public HashSet <long> GetNearDups(SimhashResult simhash)
{
/*
* "simhash" is an instance of Simhash
* return a list of obj_id, which is in type of long (for now)
*/
if (Simhash.FpSize != _fpSize)
{
throw new Exception();
}
var ans = new HashSet <long>();
foreach (var key in GetEnumerableKeys(simhash))
{
if (!_bucket.TryGetValue(key, out var dups))
{
continue;
}
foreach (var dup in dups)
{
var parts = dup.Split(',');
var fp = Convert.ToUInt64(parts[0]);
var objId = Convert.ToInt64(parts[1]);
var sim2 = new SimhashResult(fp);
var d = simhash.Distance(sim2);
if (d <= _kDistance)
{
ans.Add(objId);
}
}
}
return(ans);
}
