Tag IntervalConstraintProgramming.jl v0.4.0

[attobot]

Repository: dpsanders/IntervalConstraintProgramming.jl
Release: v0.4.0
Travis:
Diff: vs v0.3.0
requires vs v0.3.0:

--- 0.3.0/requires
+++ 0.4.0/requires
@@ -1,6 +1,5 @@
 julia 0.4
-ValidatedNumerics 0.5
+ValidatedNumerics 0.5.1  # plot recipes from 0.5.1
 MacroTools 0.3
 Compat 0.7.14
-FixedSizeArrays 0.1.2
-
+FixedSizeArrays

cc: @dpsanders

[tkelman]

is there no longer a minimum version requirement?

[dpsanders]

I'll put it back. How do I do that with attobot?

[tkelman]

make the commit to master, then delete and re-submit the release (without closing this PR, somewhat oddly - it'll update in place)

[simonbyrne]

See https://github.com/attobot/attobot#updatingdeleting-a-release for how to delete. It's not completely obvious.

[tkelman]

You have a couple of pretty huge ipynb files in here, it's not always a good idea to put those in the git history.

[dpsanders]

You have a couple of pretty huge ipynb files in here, it's not always a good idea to put those in the git history.

What's the alternative?

[tkelman]

separate repo, non-repo hosting, or attach them to github releases?

[dpsanders]

None of those seem to be particularly good solutions IMO.

[tkelman]

Why not? These are semi-independent documentation, not part of the library code. Alternate branch would be another option but would make less of a difference.

[dpsanders]

Yes, OK, true.

[dpsanders]

Do you suggest I strip them completely from the history? If so, what's a sensible way to do that?

[tkelman]

Were they added in a single commit? Or did they exist at previously registered tags? This is one of the few cases where rebasing and force pushing to master can be worth the inconvenience, but should be careful to only rebase commits that aren't yet part of a published release tag. If you added them in a single commit, you can add a new commit that deletes them, then rebase relative to the commit right before the one that added them, and squash the deletion commit into the addition commit.

[dpsanders]

They weren't added in a single commit. Some of them were in previous tags, although possibly with different names. It's a bit of a mess.

[tkelman]

Looks like JuliaIntervals/IntervalConstraintProgramming.jl@5a1caad is responsible for adding the really large ones. For smaller ones it's not worth rewriting history.

[attobot]

Repository: dpsanders/IntervalConstraintProgramming.jl
Release: v0.4.0
Travis:
Diff: vs v0.3.0
requires vs v0.3.0:

--- 0.3.0/requires
+++ 0.4.0/requires
@@ -1,6 +1,5 @@
 julia 0.4
-ValidatedNumerics 0.5
+ValidatedNumerics 0.6 0.6.1
 MacroTools 0.3
 Compat 0.7.14
 FixedSizeArrays 0.1.2
-

cc: @dpsanders

[dpsanders]

I have (EDIT: rewritten history to) removed all the example notebooks out into a separate repo (linked from the README).
This is now ready to merge.

[tkelman]

Does that upper bound need to apply to past versions of the package?

[dpsanders]

Good point. I'll send a separate PR for that. Should the upper bound be written as "0.6+"?

…

On Mon, Feb 6, 2017 at 12:11 PM, Tony Kelman ***@***.***> wrote: Does that upper bound need to apply to past versions of the package? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#7524 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AALtTuKGjmcopvssZ7PUZrJqs6l6hLnDks5rZ2JIgaJpZM4Lfk7R> .

-- Dr. David P. Sanders Profesor Titular "B" / Associate Professor Departamento de Física, Facultad de Ciencias Universidad Nacional Autónoma de México (UNAM) dpsanders@g <dpsanders@ciencias.unam.mx>mail.com / Twitter: @davidpsanders <https://twitter.com/DavidPSanders> http://sistemas.fciencias.unam.mx/~dsanders / GitHub: dpsanders <https://github.com/dpsanders> Cubículo / office: #414, 4o. piso del Depto. de Física Tel.: (+52 55) 5622 4965

[tkelman]

what do you want the upper bound to be?

[dpsanders]

Allow 0.6.0 but not 0.6.1.

…

On Mon, Feb 6, 2017 at 12:51 PM, Tony Kelman ***@***.***> wrote: what do you want the upper bound to be? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#7524 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AALtThJ-6__5x7yNaJvOmC8TkoUc732zks5rZ2vJgaJpZM4Lfk7R> .

-- Dr. David P. Sanders Profesor Titular "B" / Associate Professor Departamento de Física, Facultad de Ciencias Universidad Nacional Autónoma de México (UNAM) dpsanders@g <dpsanders@ciencias.unam.mx>mail.com / Twitter: @davidpsanders <https://twitter.com/DavidPSanders> http://sistemas.fciencias.unam.mx/~dsanders / GitHub: dpsanders <https://github.com/dpsanders> Cubículo / office: #414, 4o. piso del Depto. de Física Tel.: (+52 55) 5622 4965

[tkelman]

then 0.6.1 is most clear, the upper bounds are exclusive

[tkelman]

Also what are you breaking between patch releases? That's not usually supposed to happen.

[dpsanders]

It turns out that v0.6.1 of VN does not play well once you import it into another package. (It works (?) but gives a lot of scary warnings.)
I guess I can't delete that tag?

[tkelman]

you could supersede it with a v0.6.2 that reverts the offending commit or something?

[dpsanders]

Please merge this once ValidatedNumerics 0.6.2 (reverting the offending commit) has been merged.

[tkelman]

won't you want to remove the upper bound then?

[dpsanders]

Yes, but I don't know how to do that before until it is merged!

[tkelman]

there are instructions on the attobot readme

[dpsanders]

I mean that master has changed since then. So I would make a new branch and do the release from that branch?

[attobot]

Repository: dpsanders/IntervalConstraintProgramming.jl
Release: v0.4.0
Diff: vs v0.3.0
requires vs v0.3.0:

--- 0.3.0/requires
+++ 0.4.0/requires
@@ -1,6 +1,5 @@
 julia 0.4
-ValidatedNumerics 0.5
+ValidatedNumerics 0.6
 MacroTools 0.3
 Compat 0.7.14
 FixedSizeArrays 0.1.2
-

cc: @dpsanders

[tkelman]

ah, yes

[tkelman]

Why isn't ∪{N,T}(X::IntervalBox{N,T}, Y::IntervalBox{N,T}) implemented in ValidatedNumerics, if that type is from there?

[dpsanders]

Fixed, thanks.

[tkelman]

Should the tag be moved to JuliaIntervals/IntervalConstraintProgramming.jl@09f2467 then or were you going to make a 0.4.1 for that?

Similarly with iseven and isodd for Interval, shouldn't those methods come from the package that owns that type?

[attobot]

Repository: dpsanders/IntervalConstraintProgramming.jl
Release: v0.4.0
Travis:
Diff: vs v0.3.0
requires vs v0.3.0:

--- 0.3.0/requires
+++ 0.4.0/requires
@@ -1,6 +1,5 @@
 julia 0.4
-ValidatedNumerics 0.5
+ValidatedNumerics 0.6
 MacroTools 0.3
 Compat 0.7.14
 FixedSizeArrays 0.1.2
-

cc: @dpsanders

[dpsanders]

I have moved the tag.
I will deal with iseven and isodd in the next release.

Thanks for your careful attention to these details!