On Fri, Jul 31, 2020 at 12:07 PM Ian Denhardt <[email protected]> wrote:

> It would be nice to have a precisely specified splitting function, so
> that it was possible to write interoperable software without having to
> deal with different tools computing different splits


I heartily agree! And thanks for the suggestion. In order to contribute to
such a specification I would first need to educate myself on the pluses and
minuses of different rolling hashes (e.g. beginning here
<https://en.wikipedia.org/wiki/Rolling_hash>). What I have right now in my
hashsplit package is a little cargo-culty.

Not sure what my time will allow, but I will make this effort and keep you
posted. Meanwhile, if you make any progress I'd love to hear about it.

Cheers,
- Bob

(thereby somewhat
> negating the dedup benefits) or having to just rely on the same library
> everywhere (and thus language...). With cryptographic hash functions
> we have names for specific widely implemented functions (e.g. sha224);
> it would be nice to have the same for hash splitting.
>
> I could be talked into helping out with a formal spec.
>
> -Ian
>
> Quoting Bob Glickstein (2020-07-31 09:53:36)
> >    On Fri, Jul 31, 2020 at 1:33 AM Ian Denhardt <[1][email protected]>
> >    wrote:
> >
> >      I have kicking around on my hard drive a WIP library that just
> >      provides
> >      the splitting algorithm, without being entangled with the rest of
> >      the
> >      file upload logic.
> >
> >    I did that recently too. It's not identical to the implementation in
> >    Perkeep though. It's at� [2]https://github.com/bobg/hashsplit.
> >    Cheers,
> >    - Bob
> >    �
> >
> >      I should finish that up and publish it...
> >      Hope this is useful,
> >      -Ian
> >      --
> >      You received this message because you are subscribed to the Google
> >      Groups "Perkeep" group.
> >      To unsubscribe from this group and stop receiving emails from it,
> >      send an email to [3][email protected].
> >      To view this discussion on the web visit
> >      [4]
> https://groups.google.com/d/msgid/perkeep/159618442004.25997.1087
> >      4365701907187897%40localhost.localdomain.
> >
> >    --
> >    You received this message because you are subscribed to the Google
> >    Groups "Perkeep" group.
> >    To unsubscribe from this group and stop receiving emails from it, send
> >    an email to [5][email protected].
> >    To view this discussion on the web visit
> >    [6]
> https://groups.google.com/d/msgid/perkeep/CAEf8c4_Cq5%3D%3DEetpr1G3N
> >    C-HqzGfRrFdxg%2BdCAqX8P-5F8SbQA%40mail.gmail.com.
> >
> > Verweise
> >
> >    1. mailto:[email protected]
> >    2. https://github.com/bobg/hashsplit
> >    3. mailto:perkeep%[email protected]
> >    4.
> https://groups.google.com/d/msgid/perkeep/159618442004.25997.10874365701907187897%40localhost.localdomain
> >    5. mailto:[email protected]
> >    6.
> https://groups.google.com/d/msgid/perkeep/CAEf8c4_Cq5%3D%3DEetpr1G3NC-HqzGfRrFdxg%2BdCAqX8P-5F8SbQA%40mail.gmail.com?utm_medium=email&utm_source=footer
>

-- 
You received this message because you are subscribed to the Google Groups 
"Perkeep" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/perkeep/CAEf8c4_V3S2uKtffYtYMuuCDHMW%2B5niSXR84ep3JW1CxrqBaSQ%40mail.gmail.com.

Reply via email to