Improvements to `-Vimplicits`; errors now show complete implicit search tree #9944

tribbloid · 2022-02-19T23:00:39Z

Backports bugfixes from splain 0.5.8 & 1.0.0

The new algorithm now build a tree instead of an indented list that records implicit search process. As a result, implicit terms being considered multiple times in a search will be displayed repeatedly.

E.g. for the following scala code:

object tpes
{
  trait I1
  trait I2
  trait I3
  trait I4
  trait I5
  trait I6
  trait I7
  trait I8
  trait I9
}
import tpes._

object Tree
{
  implicit def i8(implicit p: I9): I8 = ???
  implicit def i7(implicit p: I8): I7 = ???
  implicit def i6a(implicit p: I7): I6 = ???
  implicit def i6b(implicit p: I8): I6 = ???
  implicit def i5(implicit p: I6): I5 = ???
  implicit def i4(implicit p: I5): I4 = ???
  implicit def i3a(implicit p: I4): I3 = ???
  implicit def i3b(implicit p: I4): I3 = ???
  implicit def i2(implicit p: I3): I2 = ???
  implicit def i1a(implicit p: I2): I1 = ???
  implicit def i1b(implicit p: I6): I1 = ???
  implicitly[I1]
}

The old error message can only show term i6b once, this is problematic as i6b: I6 was summoned for multiple p: I6 arguments:

newSource1.scala:28: error: implicit error;
!I e: tpes.I1
i1a invalid because
!I p: tpes.I2
――i2 invalid because
  !I p: tpes.I3
――――i3a invalid because
    !I p: tpes.I4
――――――i4 invalid because
      !I p: tpes.I5
――――――――i5 invalid because
        !I p: tpes.I6
――――――――――i6a invalid because
          !I p: tpes.I7
――――――――――――i7 invalid because
            !I p: tpes.I8
――――――――――――――i8 invalid because
              !I p: tpes.I9

――――――――――i6b invalid because
          !I p: tpes.I8
――――――――――――i8 invalid because
            !I p: tpes.I9

――――i3b invalid because
    !I p: tpes.I4
――――――i4 invalid because
      !I p: tpes.I5
――――――――i5 invalid because
        !I p: tpes.I6
――――――――――i6a invalid because
          !I p: tpes.I7
――――――――――――i7 invalid because
            !I p: tpes.I8
――――――――――――――i8 invalid because
              !I p: tpes.I9

i1b invalid because
!I p: tpes.I6
――i6a invalid because
  !I p: tpes.I7
――――i7 invalid because
    !I p: tpes.I8
――――――i8 invalid because
      !I p: tpes.I9
  implicitly[I1]
            ^

after this patch, the error message becomes:

newSource1.scala:28: error: implicit error;
!I e: tpes.I1
i1a invalid because
!I p: tpes.I2
――i2 invalid because
  !I p: tpes.I3
――――i3a invalid because
    !I p: tpes.I4
――――――i4 invalid because
      !I p: tpes.I5
――――――――i5 invalid because
        !I p: tpes.I6
――――――――――i6a invalid because
          !I p: tpes.I7
――――――――――――i7 invalid because
            !I p: tpes.I8
――――――――――――――i8 invalid because
              !I p: tpes.I9
――――――――――i6b invalid because
          !I p: tpes.I8
――――――――――――i8 invalid because
            !I p: tpes.I9
――――i3b invalid because
    !I p: tpes.I4
――――――i4 invalid because
      !I p: tpes.I5
――――――――i5 invalid because
        !I p: tpes.I6
――――――――――i6a invalid because
          !I p: tpes.I7
――――――――――――i7 invalid because
            !I p: tpes.I8
――――――――――――――i8 invalid because
              !I p: tpes.I9
――――――――――i6b invalid because
          !I p: tpes.I8
――――――――――――i8 invalid because
            !I p: tpes.I9
i1b invalid because
!I p: tpes.I6
――i6a invalid because
  !I p: tpes.I7
――――i7 invalid because
    !I p: tpes.I8
――――――i8 invalid because
      !I p: tpes.I9
――i6b invalid because
  !I p: tpes.I8
――――i8 invalid because
    !I p: tpes.I9
  implicitly[I1]
            ^

which shows multiple summoning attempts of i6b correctly

SethTisue · 2022-04-20T16:20:39Z

@tribbloid interested in returning to this?

tribbloid · 2022-04-21T17:27:06Z

Of course. I'm just fixing a bug for splain 1.0

SethTisue · 2022-10-26T11:57:12Z

@tribbloid interested in returning to this?

tribbloid · 2022-10-28T16:53:09Z

Yes, don't want to drag it any longer, it will be 2.13.11 or monastery

SethTisue · 2023-01-31T02:21:24Z

I hear that a choice cell just opened up at the monastery ;-)

tribbloid · 2023-01-31T02:26:05Z

NOOOO you can't just drag me in without me submitting it!

