Boost logo

Boost :

Subject: Re: [boost] Any interest in Compile Time Hash Containers library?
From: degski (degski_at_[hidden])
Date: 2019-01-25 08:13:46

On Fri, 25 Jan 2019 at 08:37, Ivan Matek <libbooze_at_[hidden]> wrote:

> How is this [whatever it is] different from frozen (
>> )?
>From your doc:


Another example of library doing very similar thing to what we want to do
is Frozen
For details you should consult the documentation,

but the general idea is similar: try to get good hashing based on the fact
the data is known at compilation time. Difference to us is that frozen does
perfect hashing.

"For details you should consult the documentation", right, we're getting
somewhere, oh, maybe not ... "Difference to us is that frozen does perfect
hashing": ... and your idea is ... to do imperfect hashing? The document is
far too long and there are far too many diversions/excursions into
la-la-land. Take as example the from frozen and put something
similar together, please.

> If you want the code feel free to ping me directly on my mail ...

Just stick it on github, that's what the rest of the world does ... if you
want to have it included in Boost, that's what you'll need to do in the end
anyway ...

> , only things I require is that:
> you do not claim ownership of code
> you do not blame me for bugs in case you do use code in real production
> and it causes damage, in other words no liability on my part as permissible
> by law
> you do not expect me to support you in x years(if I can help you today I
> will) or implement features on top of this code(I think there is a good
> chance this code will be changed significantly including: names and hashers
> avaliable)
> you do compile with clang(GCC/VC++ are partly/fully broken for this code)

... and put an appropriate license (BSL ( ) comes to mind, as that's what
you are proposing anyway) to deal with the above...


*“If something cannot go on forever, it will stop" - Herbert Stein*

Boost list run by bdawes at, gregod at, cpdaniel at, john at