Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #79266

Re: Sqlite3 help

Show key headers only | View raw

Am 14.10.2014 15:36, schrieb Chuck:
> I am building a simple podcast program where I download all the data from a feed with feedparser and store the data in sqlite3.  I am spanking new to sqlite and database programming.  Should I create the database in the __init__ method of my class, or is that a bad idea.  I haven't done any designing that included databases.
>
> Thanks!
> Chuck
>
When I first answered this question, I sent the answer erroneously to 
the OP and not to the list. So here it is again.

On Tue, Oct 14, 2014 at 1:19 PM, Sibylle Koczian wrote:
 >
 > As I don't know anything at all about podcasts and feeds, I'll just
 > answer your question about creating the database in the __init__
 > method of your class:
 >
 > That would destroy the existing database and create a new one without
 > any data. If you don't want to keep data from former runs of your
 > application, this wouldn't be necessary, because you could just
 > delete the data and keep the database. But most probably you want to
 > keep your data and add to it. In this case, you certainly don't want
 > to create a new database.
 >
 > The __init__ method of your class might be the right place to open
 > the database, though.
 >
 > You know that the documentation for the sqlite3 module is part of the
 > Python documentation? It contains links to the SQLite web page and to
 > a website with beginner information about SQL itself. So that should
 > help to get you started.

Am 15.10.2014 18:51, schrieb Chuck Johnson:
 >
 > I was thinking that I could fix that by using ' ' ' CREATE TABLE IF
 > NOT EXISTS ' ' '   Should I make the sqlite3.connect() command a
 > global variable?  I am not sure about how to design this properly.
 >
That should work. I wouldn't make the connection a global variable. Use 
it as a parameter for every function that needs it. Roughly like this:

import sqlite3
# other imports as needed

def one_of_my_functions(myconn, my_other_parameters):
     mycurs = myconn.cursor()
     try:
         # do things with mycurs and my_other_parameters
     finally:
         mycurs.close()
         myconn.commit()

# more of the same ...

def main():
     conn = sqlite3.connect('path/to/my/database')
     try:
         one_of_my_functions(conn, params)
         # more of the same
     finally:
         conn.close()

if __name__ == "__main__":
     main()


Or, if your program gets bigger, you might start and stop a connection 
inside a function instead of keeping one connection open for all of the 
application.

HTH
Sibylle

Back to comp.lang.python | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Sqlite3 help Chuck <galois271@gmail.com> - 2014-10-14 06:36 -0700
  Re: Sqlite3 help Sibylle Koczian <nulla.epistola@web.de> - 2014-10-18 12:27 +0200
    Re: Sqlite3 help Chuck <galois271@gmail.com> - 2014-10-20 19:48 -0700

csiph-web