We Keep Coding

SQLlite3 Insert values to columns based on Where condition

I'm trying to insert values to new columns only for rows which answer a specific where condition (when a the value in Name column is equal to a given name).
This is the code I've written so far:
cours.execute("""INSERT INTO NewTable(Straight, Right, Left) WHERE Name = (?)
VALUES (?, ?, ?) """,
(name, DIRECTIONS['straight'], DIRECTIONS['right'], DIRECTIONS['left']))
conn.commit()
after I do the commit I get the following error:
sqlite3.OperationalError: near "WHERE": syntax error
Thanks.

Try with an UPDATE statement:
cours.execute("""UPDATE NewTable
SET Straight = ?, Right = ?, Left = ?
WHERE Name = ?""",
(DIRECTIONS['straight'], DIRECTIONS['right'], DIRECTIONS['left'], name))

python sqlite3 parameterization - insert throws no such column error

Insert in columns with parameterized query throws no such column error
First (working) example:
# unit test input
name = "issue_number_1"
text = "issue_text"
rating_sum = 0
if name:
# check if issue is already in db
with self.conn: # this should release the connection when finished
test = cursor.execute("SELECT name, text FROM issue WHERE name = ?", (name,))
data = test.fetchall()
print(data)
this is working and prints:
[('issue_number_1', 'issue_text')]
Second (non working) example:
# unit test input
name = "issue_number_2"
text = "issue_text"
rating_sum = 0
if name:
with self.conn:
sql_string = "INSERT INTO issue (name, text, rating_sum) VALUES (name = ?, text = ?, rating_sum = ?)"
cursor.execute(sql_string, (name, text, rating_sum,))
throws this error:
cursor.execute(sql_string, (name, text, rating_sum,))
sqlite3.OperationalError: no such column: name
the column name exists, the first example proofed that
the name: "issue_number_2" does not exist in the DB
the second example fails exactly same with only name to insert (only one parameter)
i had no problems inserting with string concatenation so the problem should be in my second example code somewhere

You need to add single quote.for example:
"INSERT INTO table (field) VALUES ('$1')"
add just values in second () and add single quote around string values.

After a lot of experiments i was a little bit confused....
This is the right syntax:
sql_string = "INSERT INTO issue (name, text, rating_sum) VALUES (?, ?, ?)"
cursor.execute(sql_string, (name, text, rating_sum,))

The statement:
INSERT INTO .... VALUES ....
is an SQL statement and the correct syntax is:
INSERT INTO tablename (col1, col2, ...) VALUES (expr1, expr2, ...)
where col1, col2, ... are columns of the table tablename and expr1, expr2, ... are expressions or literals that are evaluated and assigned to each of the columns col1, col2, ... respectively.
So the syntax that you use is not valid SQL syntax.
The assignment of the values is not performed inside VALUES(...).
The correct syntax to use in Python would be:
INSERT INTO issue (name, text, rating_sum) VALUES (?, ?, ?)

Pass many list parameter to a query in Python

I have three lists:
countries=['AUSTRALIA', 'FRANCE', 'ITALY']
segments=['little','medium','big','huge']
costumers = ['ERIK','FRANK','BOB', 'JANE']
I have to write something like this:
cursor.execute("SELECT COUNTRY, SEGMENT, CUSTOMER, VALUE
WHERE COUNTRY IN ?, SEGMENT IN ?, VALUE IN ?, (countries, segments, costumers))
Any idea? The three lists may vary in terms of number of components for each list.
Thanks in advance!

This can be solved in the following way:
cursor.execute("SELECT country, segment, customer, value" +
"WHERE country IN %(1)s, SEGMENT IN %(2)s, VALUE IN %(3)s",
{"1": tuple(countries), "2": tuple(segments), "3": tuple(customers)})
This ensures you still have the protection from SQL injections you had earlier. It also doesn't fail on strings.

You can do something like this:
countries=['AUSTRALIA', 'FRANCE', 'ITALY']
segments=['little','medium','big','huge']
costumers = ['ERIK','FRANK','BOB', 'JANE']
query = ("SELECT COUNTRY, SEGMENT, CUSTOMER, VALUE "
" WHERE COUNTRY IN ('{}') "
" AND SEGMENT IN ('{}') "
" AND VALUE IN ('{}')").format(countries, segments, costumers)
cursor.execute(query)

This might look like more code than the other solutions which directly pass in values using string formatting, but it is much safer. Read up on SQL injection to understand why.
The difficulty in passing values as parameters (much safer than string formatting) is that the query has to look something like this before you can pass in parameters (the number of ?s varies depending on number of parameters to be passed in):
"SELECT COUNTRY, SEGMENT, CUSTOMER, VALUE FROM SOME_TABLE WHERE COUNTRY IN (?, ?, ?) AND SEGMENT IN (?, ?, ?) AND VALUE IN (?, ?, ?)"
So you need to use string formatting to pass the appropriate number of ? placeholders:
countries=['AUSTRALIA', 'FRANCE', 'ITALY']
segments=['little','medium','big','huge']
customers = ['ERIK','FRANK','BOB', 'JANE']
country_placeholders = ','.join('?' for i in range(len(countries)))
segment_placeholders = ','.join('?' for i in range(len(segments)))
customer_placeholders = ','.join('?' for i in range(len(customers)))
query = """
SELECT COUNTRY, SEGMENT, CUSTOMER, VALUE FROM SOME_TABLE WHERE COUNTRY IN ({}) AND SEGMENT IN ({}) AND VALUE IN ({})
""".format(country_placeholders, segment_placeholders, customer_placeholders)
This gives you the following query template:
"SELECT COUNTRY, SEGMENT, CUSTOMER, VALUE FROM SOME_TABLE WHERE COUNTRY IN (?,?,?) AND SEGMENT IN (?,?,?,?) AND VALUE IN (?,?,?,?)"
Now you just need to execute and pass your parameters:
params = tuple(countries + segments + customers)
cursor.execute(query, params)

