Remove Thread and Socket classes

[Jeffrey Walton]

Hi Everyone,

We have an open issue for removing Thread and Socket classes. Also see https://github.com/weidai11/cryptopp/issues/208.

I think Denis Bider summed it up nicely in the report. Effectively he says we should focus on core capabilities and avoid some of the less frequently used extras like Sockets, Thread Local Storage and Pipes. I personally like the position because it reduces code clutter and avoids some points of failure.

I checked in the removal on my testing branch at https://github.com/noloader/cryptopp . If all goes well I would like to merge it soon.

I plan on moving Remove Thread and Socket classes to the Patch Page on the wiki. If anyone needs them they will be located there.

Are there any objections to removing Remove Thread and Socket classes?

[Jeffrey Walton]

Merged at https://github.com/weidai11/cryptopp/pull/703 .

Jeff

[Jeffrey Walton]

We added a wiki page as a sign post at https://www.cryptopp.com/wiki/ThreadLocalStorage . ThreadLocalStorage  seemed most important as it was wired into the FIPS gear. Everything else was mostly stand-alone.

The wiki page provides a list of affected files and classes, and provides a brief discussion of the reasons why.

Jeff

[Andrew Marlow]

Does this mean that FIPS support is no more after this change? I don't mind, I just need to know.

-Andrew M.
 

[Jeffrey Walton]

On Tue, Aug 21, 2018 at 5:39 AM, Andrew Marlow <marlow...@gmail.com> wrote:
> On Saturday, 18 August 2018 16:07:13 UTC+1, Jeffrey Walton wrote:

>> ...

>> We added a wiki page as a sign post at
>> https://www.cryptopp.com/wiki/ThreadLocalStorage . ThreadLocalStorage
>> seemed most important as it was wired into the FIPS gear. Everything else
>> was mostly stand-alone.
>>
>> The wiki page provides a list of affected files and classes, and provides
>> a brief discussion of the reasons why.
>
> Does this mean that FIPS support is no more after this change? I don't mind,
> I just need to know.

The Crypto++ FIPS validation was withdrawn by the CMVP when it place
Crypto++ on the Historical Validation List. Also see Denis Bider's
report at https://github.com/weidai11/cryptopp/issues/149 .

Jeff