Seriously when will be the feature freeze date? I'll just need a week to wrap it up, with all test cases

SethTisue · 2023-01-31T03:34:05Z

when will be the feature freeze date

we're not exactly sure. it's not imminent. but we're coming up on the 4 month mark since 2.13.10, and it would be highly unusual of us to let more than 6 months pass, so...

you're safe for at least 2 or 3 weeks, say. after that, can't predict.

tribbloid · 2023-02-12T23:20:10Z

Backport almost finished and passed partest locally. Let's wait for the CI

Still, this doesn't rule out the monastery as there may be numerous problems in downstream testing, format compliance & release schedule.

SethTisue · 2023-02-13T19:35:02Z

(CI likes the last commit, but you'll need to rebase such that every commit passes — that's the "combined" check.)

…on par with splain 1.0.0

tribbloid · 2023-02-14T20:01:14Z

Sorry for being slow, Travis CI is getting a bit squishy, got "504 Gateway Time-out" in my last run.

SethTisue · 2023-02-14T20:46:46Z

Don't worry about the Travis-CI failure — it's unrelated to your changes. I've opened scala/scala-dev#832 on it

Let us know if you consider this ready for review

tribbloid · 2023-02-14T20:48:37Z

thanks a lot! Will remove "Draft" tag once I'm comfortable

tribbloid · 2023-02-16T22:21:54Z

Finished reviewing, the only issue left is formatting difference. Is there a formatter configuration I could use to enforce it automatically?

…on par with splain 1.0.0

SethTisue · 2023-02-16T23:02:49Z

We don't have a repo-wide style, no. We try to match a file's existing style when tweaking existing code, prioritizing minimal/reviewable diffs. But if you're really tearing something up you can suit yourself as long as it's within the (admittedly nebulous) boundaries of common, consensus style.

SethTisue · 2023-02-17T03:20:17Z

Oh, one thing we'll want before merge is to have the PR title and description be your best effort at documenting these changes from the user's point of view.

…now show complete implicit search tree

tribbloid · 2023-02-21T22:17:37Z

Oh, one thing we'll want before merge is to have the PR title and description be your best effort at documenting these changes from the user's point of view.

I believe it has been done. @SethTisue Can we start the review if it is up to your standard?

tribbloid · 2023-02-21T22:19:45Z

@tek.

This patch only synchronise behaviour & test cases of all features of splain-plugin before 1.x. New features are not included

SethTisue · 2023-02-28T18:40:39Z

I revised the PR title and description a bit further, but I think the PR description could use further work, both for reviewers and for end users. Can you expand it to show an example of the improved behavior? "Suppose we're debugging the following error. The following implicit search tree is now produced, and it shows that the problem is X", something like that maybe?

tribbloid · 2023-03-01T09:05:46Z

sounds good, updated

tribbloid · 2023-03-08T16:59:50Z

Any plan to integrate it before the deadline?

SethTisue · 2023-03-08T17:23:26Z

I don't anticipate any trouble with this making 2.13.11

SethTisue · 2023-03-13T21:49:52Z

thank you!

scala-jenkins added this to the 2.13.9 milestone Feb 19, 2022

tribbloid marked this pull request as draft February 19, 2022 23:01

SethTisue modified the milestones: 2.13.9, 2.13.10 Apr 25, 2022

tribbloid force-pushed the splain1Backport/dev1 branch from 2461e39 to a7220ae Compare September 1, 2022 17:13

lrytz modified the milestones: 2.13.10, 2.13.11 Sep 26, 2022

tribbloid force-pushed the splain1Backport/dev1 branch from a7220ae to b99c1f9 Compare November 14, 2022 01:22

tribbloid force-pushed the splain1Backport/dev1 branch from b99c1f9 to 4b3c5c2 Compare February 12, 2023 23:13

tribbloid force-pushed the splain1Backport/dev1 branch from 5362cc3 to 37bb980 Compare February 14, 2023 18:22

tribbloid added a commit to tribbloid/scala that referenced this pull request Feb 14, 2023

scala#9944: upgrade SplainFormatting and relevant test partest to be …

7117ee4

…on par with splain 1.0.0

tribbloid force-pushed the splain1Backport/dev1 branch from 37bb980 to 7117ee4 Compare February 14, 2023 19:29

