diff --git a/rag/api_core.py b/rag/api_core.py index ee73e6f..1c36713 100644 --- a/rag/api_core.py +++ b/rag/api_core.py @@ -21,7 +21,7 @@ org 注入方式（session_env）：返回真实请求 _run_ns，get_userorgid/g """ import json import os -import uuid as _uuid +from appPublic.uniqueID import getID from ahserver.serverenv import ServerEnv from appPublic.dictObject import DictObject @@ -107,7 +107,7 @@ async def kb_create(env, ns): if emb not in _ENGINES: return _err("embedding_engine must be one of " + "/".join(_ENGINES)) org_id = await env.get_userorgid() - kb_id = _uuid.uuid4().hex + kb_id = getID() from rag.init import get_rags_base, ensure_kb_dir async with get_sor_context(env, 'rag') as sor: dup = await sor.sqlExe( @@ -289,7 +289,7 @@ async def tag_create(env, ns): if existing: r = existing[0] return _ok(tag_id=r.id, name=r.name, color=r.color, duplicate=True) - tag_id = _uuid.uuid4().hex + tag_id = getID() await sor.sqlExe( "INSERT INTO rag_tags (id, kb_id, name, color, org_id, created_at) " "VALUES (${id}$, ${k}$, ${n}$, ${c}$, ${o}$, NOW())", @@ -333,7 +333,7 @@ async def doc_set_tags(env, ns): if recs: tag_ids_in.append(recs[0].id) else: - tid = _uuid.uuid4().hex + tid = getID() await sor.sqlExe( "INSERT INTO rag_tags (id, kb_id, name, color, org_id, created_at) " "VALUES (${id}$, ${k}$, ${n}$, '#3b82f6', ${o}$, NOW())", @@ -366,7 +366,7 @@ async def doc_set_tags(env, ns): await sor.sqlExe( "INSERT INTO rag_media_tags (id, kb_id, media_type, media_id, tag_id, created_at) " "VALUES (${id}$, ${k}$, 'document', ${d}$, ${t}$, NOW())", - {"id": _uuid.uuid4().hex, "k": kb_id, "d": doc_id, "t": tid}) + {"id": getID(), "k": kb_id, "d": doc_id, "t": tid}) added += 1 await sor.sqlExe("COMMIT", {}) # 回读最终标签列表 @@ -411,7 +411,7 @@ async def doc_upload(env, ns, file_data, file_name): return _err(quota_msg, code="storage_quota_exceeded") rags_base = await get_rags_base(sor) - doc_id = _uuid.uuid4().hex + doc_id = getID() kb_dir = ensure_kb_dir(rags_base, org_id, kb_id) safe_name = file_name.replace('/', '_').replace('\\', '_') disk_name = doc_id[:8] + '_' + safe_name diff --git a/rag/init.py b/rag/init.py index 2345c9c..4837dbc 100644 --- a/rag/init.py +++ b/rag/init.py @@ -7,7 +7,8 @@ from ahserver.serverenv import ServerEnv from appPublic.registerfunction import RegisterFunction from appPublic.log import debug, exception from sqlor.dbpools import get_sor_context -import json, os, uuid, time +import json, os, time +from appPublic.uniqueID import getID async def status_handler(request, params_kw, *args, **kwargs): @@ -433,7 +434,7 @@ async def doc_upload_handler(request, params_kw, *args, **kwargs): "message": "存储配额超限：机构已用 " + _fmt_bytes(used) + "，限额 " + _fmt_bytes(quota_limit) + "，本文件 " + _fmt_bytes(file_size)}, ensure_ascii=False) # ---- 保存到机构 rags 目录：{workspace_base}/{org}/rags/{kb}/ ---- - doc_id = uuid.uuid4().hex + doc_id = getID() async with get_sor_context(env, 'rag') as sor: rags_base = await get_rags_base(sor) kb_dir = ensure_kb_dir(rags_base, userorgid, kb_id) @@ -875,7 +876,7 @@ async def dir_create_handler(request, params_kw, *args, **kwargs): ok, kb, msg = await check_kb_perm(env, sor, kb_id, 'maintain') if not ok: return