github/DB-GPT
1
0
Fork 0
mirror of https://github.com/csunny/DB-GPT.git synced 2025-09-08 20:39:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat: (0.6)New UI (#1855)

Browse Source 
Co-authored-by: 夏姜 <wenfengjiang.jwf@digital-engine.com>
Co-authored-by: aries_ckt <916701291@qq.com>
Co-authored-by: wb-lh513319 <wb-lh513319@alibaba-inc.com>
Co-authored-by: csunny <cfqsunny@163.com>
This commit is contained in:
明天
2024-08-21 17:37:45 +08:00
committed by GitHub
parent 3fc82693ba
commit b124ecc10b
824 changed files with 93371 additions and 2515 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
dbgpt/app/scene/chat_data/chat_excel/excel_analyze/chat.py
View File

@@ -32,19 +32,15 @@ class ChatExcel(BaseChat):
- model_name:(str) llm model name
- select_param:(str) file path
"""
chat_mode = ChatScene.ChatExcel
self.select_param = chat_param["select_param"]
self.model_name = chat_param["model_name"]
chat_param["chat_mode"] = ChatScene.ChatExcel
if has_path(self.select_param):
self.excel_reader = ExcelReader(self.select_param)
else:
self.excel_reader = ExcelReader(
os.path.join(
KNOWLEDGE_UPLOAD_ROOT_PATH, chat_mode.value(), self.select_param
)
)
self.chat_param = chat_param
self.excel_reader = ExcelReader(
chat_param["chat_session_id"], self.select_param
)
self.api_call = ApiCall()
super().__init__(chat_param=chat_param)
@@ -65,9 +61,10 @@ class ChatExcel(BaseChat):
"chat_session_id": self.chat_session_id,
"user_input": "[" + self.excel_reader.excel_file_name + "]" + " Analyze！",
"parent_mode": self.chat_mode,
"select_param": self.excel_reader.excel_file_name,
"select_param": self.select_param,
"excel_reader": self.excel_reader,
"model_name": self.model_name,
"user_name": self.chat_param.get("user_name", None),
}
learn_chat = ExcelLearning(**chat_param)
result = await learn_chat.nostream_call()

2
dbgpt/app/scene/chat_data/chat_excel/excel_learning/chat.py
View File

@@ -19,6 +19,7 @@ class ExcelLearning(BaseChat):
select_param: str = None,
excel_reader: Any = None,
model_name: str = None,
user_name: str = None,
):
chat_mode = ChatScene.ExcelLearning
""" """
@@ -30,6 +31,7 @@ class ExcelLearning(BaseChat):
"current_user_input": user_input,
"select_param": select_param,
"model_name": model_name,
"user_name": user_name,
}
super().__init__(chat_param=chat_param)
if parent_mode:

72
dbgpt/app/scene/chat_data/chat_excel/excel_reader.py
View File

@@ -1,3 +1,5 @@
import io
import json
import logging
import os
@@ -17,6 +19,7 @@ from pyparsing import (
delimitedList,
)
from dbgpt.util.file_client import FileClient
from dbgpt.util.pd_utils import csv_colunm_foramt
from dbgpt.util.string_utils import is_chinese_include_number
@@ -227,28 +230,67 @@ def is_chinese(text):
class ExcelReader:
def __init__(self, file_path):
file_name = os.path.basename(file_path)
self.file_name_without_extension = os.path.splitext(file_name)[0]
encoding, confidence = detect_encoding(file_path)
logger.info(f"Detected Encoding: {encoding} (Confidence: {confidence})")
self.excel_file_name = file_name
self.extension = os.path.splitext(file_name)[1]
def __init__(self, conv_uid, file_param):
self.conv_uid = conv_uid
self.file_param = file_param
if isinstance(file_param, str) and os.path.isabs(file_param):
file_name = os.path.basename(file_param)
self.file_name_without_extension = os.path.splitext(file_name)[0]
encoding, confidence = detect_encoding(file_param)
self.excel_file_name = file_name
self.extension = os.path.splitext(file_name)[1]
file_info = file_param
else:
if isinstance(file_param, dict):
file_path = file_param.get("file_path", None)
if not file_path:
raise ValueError("Not find file path!")
else:
file_name = os.path.basename(file_path.replace(f"{conv_uid}_", ""))
else:
temp_obj = json.loads(file_param)
file_path = temp_obj.get("file_path", None)
file_name = os.path.basename(file_path.replace(f"{conv_uid}_", ""))
self.file_name_without_extension = os.path.splitext(file_name)[0]
self.excel_file_name = file_name
self.extension = os.path.splitext(file_name)[1]
file_client = FileClient()
file_info = file_client.read_file(
conv_uid=self.conv_uid, file_key=file_path
)
result = chardet.detect(file_info)
encoding = result["encoding"]
confidence = result["confidence"]
logger.info(
f"File Info:{len(file_info)},Detected Encoding: {encoding} (Confidence: {confidence})"
)
# read excel file
if file_path.endswith(".xlsx") or file_path.endswith(".xls"):
df_tmp = pd.read_excel(file_path, index_col=False)
if file_name.endswith(".xlsx") or file_name.endswith(".xls"):
df_tmp = pd.read_excel(file_info, index_col=False)
self.df = pd.read_excel(
file_path,
file_info,
index_col=False,
converters={i: csv_colunm_foramt for i in range(df_tmp.shape[1])},
)
elif file_path.endswith(".csv"):
df_tmp = pd.read_csv(file_path, index_col=False, encoding=encoding)
self.df = pd.read_csv(
file_path,
elif file_name.endswith(".csv"):
df_tmp = pd.read_csv(
file_info if isinstance(file_info, str) else io.BytesIO(file_info),
index_col=False,
encoding=encoding,
)
self.df = pd.read_csv(
file_info if isinstance(file_info, str) else io.BytesIO(file_info),
index_col=False,
encoding=encoding,
# csv_colunm_foramt 可以修改更多，只是针对美元人民币符号，假如是“你好¥¥¥”则会报错！
converters={i: csv_colunm_foramt for i in range(df_tmp.shape[1])},
)
else: