时间戳

首页 > 解决方案 > Streamlit - 在线程中创建的 SQLite 对象只能在同一线程中使用

python - Streamlit - 在线程中创建的 SQLite 对象只能在同一线程中使用

问题描述

我使用Streamlit作为我的数据处理和可视化框架。

遵循有关使用 Streamlit 进行高级缓存的文档，关于如何创建数据库连接并将它们传递给其他函数，我正在使用sqlite3并尝试在运行时创建一个连接并传递“光标”，方法是：

import streamlit as st

# create database connection
@st.cache(allow_output_mutation=True)
def get_database_connection():
    conn = sqlite3.connect('collect/Data/fpl.db')
    c = conn.cursor()   
    return c

在这里，我尝试通过 传递我的连接 ID hash_funcs，返回创建的光标：

# pass connections around
@st.cache(allow_output_mutation=True, hash_funcs={sqlite3.Cursor:id})
def get_teams_basic(c):
    df_teams = sql('SELECT * FROM TeamsBasic', c)
    return df_teams

@st.cache(allow_output_mutation=True, hash_funcs={sqlite3.Cursor:id})
def get_players_basic(c):   
    df_player_basic = sql('SELECT * FROM PlayersBasic', c)
    return df_player_basic

# and so on...

这是我的sql()：

def sql(query, cursor):
    '''
    Takes an SQL query string, and outputs a
    dataframe representation of the query result.
    '''
    # Execute the sql query
    cursor.execute(query)

    # Get the query into a dataframe and set columns
    df_temp = pd.DataFrame(cursor.fetchall())
    df_temp.columns = [x[0] for x in cursor.description]

    # Set the sql id as the dataframe index
    index_column = df_temp.columns[0]
    df_temp.set_index(index_column, drop=True, inplace=True)

    return df_temp

然后，在 处main()，我实例化数据库连接并将光标传递给我的函数：

def main():
    c = get_database_connection()
    teams(c)
    players(c)

if __name__ == '__main__':
  main()

但是当代码到达第二个连接时，players()我收到以下错误：

ProgrammingError: SQLite objects created in a thread can only be used in that same thread. The object was created in thread id 123145444151296 and this is thread id 123145454661632.

追溯：

File "/Users/me/anaconda2/envs/data_science/lib/python3.7/site-packages/streamlit/ScriptRunner.py", line 322, in _run_script
    exec(code, module.__dict__)
File "/Volumes/Dados/Documents/Code/Apps/app.py", line 280, in <module>
    main()
File "/Volumes/Dados/Documents/Code/Apps/app.py", line 273, in main
    managers(c)
File "/Volumes/Dados/Documents/Code/Apps/app.py", line 149, in managers
    df_players = get_players_basic(c)
File "/Users/me/anaconda2/envs/data_science/lib/python3.7/site-packages/streamlit/caching.py", line 594, in wrapped_func
    return get_or_create_cached_value()
File "/Users/me/anaconda2/envs/data_science/lib/python3.7/site-packages/streamlit/caching.py", line 578, in get_or_create_cached_value
    return_value = func(*args, **kwargs)
File "/Volumes/Dados/Documents/Code/Apps/app.py", line 64, in get_players_basic
    df_player_basic = sql('SELECT * FROM PlayersBasic', c)
File "/Volumes/Dados/Documents/Code/Apps/collect/PythonFunctions/sqlfunctions.py", line 111, in sql
    cursor.execute(query)

我错过了什么？

标签： pythonsqlitestreamlit

解决方案

我不确定你是否还在解决这个问题，但如果你是，我发现了一个可能有用的关于这个主题的讨论线程。我很快就会在我的一个项目上进行尝试。

https://discuss.streamlit.io/t/prediction-analysis-and-creating-a-database/3504

推荐阅读