从Python SQLite3查询中求和

Question

I would like to do some math with numbers I have stored in an SQLite database. My query seems to work. My issue is having python treat the numbers that I have queryed from the database as numbers.

The first definition selects a unique row from my database. The second definition cycles thru a number of rows in my database and adds them to a list of numbers which I would like to sum.

The problem occurs when I try to sum the numbers. Can anyone point me in the right direction?

Here is some of my code:

def query_symbol(symbolId):
    uniqueId = str(symbolId) + '@' + str(datetime.date.today())

    conn = sqlite3.connect('sql/databse.db')
    c = conn.cursor()
    with conn:
        c.execute("SELECT number FROM symbols WHERE uniqueId= ?", (uniqueId,))
    return c.fetchall()

def calc(filename):
    symbolIds = def_sec.load_symbolIds(filename)

    print(len(symbolIds))

    list = []

    for symbolId in symbolIds:
        data = query_symbol(symbolId)
        #data = data.replace('()','')
        list.append(data)
    print(list)

        total = sum(list)
        print(total)

calc('index_symbolids')

My error message looks like this:

65
[[(13506636000.0,)], [(20156784500.0,)], [(21361120000.0,)], [(4650564600.0,)], [(18572773200.0,)], [(13889340000.0,)], [(21911477100.0,)], [(19014765000.0,)], [(8592582800.0,)], [(12399850600.0,)], [(26021607500.0,)], [(17344514400.0,)], [(28396342200.0,)], [(10444843100.0,)], [(13894385900.0,)], [(26429184100.0,)], [(9193019800.0,)], [(18356516200.0,)], [(13693344800.0,)], [(39135783700.0,)], [(64988933000.0,)], [(52588381800.0,)], [(53514752300.0,)], [(8205312900.0,)], [(18563139800.0,)], [(34542681400.0,)], [(10626282600.0,)], [(14568874300.0,)], [(52083201800.0,)], [(21204153700.0,)], [(13380654000.0,)], [(24821311300.0,)], [(8232241800.0,)], [(148515191500.0,)], [(31669795700.0,)], [(97989223400.0,)], [(135145143799.0,)], [(178696200.0,)], [(9474728600.0,)], [(77661549000.0,)], [(33649778800.0,)], [(10061871500.0,)], [(23682872900.0,)], [(5196629500.0,)], [(54706667400.0,)], [(13934478600.0,)], [(5141383100.0,)], [(81343002200.0,)], [(16173162200.0,)], [(17649907400.0,)], [(32514907200.0,)], [(9783995600.0,)], [(75825589800.0,)], [(6205111500.0,)], [(53908007900.0,)], [(7615559400.0,)], [(17484345800.0,)], [(16072715900.0,)], [(53990182900.0,)], [(25798084100.0,)], [(28311485300.0,)], [(7296894200.0,)], [(19297000000.0,)], [(13271169800.0,)], [(22862203000.0,)]]
Traceback (most recent call last):
  File "/Users/michael/atomProjects/calc.py", line 53, in <module>
    index_calc('index_symbolids')
  File "/Users/michael/atomProjects/calc.py", line 49, in calc
    total = sum(list)
TypeError: unsupported operand type(s) for +: 'int' and 'list'
[Finished in 0.822s]

Answer 1

The problem occurs when I try to sum the numbers.

No, you are trying to sum list of lists.

total = sum(map(lambda x: x[0][0], list)

Also, you should avoid using list as a variable because it shadows the builtin list .

Answer 2

I saw two problems. Let's take a look at your list which you printed out:

[[(13506636000.0,)], [(20156784500.0,)],  ... ]

What it told me is you have a list of nested lists (eg [(13506636000.0,)] ). This represents the a single row and a single column, because you return c.fetchall() . In this case, you know for sure that at most 1 row of data is returned, so we can use c.fetchone() to reduce the number of nested lists.

Next, take a look at this line:

        data = query_symbol(symbolId)

Since c.fetchone() returns (13506636000.0,) , which is a tuple of 1 element, you can use the following trick to extract the number:

        (data,) = query_symbol(symbolId)

Putting it together:

def query_symbol(symbolId):
    uniqueId = '{}@{}'.format(symbolId, datetime.date.today())

    with sqlite3.connect('sql/database.db') as conn:
        c = conn.execute(
            "SELECT number FROM symbols WHERE uniqueId=?",
             (uniqueId,))
    return c.fetchone()

def calc(filename):
    symbolIds = def_sec.load_symbolIds(filename)

    print(len(symbolIds))

    numbers = []

    for symbolId in symbolIds:
        (data, ) = query_symbol(symbolId)  # instead of `data =`
        numbers.append(data)

    print(numbers)
    total = sum(numbers)
    print(total)

calc('index_symbolids')

At this point, you should get the desired result.

Update

This update explains how (data,) = works. Let say, if you have a tuple of 3 elements and want to assign to variables a , b , c :

(a, b, c) = (1, 2, 3)  # a=1, b=2, c=3

In practice, we can drop the parentheses from the left side of the equal sign:

a, b, c = (1, 2, 3)  # a=1, b=2, c=3

For a 1-element tuple, the notation is (1,) , not (1) (which is simply 1 ):

data = (1,)     # data = (1,), which is the whole tuple
(data,) = (1,)  # data = 1, which is what we want
data, = (1,)    # data = 1, but the syntax looks kind of wrong

Back to your problem, the function query_symbol() returns a 1-element tuple, thus we have to use:

(data,) = query_symbol(symbolId)
data, = query_symbol(symbolId)

I personally like the first syntax because the second looks like the comma was there by a mistake.