Why not read in just the date and ID columns to start with, then do a
numpy.unique() or python set() on theses, then query based on the unique
values? Seems like it might be faster
Be Well
Anthony
On Mon, Jul 2, 2012 at 5:16 PM, Aquil H. Abdullah
wrote:
> Hello All,
>
> I have a table that
No worries ;)
On Mon, Jul 2, 2012 at 5:41 PM, Jacob Bennett wrote:
> Cool, this seems pretty straightforward. Thanks again Anthony!
>
> -Jacob
>
>
> On Mon, Jul 2, 2012 at 1:10 PM, Anthony Scopatz wrote:
>
>> Hello Jacob,
>>
>> It seems like you have answered your own question ;). The thing is
Cool, this seems pretty straightforward. Thanks again Anthony!
-Jacob
On Mon, Jul 2, 2012 at 1:10 PM, Anthony Scopatz wrote:
> Hello Jacob,
>
> It seems like you have answered your own question ;). The thing is that
> the locking doesn't have to be all that hard. You can simply check if the
>
Hello All,
I have a table that is indexed by two keys, and I would like to search for
duplicate keys. So here is my naive slow implementation: (code I posted on
stackoverflow)
import tables
h5f = tables.openFile('filename.h5')
tbl = h5f.getNode('/data','data_table') # assumes group data and tabl
Hello Jacob,
It seems like you have answered your own question ;). The thing is that
the locking doesn't have to be all that hard. You can simply check if the
file is already in the open files cache (from previous thread). If it
isn't, the thread will open the file. Or you don't even have to d
Hello PyTables Users,
I am developing an API to access the current data stored in my pytables
instance. Note at this point that this is only reading, no writing to the
files. The big question on my mind at this point is how am I supposed to
handle the opening and closing of files on read requests
