Efficient way of specifying data type when importing .csv file to SQLite Table using Python?

Question

I am importing data from a .csv file with headers into a SQLite table. This is easy to do of course and I could simply do the following:

sqlite> .separator ","
sqlite> .import 2015.csv tennisresults2015

However if I do this then all the columns default to the 'TEXT' data type. Instead I want my numeric field to default to either the 'INTEGER' or 'REAL' data types, which ever is appropriate.

Now I realize I can define each column explicitly in Python when I create a table via the following lines of code:

conn = sqlite3.connect('tennisresults.db') #connection
c = conn.cursor()  #get a cursor object, all SQL commands are processed by it            
c.execute('CREATE TABLE tennisresults2015(ATP INTEGER, Location TEXT, Tournament TEXT)') #etc

However this .csv file has a lot of columns, as you can see from the first two lines of the file shown below:

ATP,Location,Tournament,Date,Series,Court,Surface,Round,Best of,Winner,Loser,WRank,LRank,WPts,LPts,W1,L1,W2,L2,W3,L3,W4,L4,W5,L5,Wsets,Lsets,Comment,B365W,B365L,EXW,EXL,LBW,LBL,PSW,PSL,MaxW,MaxL,AvgW,AvgL
1,Brisbane,Brisbane International,01/05/2015,ATP250,Outdoor,Hard,1st Round,3,Duckworth J.,Simon G.,125,21,430,1730,6,2,6,2,,,,,,,2,0,Completed,4.5,1.18,4.3,1.2,4.33,1.2,4.67,1.23,4.73,1.23,4.31,1.2

I was wondering if there is a more efficient way of doing this instead of me manually specifying each field name and data type in Python?

Answer 1

Sadly I think the answer is simply no, there is no built-in way to do what you're asking. You need to either:

• Explicitly use a create statement like the one you included in your question.
• Or, include the datatypes in the csv file, and write your own parsing mechanism to read them and exclude them from the actual import (which you would need to perform yourself instead of using the built in import.)