🚀 go-pugleaf

On Thu, 26 Apr 2012 11:54:58 -0700 (PDT), kazik wrote:

> DataTable t = new DataTable();
> a.Fill(t);
> for (int i = 0; i < t.Rows.Count; i++)
> {
> txt += "\n" + t.Rows[i][0] + "\t" + t.Rows[i][1] + "\t" + t.Rows[i][2]
> + "\t" + t.Rows[i][3] + "\t" + t.Rows[i][4] + "\t" + t.Rows[i][5] +
> "\t" + t.Rows[i][6] + "\t" + t.Rows[i][7] + "\t" + t.Rows[i][8] + "\t"
> + t.Rows[i][9] + "\t" + t.Rows[i][10] + "\t" + t.Rows[i][11] + "\t" +
> t.Rows[i][12] + "\t" + t.Rows[i][13] + "\t" + t.Rows[i][14] + "\t" +
> t.Rows[i][15] + "\t" + t.Rows[i][16] + "\t" + t.Rows[i][17] + "\t" +
> t.Rows[i][18] + "\t" + t.Rows[i][19];
> }
> 
> I need this tabs between columns to generate txt file later.
> Problem is that t.Rows.Count = 600000 and this loop works 9 hours.
> Any ideas how to make this faster?

I assume txt is a string variable, use a StringBuilder instead, and 
preallocate it if you can even roughly estimate the total string
length. Most of the execution time is consumed by string 
reallocations - whenever in doubt, use a profiler...

...or just push the data to the destination file directly instead of 
buffering it in a huge string and then saving it.

You might as well cache the DataRow in a Variable so the t.Rows 
indexer propertny isn't called repeatedly for a single row.

p.s. To jest polsko-języczna grupa ;-).

-- 
Pozdrawiam,
Łukasz 'Maly' Ostrowski.      http://l3v.pl/