python-noob - which container is appropriate for later exporting into mySql + matplotlib ?

[someone]

On 2013-04-11 19:58, Cousin Stanley wrote:
> someone wrote:
>
>> ....
>> I want to put this table into an appropriate container
>> such that afterwards I want to:
>>
>> 1) Put the data into a mySql-table
>> ....
>
>    You might consider using sqlite3 as a database manager
>    since it is  "batteries included"  with python ....
>
>    The stand-alone sqlite interpreter can first be used
>    to create an empty database named some.sql3
>    and create a table named  xdata  in that data base ....
>
>      sqlite3 some.sql3 '.read xdata_create.sql'
>
>    where the file  xdata_create.sql  contains ....
>
>      create table xdata
>      (
>       xdate  integer ,
>       xtime  integer ,
>       col1   real ,
>       col2   integer ,
>       col3   integer ,
>       col4   real ,
>       col5   integer ,
>       col6   integer
>      ) ;

Oh, thank you very much! Now I understand this (I haven't really worked 
much with sql before, so this was/is new to me, thanks!).

>    The csv data file can then be inserted into the xdata table
>    in the  some.sql3  database via python ....
........ and .......
> # python data selection example
> # for column 4 between 8 and 9

I combined both code snippets into:

==============================
#!/usr/bin/python

import sqlite3 as DBM
import ipdb

fs = open( 'some.csv' )
ls = [ ]
dbc = DBM.connect( 'some.sql3' )
cur = dbc.cursor()
if 0:
     sql = 'insert into xdata values( ? , ? , ? , ? , ? , ? , ? , ? )'
     for row in fs :
         dt, col1, col2, col3, col4,col5, col6 = row.strip().split(',' )
         xdate , xtime = dt.split( 'T' )
         xtuple = ( xdate, xtime, col1, col2, col3, col4, col5, col6 )
         cur.execute( sql , xtuple )
     dbc.commit()
else:
     list_sql = [
       'select xtime , col4' ,
       'from   xdata' ,
       'where  xtime >= 80000  and  xtime <= 90000 ; ' ]
     str_sql = '\n'.join( list_sql )
     cur.execute( str_sql )
     for row in cur :
         #ipdb.set_trace()
         # I get: TypeError: "tuple indices must be integers, not str"
         # "ipdb> row" says: "(80000, 46120.0)"
         #print row[ 'xtime' ] , row[ 'col4' ]
         print row[0] , row[1]

fs.close()
dbc.close()
==============================

I don't fully understand it yet, but it's nice to see that it works! 
Thank you very much for that! Now I'll have to concentrate on figuring 
out how/why it works :-)

>    You can be creative with the data selections
>    and pass them off to be plotted as needed ....

Yes, I understand. Thank you very much. As you can see, on my system I 
had to use:

         print row[0] , row[1]

instead of:

         print row[ 'xtime' ] , row[ 'col4' ]

I'm not sure exactly why - is it because you have another version of 
sqlite3 ? This is a bit strange, but anyway I can live with that - at 
least for now...

>    If mysql is used instead of sqlite3
>    you should only have to monkey with
>    the data type declarations in xdata_create.sql
>    and the  dbc.connect  strings in the python code ....

Actually I didn't knew anything about this sqlite3 before now. This is 
the first time I try it out, so I don't really know what's the 
difference between sqlite3 and mysql...

But thank you very much for providing some code I can now study and 
learn from !!! Much appreciated....