🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
00:02 reportable6 left 00:05 reportable6 joined
japhb We should at least give people some guidance as to what is recommended for different use cases ... I would be surprised if new users didn't have some confusion about what libraries to use to get certain tasks done; I struggle with that still sometimes, and I've been here for a LONG time. 00:26
It doesn't have to be in the form of Rakudo Star, or a Task/Bundle like in the CPAN world, but we ought to provide *some* guidance so that people don't have to yak shave the first time they try something non-trivial. 00:27
(Some people *like* yak shaving, and that's fine, I'm just saying people shouldn't be forced into it.)
[Coke] exit 00:37
ugexe part of the magic of rakudo star is likely (im guessing) that it doesnt run the tests of all the modules it installs 00:41
00:56 frost joined 01:22 squashable6 left 01:23 squashable6 joined
[Coke] so the install is faster, but there's risk? 01:42
curating a list of modules is a big ask. I'm not sure how much curation has been done on the list that was included with R* since it was initially put together. And it doesn't include a list of why you'd want A or B, they're just there. 01:43
so putting that list together is a good idea regardless. One could create a Star::Bundle or somesuch as well that could be used to achieve a similar setup (but would be installed after the fact). 01:44
ugexe well you can also just tell users to use `zef --/test ...` 01:49
but generally you probably dont want to instill that type of behavior
[Coke] nope. I'd rather run the tests, agreed. 01:51
japhb Definitely. I still have to keep a list of dists that have to be installed with --/test in my "rebuild everything" script, but I regularly review that list to keep it as small as possible so I don't miss regressions. 02:07
And I think a Bundle::Star would be great, but even just explaining (as [Coke] said) why people would want to install one module or another would be valuable. 02:08
I wonder if we should do a Kwalitee-like think in the Raku ecosystem -- automated analysis of module dists for minimal adherence to best practices -- so that we could make that a filter for people searching for modules 02:10
Might also be good to start socializing some level of keyword tagging or SEO techniques that would allow people to find things easier. Our current module search algorithms are pretty pedantic. 02:11
[Coke] Do we have enough hosting to consider throwing an elasticsearch or solr instance in there? 02:12
I wonder if Coveo provides any OSS cloud services for free.
(I will investigate that)
japhb If we don't want to host it ourselves, we might be able to make use of DuckDuckGo. 02:38
06:02 reportable6 left 06:04 reportable6 joined 06:44 patrickb joined 08:02 patrickb left 09:22 Altai-man joined 09:39 Altai-man left 10:39 linkable6 left, evalable6 left 10:40 evalable6 joined 10:41 linkable6 joined 11:29 brown121407 is now known as smarton 12:02 reportable6 left 12:04 reportable6 joined
Geth rakudo: bdc5092fa0 | (Elizabeth Mattijsen)++ | 2 files
Abstract some SlippyIterator logic

  - add method "push-rest" for pushing the rest of a Slip to a target
  - add method "sink-rest: for sinking the rest of a Slip
  - use these in the iterators where applicable
This reduces the bytecode footprint of some iterator methods increasing the chance they will be inlined, as well as becoming more maintainable.
13:04 evalable6 left, linkable6 left 13:06 linkable6 joined 13:07 evalable6 joined
[TuxCM] Rakudo v2021.06-48-gbdc5092fa (v6.d) on MoarVM 2021.06-16-g74e828f0a
csv-ip5xs0.870 - 0.889
csv-ip5xs-209.393 - 9.580
csv-parser29.757 - 32.454
csv-test-xs-200.375 - 0.401
test8.380 - 8.624
test-t2.083 - 2.168
test-t --race1.004 - 1.150
test-t-2034.754 - 37.160
test-t-20 --race11.225 - 11.507
Geth rakudo: 1bd0411c0f | (Elizabeth Mattijsen)++ | 3 files
Remove unneeded attribute from SlippyIterator

The attribute holding the iterator to be used, is now the flag whether the Slip being processed is exhausted or not.
The name of the $!slip-iter was changed to $!slipper to make sure that any client of the SlippyIterator role (which is an implementation detail) will break.
lizmat unless changed of course :-)
Geth roast: fb61fcfc26 | (Elizabeth Mattijsen)++ | spectest.data
Remove S14-roles/generic-subtyping.t for now