tribbloid added a commit to tribbloid/scala that referenced this pull request Feb 16, 2023

scala#9944: upgrade SplainFormatting and relevant test partest to be …

cb6b3a1

…on par with splain 1.0.0

tribbloid force-pushed the splain1Backport/dev1 branch from 7117ee4 to cb6b3a1 Compare February 16, 2023 22:35

tribbloid changed the title ~~[work in progress, DO NOT MERGE] Backport bugfixes from splain 0.5.8-SNAPSHOT & 1.0.0~~ Backport bugfixes from splain 0.5.8-SNAPSHOT & 1.0.0 Feb 16, 2023

tribbloid marked this pull request as ready for review February 16, 2023 22:37

tribbloid changed the title ~~Backport bugfixes from splain 0.5.8-SNAPSHOT & 1.0.0~~ Backport bugfixes from splain 0.5.8 & 1.0.0, errors from -VImplicits now show complete implicit search tree Feb 20, 2023

Backport bugfixes from splain 0.5.8 & 1.0.0, errors from -VImplicits …

6573972

…now show complete implicit search tree

tribbloid force-pushed the splain1Backport/dev1 branch from cb6b3a1 to 6573972 Compare February 20, 2023 04:47

SethTisue changed the title ~~Backport bugfixes from splain 0.5.8 & 1.0.0, errors from -VImplicits now show complete implicit search tree~~ Improvements to -Vimplicits; errors now show complete implicit search tree Feb 28, 2023

SethTisue added the prio:hi high priority (used only by core team, only near release time) label Mar 8, 2023

SethTisue merged commit f3bfc05 into scala:2.13.x Mar 13, 2023

SethTisue added release-notes worth highlighting in next release notes and removed prio:hi high priority (used only by core team, only near release time) labels May 3, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Improvements to `-Vimplicits`; errors now show complete implicit search tree #9944

Improvements to `-Vimplicits`; errors now show complete implicit search tree #9944

tribbloid commented Feb 19, 2022 •

edited

SethTisue commented Apr 20, 2022

tribbloid commented Apr 21, 2022 •

edited by SethTisue

SethTisue commented Oct 26, 2022

tribbloid commented Oct 28, 2022

SethTisue commented Jan 31, 2023

tribbloid commented Jan 31, 2023

SethTisue commented Jan 31, 2023

tribbloid commented Feb 12, 2023

SethTisue commented Feb 13, 2023

tribbloid commented Feb 14, 2023

SethTisue commented Feb 14, 2023 •

edited

tribbloid commented Feb 14, 2023

tribbloid commented Feb 16, 2023

SethTisue commented Feb 16, 2023

SethTisue commented Feb 17, 2023

tribbloid commented Feb 21, 2023

tribbloid commented Feb 21, 2023

SethTisue commented Feb 28, 2023 •

edited

tribbloid commented Mar 1, 2023

tribbloid commented Mar 8, 2023

SethTisue commented Mar 8, 2023

SethTisue commented Mar 13, 2023

Improvements to -Vimplicits; errors now show complete implicit search tree #9944

Improvements to -Vimplicits; errors now show complete implicit search tree #9944

Conversation

tribbloid commented Feb 19, 2022 • edited

SethTisue commented Apr 20, 2022

tribbloid commented Apr 21, 2022 • edited by SethTisue

SethTisue commented Oct 26, 2022

tribbloid commented Oct 28, 2022

SethTisue commented Jan 31, 2023

tribbloid commented Jan 31, 2023

SethTisue commented Jan 31, 2023

tribbloid commented Feb 12, 2023

SethTisue commented Feb 13, 2023

tribbloid commented Feb 14, 2023

SethTisue commented Feb 14, 2023 • edited

tribbloid commented Feb 14, 2023

tribbloid commented Feb 16, 2023

SethTisue commented Feb 16, 2023

SethTisue commented Feb 17, 2023

tribbloid commented Feb 21, 2023

tribbloid commented Feb 21, 2023

SethTisue commented Feb 28, 2023 • edited

tribbloid commented Mar 1, 2023

tribbloid commented Mar 8, 2023

SethTisue commented Mar 8, 2023

SethTisue commented Mar 13, 2023

Improvements to `-Vimplicits`; errors now show complete implicit search tree #9944

Improvements to `-Vimplicits`; errors now show complete implicit search tree #9944

tribbloid commented Feb 19, 2022 •

edited

tribbloid commented Apr 21, 2022 •

edited by SethTisue

SethTisue commented Feb 14, 2023 •

edited

SethTisue commented Feb 28, 2023 •

edited