Skip to content

Conversation

@szeiger
Copy link
Contributor

@szeiger szeiger commented Feb 16, 2017

Types in the applicable overload alternatives need to be seen from
the respective owners of the individual alternative, not from the
target’s owner (which can be a subtype of the types that define the
methods).

Copy link
Contributor

@adriaanm adriaanm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good catch! Would you mind doing the refactoring I suggested if it makes sense?

def functionProto(argTps: List[Type]): Type =
try functionType(funArgTypes(argTps).transpose.map(lub), WildcardType)
def functionProto(argTpWithAlt: List[(Type, Symbol)]): Type =
try functionType(funArgTypes(argTpWithAlt.map { case (tp, alt) => tp.asSeenFrom(pre, alt.owner) }).transpose.map(lub), WildcardType)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why not keep theasSeenFrom in funArgTypes? There's a lot going on on this line already.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I tried to simplify it but that didn't really work out.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@szeiger I don't understand what didn't work out - it seems it should be easy to move the asSeenFrom into funArgTypes?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@lrytz That's where it used to be. Moving it out didn't really make the code any simpler.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Moving it out didn't really make the code any simpler.

Then why was it moved out? I feel like we're talking past each other.. :)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's such a detail that I'm sorry to go back and forth so many times.. I just don't understand.. IIUC adriaan suggests to pass argTpWithAlt as argument to funArgTypes directly and do the asSeenFrom there, which also looks better to me. You even seem to agree, as you say

Moving it out [of funArgTypes?] didn't really make the code any simpler

so why did you move it out?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Because I expected it would. There really isn't any deeper or better explanation...

Copy link
Contributor

@adriaanm adriaanm Feb 20, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Updated (moving asSeenFrom back down into funArgTypes)

I'm not seeing this update in the diff.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Huh, still had some unstaged changes lying around. Updated again.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That explains my confusion :-)

@szeiger
Copy link
Contributor Author

szeiger commented Feb 17, 2017

Updated (moving asSeenFrom back down into funArgTypes) and rebased (to clear up the MiMa failure).

@lrytz
Copy link
Member

lrytz commented Feb 20, 2017

/rebuild - instable test, scala/scala-dev#296

Types in the applicable overload alternatives need to be seen from
the respective owners of the individual alternative, not from the
target’s owner (which can be a subtype of the types that define the
methods).
@lrytz lrytz merged commit 76bfb9e into scala:2.12.x Feb 21, 2017
SethTisue added a commit to SethTisue/scala that referenced this pull request Mar 6, 2017
only trivial merge conflict involving a method renaming

