Displaying #maven-dev/2017-03-19.log:

Sun Mar 19 09:08:51 2017  tibor_:Joined the channel
Sun Mar 19 09:26:12 2017  rfscholte:Joined the channel
Sun Mar 19 12:21:47 2017  rfscholte:stephenc: yt?
Sun Mar 19 13:15:01 2017  tibor_:Joined the channel
Sun Mar 19 15:59:56 2017  tibor_:Joined the channel
Sun Mar 19 16:39:49 2017  stephenc:rfscholte: I am now
Sun Mar 19 16:40:00 2017  stephenc:Oh well
Sun Mar 19 16:40:58 2017  stephenc:hboutemy: how would you feel about a more strict branch naming scheme?
Sun Mar 19 16:41:27 2017  stephenc:Something like: owner/targetBranch/mng-XXXX
Sun Mar 19 16:41:28 2017  hboutemy:why not, if this can ease understanding
Sun Mar 19 16:42:29 2017  stephenc:if we know the target branch we can have the Jenkinsfile do a speculative merge to the target branch as part of the validation
Sun Mar 19 16:42:43 2017  hboutemy:?
Sun Mar 19 16:43:20 2017  hboutemy:everything looks too complex now
Sun Mar 19 16:43:49 2017  hboutemy:there are too many tasks in progress
Sun Mar 19 16:44:12 2017  hboutemy:work on 3.5.1 but we still don't have 3.5.0 out
Sun Mar 19 16:44:29 2017  stephenc:I know
Sun Mar 19 16:44:38 2017  stephenc:We have some
Sun Mar 19 16:44:48 2017  stephenc:Moving to fast for the number of people we have
Sun Mar 19 16:44:57 2017  hboutemy:and too much on ML
Sun Mar 19 16:45:41 2017  stephenc:I'd rather too much on ML than too little ;-)
Sun Mar 19 16:45:48 2017  hboutemy:now, we are missing confidence in what others do
Sun Mar 19 16:46:21 2017  hboutemy:either too much explanations => TLDR
Sun Mar 19 16:46:40 2017  stephenc:opens private chat
Sun Mar 19 16:46:50 2017  hboutemy:or not sufficient explanations: "improvements" without explaining what is the new behaviour
Sun Mar 19 17:00:38 2017  hboutemy:stephenc: what's missing for 3.5.0?
Sun Mar 19 17:00:53 2017  stephenc:OK, there is the decision on MNG-6182
Sun Mar 19 17:01:06 2017  stephenc:now that there are some tests, it's not totally dead code
Sun Mar 19 17:01:21 2017  stephenc:I need to review the logic in the tests... but assuming it's sound I'm ok with that
Sun Mar 19 17:01:37 2017  stephenc:There is the m-assembler-p upgrade... has that landed yet?
Sun Mar 19 17:01:44 2017  hboutemy:I didn't take time to review: TLDR
Sun Mar 19 17:01:55 2017  hboutemy:yes, I upgraded m-assembler-p
Sun Mar 19 17:02:03 2017  stephenc:there is your warning
Sun Mar 19 17:02:07 2017  hboutemy:it was required, of the .zip was unusable on Unix
Sun Mar 19 17:02:17 2017  stephenc:ack
Sun Mar 19 17:03:16 2017  stephenc:from https://issues.apache.org/jira/issues/?jql=project%20%3D%20MNG%20AND%20fixVersion%20%3D%203.5.0-alpha-2%20AND%20resolution%20%3D%20Unresolved%20AND%20priority%20%3D%20Major%20ORDER%20BY%20key%20DESC it's just your warning
Sun Mar 19 17:03:29 2017  stephenc:and then a decision on MNG-6182
Sun Mar 19 17:04:00 2017  stephenc:oh whoop de do... MNG-6182 was merged! ffs
Sun Mar 19 17:04:21 2017  stephenc:trys not to channel bruce banner
Sun Mar 19 17:04:22 2017  tibor_:Joined the channel
Sun Mar 19 17:05:02 2017  hboutemy:yes, MNG-6182 was merged: what remains is to merge tests or not...
Sun Mar 19 17:05:17 2017  stephenc:it's merge tests or revert MNG-6182 IMHO
Sun Mar 19 17:06:07 2017  hboutemy:description of https://issues.apache.org/jira/browse/MNG-6182
Sun Mar 19 17:06:12 2017  hboutemy:"enhancement"
Sun Mar 19 17:06:24 2017  hboutemy:nothing more
Sun Mar 19 17:06:33 2017  stephenc:still trying not to channel bruce banner... hervé is not helping
Sun Mar 19 17:07:11 2017  hboutemy:in fact, is MNG-4463 for 3.5.1 or 3.6.0?
