1.7 freeze

[DGDanforth]

Are there any more outstanding issues or can we freeze development of 1.7?

If we freeze now that does not necessarily mean we release now. Some time must past
for people to USE the unstable version to shake out any unforeseen bugs.
However Edsger Dijkstra said "Testing shows the presence, not the absence of bugs " so we can't
wait forever.

-Doug

[Josef Templ]

Issue-#125 (trap on WITH ELSE END) has not been voted on.
This needs to be merged to master.
Then we can release 1.7.

- Josef

[Robert]

If we freeze now that does not necessarily mean we release now.

The problem with waiting is that new bugs / issues will inevitably arise during that period.

Any moment we get a clear bug list we should publish immediately - I think.

[Ivan Denisov]

I agree with Josef, that we should publish 1.7 after adopting issue-#125 fix.

Robert, clear bug list is impossible because there are some rare bugs like one with files damage. So it is hard to analyse and fix them now. Also there should be more thinking and understanding the difference between two bug fixes of issue-#126, so we can leave it for later patch.

[Josef Templ]

issue-#126 is also ready for voting and merging with master.
Sorry, I forget that in my previous posting.

- Josef

[Robert]

Robert, clear bug list is impossible because ...

If we can sort out 125 & 126 the bug list for 1.7 will be clear, which is what I meant.

[Ivan Denisov]

issue-#126 is also ready for voting and merging with master.
Sorry, I forget that in my previous posting.

It is not essential to include this fix in the back-end of the compiler in last minute without any discussion and testing.
With a great respect to Josef experience anyway it can led to unknown results.
I think, that we can leave this for 1.7.1 patch.

[Josef Templ]

issue-#126 is also ready for voting and merging with master.
Sorry, I forget that in my previous posting.

It is not essential to include this fix in the back-end of the compiler in last minute without any discussion and testing.
With a great respect to Josef experience anyway it can led to unknown results.
I think, that we can leave this for 1.7.1 patch.

With great respect to Ivan's inexperience, it is impossible that this fix introduces problems because it
is only changing the error reporting behavior from trapping to non-trapping.

And of course there should be a discussion and actually there is one going on.

- Josef

[Ivan Denisov]

IMHO, we should release 1.7 now.

[Josef Templ]

Then you have a known compiler TRAP in the release.
Are you serious?