriptionCache）"""
try:
# 清理Images表中type为emoji的记录
deleted_images = Images.delete().where(Images.type == "emoji").execute()
# 清理ImageDescriptions表中type为emoji的记录
deleted_descriptions = (
ImageDescriptions.delete().where(ImageDescriptions.type == "emoji").execute()
)
total_deleted = deleted_images + deleted_descriptions
if total_deleted > 0:
logger.info(
f"[清理完成] 从Images表中删除 {deleted_images} 条emoji类型记录, "
f"从ImageDescriptions表中删除 {deleted_descriptions} 条emoji类型记录, "
f"共删除 {total_deleted} 条记录"
)
else:
logger.info("[清理完成] Images和ImageDescriptions表中未发现emoji类型记录")
except Exception as e:
logger.error(f"清理Images和ImageDescriptions中的emoji记录时出错: {str(e)}")
raise
async def get_emoji_tag(self, image_base64: str) -> str:
from src.chat.emoji_system.emoji_manager import get_emoji_manager
@@ -135,7 +165,7 @@ class ImageManager:
return f"[表情包：{tag_str}]"
async def get_emoji_description(self, image_base64: str) -> str:
"""获取表情包描述，优先使用Emoji表中的缓存数据"""
"""获取表情包描述，优先使用EmojiDescriptionCache表中的缓存数据"""
try:
# 计算图片哈希
# 确保base64字符串只包含ASCII字符
@@ -158,10 +188,19 @@ class ImageManager:
except Exception as e:
logger.debug(f"查询EmojiManager时出错: {e}")
# 查询ImageDescriptions表的缓存描述
if cached_description := self._get_description_from_db(image_hash, "emoji"):
logger.info(f"[缓存命中] 使用ImageDescriptions表中的描述: {cached_description[:50]}...")
return f"[表情包：{cached_description}]"
# 查询EmojiDescriptionCache表的缓存（包含描述和情感标签）
try:
cache_record = EmojiDescriptionCache.get_or_none(EmojiDescriptionCache.emoji_hash == image_hash)
if cache_record:
# 优先使用情感标签，如果没有则使用详细描述
if cache_record.emotion_tags:
logger.info(f"[缓存命中] 使用EmojiDescriptionCache表中的情感标签: {cache_record.emotion_tags[:50]}...")
return f"[表情包：{cache_record.emotion_tags}]"
elif cache_record.description:
logger.info(f"[缓存命中] 使用EmojiDescriptionCache表中的描述: {cache_record.description[:50]}...")
return f"[表情包：{cache_record.description}]"
except Exception as e:
logger.debug(f"查询EmojiDescriptionCache时出错: {e}")
# === 二步走识别流程 ===
@@ -221,45 +260,35 @@ class ImageManager:
logger.debug(f"[emoji识别] 详细描述: {detailed_description[:50]}... -> 情感标签: {final_emotion}")
if cached_description := self._get_description_from_db(image_hash, "emoji"):
logger.warning(f"虽然生成了描述，但是找到缓存表情包描述: {cached_description}")
return f"[表情包：{cached_description}]"
# 保存表情包文件和元数据（用于可能的后续分析）
logger.debug(f"保存表情包: {image_hash}")
current_timestamp = time.time()
filename = f"{int(current_timestamp)}_{image_hash[:8]}.{image_format}"
emoji_dir = os.path.join(self.IMAGE_DIR, "emoji")
os.makedirs(emoji_dir, exist_ok=True)
file_path = os.path.join(emoji_dir, filename)
# 再次检查缓存（防止并发情况下其他线程已经保存）
try:
# 保存文件
with open(file_path, "wb") as f:
f.write(image_bytes)
# 保存到数据库 (Images表) - 包含详细描述用于可能的注册流程
try:
img_obj = Images.get((Images.emoji_hash == image_hash) & (Images.type == "emoji"))
img_obj.path = file_path
img_obj.description = detailed_description # 保存详细描述
img_obj.timestamp = current_timestamp
img_obj.save()
except Images.DoesNotExist: # type: ignore
Images.create(
image_id=str(uuid.uuid4()),
emoji_hash=image_hash,
path=file_path,
type="emoji",
description=detailed_description, # 保存详细描述
timestamp=current_timestamp,
vlm_processed=True,
)
cache_record = EmojiDescriptionCache.get_or_none(EmojiDescriptionCache.emoji_hash == image_hash)
if cache_record and cache_record.emotion_tags:
logger.warning(f"虽然生成了描述，但是找到缓存表情包情感标签: {cache_record.emotion_tags}")
return f"[表情包：{cache_record.emotion_tags}]"
except Exception as e:
logger.error(f"保存表情包文件或元数据失败: {str(e)}")
logger.debug(f"再次查询EmojiDescriptionCache时出错: {e}")
# 保存最终的情感标签到缓存 (ImageDescriptions表)
self._save_description_to_db(image_hash, final_emotion, "emoji")
# 保存识别出的详细描述和情感标签到 emoji_description_cache
try:
current_timestamp = time.time()
cache_record, created = EmojiDescriptionCache.get_or_create(
emoji_hash=image_hash,
defaults={
"description": detailed_description,
"emotion_tags": final_emotion,
"timestamp": current_timestamp,
},
)
if not created:
# 更新已有记录
cache_record.description = detailed_description
cache_record.emotion_tags = final_emotion
cache_record.timestamp = current_timestamp
cache_record.save()
logger.info(f"[缓存保存] 表情包描述和情感标签已保存到EmojiDescriptionCache: {image_hash[:8]}...")
except Exception as e:
logger.error(f"保存表情包描述和情感标签缓存失败: {str(e)}")
return f"[表情包：{final_emotion}]"