Sun Mar 19 17:07:34 2017  hboutemy:because IIUC, this "enhancement" is to be able to add a feature in 3.5.1
Sun Mar 19 17:07:46 2017  stephenc:well depends on PoV
Sun Mar 19 17:08:15 2017  stephenc:is scope import not working with ranges a bug or is adding range support a feature
Sun Mar 19 17:08:43 2017  stephenc:everything in <dependencies> except scope import works with ranges
Sun Mar 19 17:09:04 2017  stephenc:so from that PoV one *could* argue that range support is a core feature of dependencies
Sun Mar 19 17:09:06 2017  hboutemy:I don't understand what people expect from ranges
Sun Mar 19 17:09:18 2017  stephenc:and therefore adding range support fixes a bug
Sun Mar 19 17:09:21 2017  hboutemy:but ranges on imports are even more strange
Sun Mar 19 17:09:35 2017  stephenc:and bug fixes are eligible in patch releases
Sun Mar 19 17:09:45 2017  stephenc:yep... I know
Sun Mar 19 17:10:20 2017  stephenc:ranges are like a (╯°□°)╯︵ ┻━┻
Sun Mar 19 17:10:36 2017  hboutemy:notice: instead of downloading a wrong artifact (which is a bug), I'd expect a clean error
Sun Mar 19 17:10:43 2017  stephenc:wonders if that gets rendered correctly over IRC
Sun Mar 19 17:10:53 2017  hboutemy:"ranges are not supported for imports"
Sun Mar 19 17:11:02 2017  stephenc:yes we state that up front
Sun Mar 19 17:11:20 2017  stephenc:but was that "we're lazy so we'll call a bug a feature"
Sun Mar 19 17:11:38 2017  stephenc:or was that "we are designing it this way to save you from your own stupidity"
Sun Mar 19 17:11:48 2017  hboutemy:+1 on the second
Sun Mar 19 17:12:09 2017  stephenc:well if that is the case then we should just close MNG-4463 as won't fix
Sun Mar 19 17:12:12 2017  stephenc:and revert 6182
Sun Mar 19 17:12:14 2017  hboutemy:but I'll change "stupidity" to "risky choices"
Sun Mar 19 17:12:55 2017  hboutemy:in fact, we'd need to have a good discussion to find why they want this
Sun Mar 19 17:13:07 2017  hboutemy:but I fear that won't be constructive
Sun Mar 19 17:13:12 2017  hboutemy:or a loof TLDR
Sun Mar 19 17:13:21 2017  hboutemy:s/loof/lot of/
Sun Mar 19 17:13:46 2017  stephenc:there was me thinking you were referring to a Nicolas de style melt-down ;-)
Sun Mar 19 17:14:22 2017  hboutemy::)
Sun Mar 19 17:14:53 2017  stephenc:only thing with closing MNG-4463 is that there has been a lot of comments on the issue from core committers since 2015
Sun Mar 19 17:15:07 2017  stephenc:nobody has yet said... you know, we don't want to do this thing
Sun Mar 19 17:15:45 2017  stephenc:https://github.com/apache/maven/pull/64#issuecomment-139687217
Sun Mar 19 17:16:38 2017  stephenc:I feel at this stage that the issue has gone on too long without being stopped
Sun Mar 19 17:16:51 2017  stephenc:perhaps we should just let users be stupid if they want to
Sun Mar 19 17:17:01 2017  hboutemy:+1
Sun Mar 19 17:17:12 2017  hboutemy:if this does not cause too much complexity
Sun Mar 19 17:17:32 2017  stephenc:ok then we merge the tests and let MNG-4463 be seen as a bugfix so eligible for 3.5.1
Sun Mar 19 17:17:56 2017  hboutemy:+1
Sun Mar 19 17:18:11 2017  hboutemy:won't cause harm for people not using ranges
Sun Mar 19 17:18:15 2017  hboutemy:= people I care about
Sun Mar 19 17:19:24 2017  stephenc:what do we feel about MNG-6069
Sun Mar 19 17:19:31 2017  stephenc:commons-cli deprecation tidy-up
Sun Mar 19 17:24:38 2017  hboutemy:in fact, there is some public API change
Sun Mar 19 17:24:57 2017  hboutemy:during the tidy-up
Sun Mar 19 17:25:20 2017  hboutemy:I don't know if some integrators use the public API
Sun Mar 19 17:25:44 2017  hboutemy:I'd wait: this is not urgent
Sun Mar 19 17:26:03 2017  hboutemy:and this question of API change could cause issues
Sun Mar 19 17:26:55 2017  stephenc:Yep... I tend to agree
Sun Mar 19 17:27:12 2017  stephenc:OTOH we could merge it for beta-1 and revert if there are issues identified
Sun Mar 19 17:32:58 2017  hboutemy:yes, that would be less hard for good old committers
Sun Mar 19 17:33:21 2017  hboutemy:that make some cleaning...
