I finally updated METADATA to point to the fixed version of BloomFilters.

 — John

On Dec 6, 2014, at 9:23 AM, Stefan Karpinski <[email protected]> wrote:

> On Sat, Dec 6, 2014 at 12:14 PM, David Koslicki <[email protected]> wrote:
> 
> Implementing your own Bloom filter really shouldn't be too hard. 
> Alternatively, it might not be too hard to file some issues against John's 
> package and get it into better working state. If you mention me in an issue, 
> I can also take a look at things. Having the BloomFilters package working 
> would probably be a good thing.
> 
> If you look at the link I provided, you'll see that I've previously done 
> exactly as you're suggesting.
> 
> Great. Thanks for helping out!
> 
> Exactly, and that's why I'm using my own implementation.
> 
> I have thought using using a single string to store all the smaller strings, 
> but I think it's even more computationally difficult to come up with a single 
> (shortest) string that contains all my specified substrings (and no more).  
> Correct me if I am wrong on this point though, as that would be great!
> 
> I meant to just use a single BioSeq array and concatenate all the sequences, 
> not trying to make the overlap. This just reduces the overhead, it's not a 
> compression technique.

Reply via email to