Re: [boost] Proposed templated integer_sort

From: Phil Endecott (spam_from_boost_dev_at_[hidden])
Date: 2008-07-04 17:26:13

Steven Ross wrote:
> From my reading, it actually has a larger L1 cache (and smaller L2 cache),
> and for that purpose a MAX_SPLITS as high as 13 might be better.  Increasing
> LOG_MIN_SPLIT_COUNT might also speed up your test a little.

I'm afraid adjusting those makes it slower.

>> OK, I've put a 14 Mbyte file called placenames.dat.bz2 in the directory
>> http://chezphil.org/tmp/ (I've not written the URL in full to stop search
>> engines from wasting their time on it).  When un-bziped you'll find a
>> tab-delimited 3-column text file.  I got basically the same results using
>> the latitude and longitude columns.
>>
>
> What's the best way to read data like that from a file?
> fscanf("%f %f %s\n")?  Won't that have problems if there is a space?  I
> don't do much file parsing in C/C++.

Nor me.  awk it into something else first.  You can probably use 
fscanf("%f\t%f\t%[^\n]") or try cin >> lng >> lat; getline(cin,name);.

Phil.