Sun Mar 19 17:34:06 2017  stephenc:plus we can argue that this is a 3.5.0 so we can break some small internal things
Sun Mar 19 17:35:25 2017  hboutemy:yes
Sun Mar 19 17:35:59 2017  hboutemy:in fact, these are constants: IMHO, this won't break binary compat
Sun Mar 19 17:48:46 2017  stephenc:is surprised that lamb does not roast itself
Sun Mar 19 18:07:23 2017  stephenc:In other news, potatoes do not roast themselves either
Sun Mar 19 18:13:58 2017  hboutemy:lamb could have helped, but I confirm potatoes are vegetables: they can't move from themselves
Sun Mar 19 18:14:13 2017  hboutemy:(IIUC, of course...)
Sun Mar 19 18:42:02 2017  Michael-O:Joined the channel
Sun Mar 19 22:10:41 2017  Michael-O:Joined the channel
Sun Mar 19 22:10:51 2017  Michael-O:tibor_: Sorry, got kicked.
Sun Mar 19 22:11:14 2017  Michael-O:I have seen that you were quite busy. I will read your mails tomorrow and see which tests I need to rerun.
Sun Mar 19 22:11:16 2017  tibor_:Michael-O: Hi what's up?
Sun Mar 19 22:11:26 2017  Michael-O:Been working on a patch for JMeter
Sun Mar 19 22:12:32 2017  tibor_:Maybe I would use JMeter in my JEE project.
Sun Mar 19 22:15:07 2017  Michael-O:JMeter is really awesome, highly recommened
Sun Mar 19 22:15:16 2017  Michael-O:recommended*
Sun Mar 19 22:15:46 2017  tibor_:Michael-O: I have several Shelfs for Surefire. Later I will create branches, but before I would wait for you two guys if you have test for Surefire141PluggableProvidersIT. Generally the result is good.
Sun Mar 19 22:16:29 2017  Michael-O:which branch is Surefire141PluggableProvidersIT?
Sun Mar 19 22:16:39 2017  tibor_:SUREFIRE-1342
Sun Mar 19 22:18:27 2017  tibor_:but I guess you tested, but Guillaume Boue tested it on FreeBSD with a failure. The *.dump file was missing and I guess because of surefire-reports/ folder did not exit. But what scares me the Time: 34 seconds in the test and it should be 5.
Sun Mar 19 22:19:00 2017  Michael-O:You haven't changed since 13th.
Sun Mar 19 22:19:08 2017  Michael-O:Rerunning single test right now
Sun Mar 19 22:19:10 2017  Michael-O:hold on
Sun Mar 19 22:20:19 2017  tibor_:I changed the branch today evening.
Sun Mar 19 22:22:30 2017  Michael-O:You did not push
Sun Mar 19 22:22:40 2017  Michael-O:see: https://github.com/apache/maven-surefire/tree/SUREFIRE-1342
Sun Mar 19 22:23:35 2017  tibor_:forced push
Sun Mar 19 22:23:42 2017  tibor_:not nice
Sun Mar 19 22:25:37 2017  Michael-O:please force push, this is ok for a dev branch
Sun Mar 19 22:26:15 2017  tibor_:like this 1342 it is ok?
Sun Mar 19 22:26:29 2017  tibor_:I would not do it in master of course.
Sun Mar 19 22:27:14 2017  tibor_:My point was to have single commit. And later merge only one commit per issue. Is that ok?
Sun Mar 19 22:29:04 2017  Michael-O:Force push is ok as long as only one person is working up. You can clean up your branch, squash and the merge back to master. So, this is ok for me.
Sun Mar 19 22:29:11 2017  Michael-O:You can simply push a second commit with your new changes now.
Sun Mar 19 22:29:31 2017  Michael-O:After testing, you can perform cleanups in a third commit and then squash altogether.
Sun Mar 19 22:29:36 2017  Michael-O:This is what I do
Sun Mar 19 22:30:32 2017  tibor_:I use IntelliJ IDEA. Do you know if squash is possible in idea?
Sun Mar 19 22:30:51 2017  tibor_:Once I was doing it but it was not simple in cli.
Sun Mar 19 22:30:58 2017  Michael-O:Nope, I always use Git in the shell
Sun Mar 19 22:31:06 2017  Michael-O:git rebase -i HEAD~n
Sun Mar 19 22:31:07 2017  Michael-O:that's it
Sun Mar 19 22:31:09 2017  Michael-O:very simple
Sun Mar 19 22:31:17 2017  Michael-O:for browsing I use tig
Sun Mar 19 22:32:16 2017  Michael-O:I my opinion, Git is so horribly complex/rich that now GUI can provide all necessary bits like a CLI
Sun Mar 19 22:43:59 2017  Michael-O:tibor_: I can run a final test before I go to sleep. Do you want to push something and have me test?
Sun Mar 19 22:44:35 2017  tibor_:No, I do not wan to push anything today.
Sun Mar 19 22:44:44 2017  tibor_:want
Sun Mar 19 22:47:54 2017  Michael-O:Alright, then good night. See you tomorrow. I will read through the rest.
Sun Mar 19 22:48:26 2017  tibor_:ok, good bye.

Comments