Re: [boost] Cygwin tests (Was: [1.61] Two weeks remaining for new libraries and breaking changes)

Subject: Re: [boost] Cygwin tests (Was: [1.61] Two weeks remaining for new libraries and breaking changes)
From: Robert Ramey (ramey_at_[hidden])
Date: 2016-02-17 12:02:30

On 2/17/16 1:25 AM, Vladimir Prus wrote:

> - It seems you have some issues on Cygwin
> - Steven has provided a patch
> - You haven't had the time to look into it yet
>
> So, it seems that on the issue of cygwin testing, you are the person who
> has to make the next step. Is that
> correct?

LOL - I'm a volunteer, I don't HAVE to do anything.

But I do have to make a choice.  My options are

a) spend time with b2 development along the lines that Stephen has 
suggested.

b) presume that testing on my os shows that the serialization library
is indeed correct and ignore the testing failures on cygwin and maybe?
mingw.

c) ignore the failures on the develop test matrix due to changes
in develop of other libraries and just merge from develop into release.

d) do nothing and wait for other stuff to get fixed.  The serialization
library won't get the latest improvements, but it won't break anything 
either.  That's the option I chose for 1.59.  I can do it again.

I fix bugs and check enhancements to the library on my local system.
At this point I presume they are correct and that I haven't introduced
new bugs.  I look to the test system to prove me wrong.  But right
now for the reasons I've cited, it can't do that.  I need the
help it's supposed to provide but it's providing it.  I'm not so
concerned about the next release as I am about this process working
smoothly.  I've made several suggestions about how we can make
this simpler - which amount to making the test system development
look more like the rest of boost.

I'm really reluctant to dive into b2 development/debug for a number
of reasons which I'm guessing should be pretty obvious.

Robert Ramey