json.dumps({"error": "kb_perm_denied", "message": msg}, ensure_ascii=False) - dir_id = uuid.uuid4().hex + dir_id = getID() await sor.sqlExe( "INSERT INTO rag_document_chunks (id, doc_id, kb_id, chunk_index, chunk_type, content, description, created_at) " "VALUES (${id}$, '', ${kb_id}$, 0, 'directory', ${name}$, ${parent}$, NOW())", @@ -952,7 +953,7 @@ async def tag_create_handler(request, params_kw, *args, **kwargs): if existing: return json.dumps({"status": "SUCCEEDED", "tag_id": existing[0].id, "name": name, "color": existing[0].color, "duplicate": True}, ensure_ascii=False) - tag_id = uuid.uuid4().hex + tag_id = getID() await sor.sqlExe( "INSERT INTO rag_tags (id, kb_id, name, color, org_id, created_at) " "VALUES (${id}$, ${kb_id}$, ${name}$, ${color}$, ${org_id}$, NOW())", @@ -1009,7 +1010,7 @@ async def tag_assign_handler(request, params_kw, *args, **kwargs): if media_type not in ("document", "face", "voice"): return json.dumps({"error": "media_type must be document/face/voice"}) async with get_sor_context(env, 'rag') as sor: - mt_id = uuid.uuid4().hex + mt_id = getID() await sor.sqlExe( "INSERT INTO rag_media_tags (id, kb_id, media_type, media_id, tag_id, created_at) " "VALUES (${id}$, ${kb_id}$, ${type}$, ${mid}$, ${tid}$, NOW())", @@ -1164,7 +1165,7 @@ async def tag_sync_handler(request, params_kw, *args, **kwargs): added = 0 for tid in wanted_ids: if tid not in current: - mt_id = uuid.uuid4().hex + mt_id = getID() await sor.sqlExe( "INSERT INTO rag_media_tags (id, kb_id, media_type, media_id, tag_id, created_at) " "VALUES (${id}$, ${kb_id}$, ${type}$, ${mid}$, ${tid}$, NOW())", @@ -1243,7 +1244,7 @@ async def engine_cfg_save_handler(request, params_kw, *args, **kwargs): from appPublic.jsonConfig import getConfig key = getConfig().password_key or 'QRIVSRHrthhwyjy176556332' enc = password(api_key_plain, key=key) if api_key_plain else "" - rid = uuid.uuid4().hex + rid = getID() await sor.sqlExe( "INSERT INTO rag_engine_configs (id, engine_type, engine_name, endpoint_url, api_key, " "model_name, is_default, priority, status, created_at, updated_at) " diff --git a/rag/pipeline.py b/rag/pipeline.py index ecf8ae3..111bd51 100644 --- a/rag/pipeline.py +++ b/rag/pipeline.py @@ -143,8 +143,8 @@ async def extract_voiceprint(file_data, file_name): async def process_upload(env, file_data, kb_id, folder_id, file_name): """Full upload pipeline — save + classify + process + DB record""" - import uuid as _uuid - doc_id = str(_uuid.uuid4()).hex[:16] + from appPublic.uniqueID import getID + doc_id = getID() ext = '.' + file_name.rsplit('.', 1)[1] if '.' in file_name else '.bin' saved_name = doc_id + ext file_path = '/d/rag/ragserver/pkgs/rag/rag/files/' + saved_name diff --git a/wwwroot/knowledge_bases_list/upload_file.dspy b/wwwroot/knowledge_bases_list/upload_file.dspy index efafe83..e2bb7c0 100644 --- a/wwwroot/knowledge_bases_list/upload_file.dspy +++ b/wwwroot/knowledge_bases_list/upload_file.dspy @@ -35,7 +35,7 @@ if used + file_size > quota_limit: web_path = await env.save_file(file_data, file_name) real_path = env.realpath(web_path) -doc_id = str(uuid()).replace('-', '')[:16] +doc_id = uuid() ext = '.' + file_name.rsplit('.', 1)[1] if '.' in file_name else '.bin' ext_l = ext.lower()