db-derby-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Thomas Vatter <thomas.vat...@network-inventory.de>
Subject Re: order by
Date Thu, 02 Mar 2006 17:03:18 GMT
Michael Segel <derby@...> writes:

> 
> On Thursday 02 March 2006 5:00 am, Thomas Vatter wrote:
> 
> This doesn't make any sense.
> You keep saying "spreadsheet".
> 
> Are you reading directly from an excel spreadsheet? .XLS or did you convert 
> the spreadsheet to a .CSV format?
> 
> The other issue of concern is that you say that you continue to die at 
> approximately 1200 rows. 
> 
> The first thought is that its a data set problem because you indicate that it 
> occurs with a single data set.
> 
> The second thought is that its a programming issue on your part because you 
> now say that you're using different portions of the data set and its still 
> hanging at 1200 rows.
> 
> A simple read/parse/insert program should not eat memory.
> 
> Are you using a prepared statement?
> 
> As you parse each line of data, how are you storing the fields and when you do 
> your insert, are you properly "flushing" the variables? (Destroying the 
> objects...)
> 
> The only negative thing is that you can't use the split() method. You have to 
> manually parse the string to pull out the data fields, strip the quotes and 
> white space.... 
> (Which is why ESQL/C  or C is a better choice for this type of work...)
> 
> HTH,
> 

spreadsheet
I am not reading the .XLS file, just reading ASCII format

should not eat memory
I absolutely agree, I have to review the code, this will take some time, I will
do the separation of database and application over two machines first, for
easier analysis

prepared statement
usually I do, I will check this in the review

tom

 





Mime
View raw message