I have a project built in Python 2.7, using SQLite3 as the da开发者_JAVA技巧tabase. I need to know how to load an item from a particular row and column into an existing Python variable.
TY!
Here are the basic steps:
import sqlite3
conn = sqlite3.connect(':memory:')
curs = conn.cursor()
results = curs.execute( """SELECT mycol
FROM mytable
WHERE somecol = ?;""", (some_var,) ).fetchall()
curs.close()
conn.close()
For further research you can look into using a context manager (with
statement), and how to fetch results into a dict. Here's an example:
with sqlite3.connect(':memory:') as conn:
curs = conn.cursor()
curs.row_factory = sqlite3.Row
try:
results = curs.execute( """SELECT mycol
FROM mytable
WHERE somecol = ?;""",
(some_var,) ).fetchall()
# you would put your exception-handling code here
finally:
curs.close()
The benefits of the context manager are many, including the fact that your connection is closed for you. The benefit of mapping the results in a dict is that you can access column values by name as opposed to a less-meaningful integer.
精彩评论