Pass variable into SQLAlchemy query

Trying to pass a variable (a set) into an sqlalchemy query.
Found this: How can I bind a list to a parameter in a custom query in sqlalchemy? But it requires you to know many items there are. The number of entries changes at any given moment.
My previous question went mostly unanswered unfortunately so I figured I'd re-iterate what I'm trying to do here.
Basically, I have this variable: sites = set(db1).intersection(db2) and I'm trying to pass it into this sql alchemy query:
'test': DBSession.query(A_School.cis_site_id.in_(sites)).all(),
But I get invalid syntax errors and invalid parameter type errors...I can't get this thing to do what I want it to do. DB1 and DB2 are, as you mightve guessed, 2 different databases.
db1 = cis_db.query(site.site_id).join(site_tag).filter(site_tag.tag_id.like(202)).all()
db2 = DBSession.query(A_School.cis_site_id).all()
Full error:
ProgrammingError: (ProgrammingError) ('Invalid parameter type. param-index=0 param-type=KeyedTuple', 'HY105') u'SELECT [A_School].cis_site_id IN (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?) AS anon_1
FROM [A_School]' ((14639,), (14709,), (14587,), (14966,), (14625,), (14589,), (15144,), (15171,) ... displaying 10 of 18 total bound parameter sets ... (15133,), (14036,))

KeyedTuple is the type of each row returned by SQLAlchemy when not querying one full model. You are making sets of keyed tuples, rather than sets of the single value in each tuple. Should look something like this instead:
db1 = set(x.site_id for x in cis_db.query(site.site_id).join(site_tag).filter(site_tag.tag_id.like(202)))
db2 = set(x.cis_site_id for x in DBSession.query(A_School.cis_site_id))
sites = db1.intersection(db2)
test = DBSession.query(A_School).filter(A_School.cis_site_id.in_(sites)).all()

Assuming you would like to load schools for the sites, how about you try:
DBSession.query(A_School).filter(A_School.cis_site_id.in_(sites)).all()
instead of:
DBSession.query(A_School.cis_site_id.in_(sites)).all()

Inserting tuple's elements to database

I have a tuple that i wanna store its elements, I'm trying to insert it as following and it gives the following error, what am i doing wrong ? records_to_be_inserted is the tuple that has 8 elements.
with self.connection:
cur = self.connection.cursor()
cur.executemany("INSERT INTO rehberim(names, phone, mobile, email, \
photo, address, note, date) VALUES(?, ?, ?, ?, ?, ?, ?, ?)", self.records_to_be_inserTed)
Traceback (most recent call last):
File "/home/tayfun/workspace/personal_guide/modules/mainwindow.py", line 57, in save_records
photo, address, note, date) VALUES(?, ?, ?, ?, ?, ?, ?, ?)", self.records_to_be_inserTed)
sqlite3.ProgrammingError: Incorrect number of bindings supplied. The current statement uses 8, and there are 0 supplied.

Note that the executemany is for inserting multiple rows, e.g.,
import sqlite3
""" the table structure is:
create table tab
a char(1),
b char(2),
c char(3)
)
"""
conn = sqlite3.connect('C:\\test.db')
stmt = "insert into tab (a, b, c) values (?, ?, ?)"
cur = conn.cursor()
## many rows
vals = [('1','2','3'), ('2','3','4'), ('3','4','5')]
cur.executemany(stmt, vals)
cur.close()
This will result in three rows in the database. If it is because you have multiple values in one query, you need to format it!
Edit: Added formatting with dictionaries
By using the following approach you do not need to consider the order of the values in the format call because the key in the dictionary is mapping the value into the {key_word} placeholder.
values = {'a' : 'value_a',
'b' : 'value_b'}
stmt = "insert into tab (col_a, col_b) values ({a}, {b})".format(**values)

The query must have all the data ready to be inserted.
You are calling a function in the query, which i guess you want that provides the data but that wont work.
You need to pass all the data in variables or locate them in the tuple index (like: tuple_name[1], tuple_name[4], etc.)
Example:
myTuple = ['a','b','c','d','e','f','g']
cur.executemany("INSERT INTO rehberim(names, phone, mobile, email, \
photo, address, note, date) VALUES({0}, {1}, {2}, {3}, {4}, {5}, {6}" .format (myTuple[1], myTuple[2], myTuple[3], myTuple[4], myTuple[5], myTuple[6], myTuple[7])