This test file needs work. And I'm not sure what it needs. But it fails atm.
14:09 frost left 14:14 squashable6 left 14:16 squashable6 joined
lizmat hmmm... do we have some kind of sentinel that we can call, but doesn't do anything ? 14:55
japhb I'd say identity, but that's all over the place 16:44
tonyo japhb: that's easily doable..i was doing something like that using virtualbox and automated testing for zef a long time ago 17:27
re: module quality 17:28
Geth rakudo/add-two-error-message-tests: 5d8158b535 | Altai-man++ | t/05-messages/03-errors.t
Add two tests for proper error messages

One is for printing something to a close handle. Two is for a weird case of adverbing.
¦ rakudo/add-two-error-message-tests: review: github.com/rakudo/rakudo/commit/5d8158b535 17:30
rakudo: Altai-man++ created pull request #4447:
Add two tests for proper error messages
japhb tonyo: Did you turn that into a public service, or was that just for zef work? 17:34
Geth roast/add-test-for-2920: e8e0028c6b | Altai-man++ | S03-sequence/misc.t
Add test for #2920
roast: Altai-man++ created pull request #747:
Add test for #2920
tonyo japhb: i can dig it out of the archives and make it public 17:39
the goal was a bit different but not that difficult to update
it was essentially when zef and fez were one utility (~6 or so years ago) and the zef website showed a matrix of a few OSes + raku versions + test results 17:40
17:43 evalable6 left 17:44 evalable6 joined
Geth roast/add-test-for-2920: c159204c8d | Altai-man++ | S32-list/create.t
Add test for #2920
17:46 Kaiepi left, Kaiepi joined
Geth roast/add-test-for-2060: fe65760bfc | Altai-man++ | S32-list/create.t
Add test for #2060
roast: Altai-man++ created pull request #748:
Add test for #2060
sena_kun bisectable6, {:foo, (if 0.5.rand < 1 { :bar })} 17:52
bisectable6 sena_kun, Will bisect the whole range automagically because no endpoints were provided, hang tight
sena_kun, Output on all releases: gist.github.com/35c8b1528fec9b7d4d...1353576794
sena_kun, Bisecting by exit code (old=2018.12 new=2019.03.1). Old exit code: 1
sena_kun, bisect log: gist.github.com/63ebb653ea79373dc8...2ef33ffaf5 17:53
sena_kun, (2019-01-04) github.com/rakudo/rakudo/commit/31...63b0942e75
sena_kun, Output on all releases and bisected commits: gist.github.com/91f590c3342273c26d...0037ce5387
Geth roast/test-for-1486: 3fd712f501 | Altai-man++ | S09-hashes/objecthash.t
Add test for #1486
17:55 Kaiepi left, Kaiepi joined
Geth roast/add-test-for-1890: 96d879a2f7 | Altai-man++ | S03-metaops/misc.t
Add test for #1890
roast: Altai-man++ created pull request #749:
Add test for #1890
18:03 reportable6 left 18:05 reportable6 joined
tonyo japhb: of how much use is something like that? 18:52
19:08 Kaipi joined, Kaiepi left
japhb tonyo: It would cover a different portion of the problem space than Blin -- whereas the latter helps to find regressions in Rakudo that would block a release by testing the ecosystem against it, but is often confounded by regressions in the ecosystem modules themselves, it sounds like what you have would help ensure that the ecosystem modules were in a better average state. 22:01
22:38 linkable6 left, evalable6 left 22:40 linkable6 joined, evalable6 joined 22:43 linkable6 joined, ilogger2_ left 22:44 ilogger2 joined 22:53 Kaipi left, Kaipi joined
[Coke] does blin automatically run latest ecosystem modules or last release, or last version we ran with and try latest if *that* works... or? 23:01
vrurg [Coke]: No it doesn't need manual work. I was considering setting up a nightly run. But it can't automatically post the resulting report, so I postponed the idea for a while. 23:35
There is actually a period after "doesn't". So, I mean it needs manual work to be ran. 23:36
23:41 kjp left