*   d1d700e - (origin/HEAD, origin/2.12.x) Merge pull request scala#5754 from Philippus/issue/html-tag-in-hover (2 days ago) <Lukas Rytz>
|\
| * cf64718 - pattern for entitylink was too narrow, cleaned up the tests (3 days ago) <Philippus Baalman>
* |   f771395 - Merge pull request scala#5671 from retronym/topic/stubby-2 (3 days ago) <Lukas Rytz>
|\ \
| * | ad13063 - Remove non-essential fix for stub symbol failure (4 days ago) <Jason Zaugg>
| * | 7f2d1fa - Avoid forcing info transforms of primitive methods (2 weeks ago) <Jason Zaugg>
| * | 37a0eb7 - Avoid stub symbol related crash in backend (2 weeks ago) <Jason Zaugg>
* | |   96a7617 - Merge pull request scala#5622 from edmundnoble/extra-errs (4 days ago) <Adriaan Moors>
|\ \ \
| * | | 466e52b - Match error lengths (4 weeks ago) <Edmund Noble>
| * | | d1fc983 - Improved error messages for identically named, differently prefixed types (9 weeks ago) <Edmund Noble>
|  / /
* | |   f2e05c2 - Merge pull request scala#5728 from Philippus/issue/html-tag-in-hover (4 days ago) <Lukas Rytz>
|\ \ \
| | |/
| |/|
| * | e3c5df8 - added missing Inline matches to inlineToStr so it is now exhaustive scala.xml.XML.loadString(tag).text will remove all html tags inside the HtmlTag (9 days ago) <Philippus Baalman>
* | |   920bc4e - Merge pull request scala#5743 from som-snytt/issue/10207-bad-update (7 days ago) <Lukas Rytz>
|\ \ \
| * | | 094f7f9 - SI-10207 Error before update conversion (8 days ago) <Som Snytt>
* | | |   1b4d36f - Merge pull request scala#5746 from paulp/pr/partest (7 days ago) <Lukas Rytz>
|\ \ \ \
| |/ / /
|/| | |
| * | | eac00e1 - Add partest paths to the list of watched sources. (8 days ago) <Paul Phillips>
|/ / /
* | |   5f1a638 - Merge pull request scala#5732 from retronym/topic/build-info-malarkey (10 days ago) <Adriaan Moors>
|\ \ \
| * | | 5e9acac - More predictable performance of SBT build startup, reload (11 days ago) <Jason Zaugg>
|  / /
* | |   759a7b7 - Merge pull request scala#5735 from SethTisue/sd-313 (10 days ago) <Adriaan Moors>
|\ \ \
| * | | ed4ddf5 - increase timeouts on some sys.process tests (11 days ago) <Seth Tisue>
* | | |   e2aaf2c - Merge pull request scala#5723 from dragos/issue/regression-assert-ide (11 days ago) <Lukas Rytz>
|\ \ \ \
| |/ / /
|/| | |
| * | | e5c957e - Fix regression in 5751763 (12 days ago) <Iulian Dragos>
* | | |   f174bfb - Merge pull request scala#5731 from janekdb/issue/scalaGH-644/fix-spec-latex-rendering (11 days ago) <Seth Tisue>
|\ \ \ \
| * | | | ba4c6d4 - scalaGH-644: Remove static html styling of spec code blocks (11 days ago) <Janek Bogucki>
|/ / / /
* | | |   8e40bef - Merge pull request scala#5729 from scala/revert-5658-topic/hashhash (12 days ago) <Adriaan Moors>
|\ \ \ \
| |_|/ /
|/| | |
| * | | 86cd70f - (origin/revert-5658-topic/hashhash) Revert "Fix erasure of the qualifier of ##" (12 days ago) <Adriaan Moors>
|/ / /
* | |   cbf7daa - Merge pull request scala#5681 from Philippus/issue/9704 (13 days ago) <Lukas Rytz>
|\ \ \
| * | | b8a8ac1 - moved Pattern and TagsNotToClose to a HtmlTag companion object (13 days ago) <Philippus Baalman>
| * | | a019082 - SI-9704 don't add a closed HtmlTag if it is already closed (4 weeks ago) <Philippus Baalman>
|  / /
* | |   effde0c - Merge pull request scala#5726 from scala/revert-5629-issue/10120-quote-err (13 days ago) <Adriaan Moors>
|\ \ \
| * | | d60f6e3 - (origin/revert-5629-issue/10120-quote-err) Revert "SI-10133 Require escaped single quote char lit" (13 days ago) <Adriaan Moors>
|/ / /
* | |   a8c4a54 - Merge pull request scala#5663 from gourlaysama/ticket/sd-256-enable-repl-colors-unix-2 (13 days ago) <Adriaan Moors>
|\ \ \
| * | | 6411170 - SD-256 enable colored output by default on unix (13 days ago) <Antoine Gourlay>
* | | |   c96a977 - Merge pull request scala#5658 from retronym/topic/hashhash (13 days ago) <Lukas Rytz>
|\ \ \ \
| |/ / /
|/| | |
| * | | f85c62e - Fix erasure of the qualifier of ## (6 weeks ago) <Jason Zaugg>
|  / /
* | |   76bfb9e - Merge pull request scala#5708 from szeiger/issue/si10194 (13 days ago) <Lukas Rytz>
|\ \ \
| * | | 1d22ee4 - SI-10194: Fix abstract type resolution for overloaded HOFs (13 days ago) <Stefan Zeiger>
|  / /
* | |   dabec1a - Merge pull request scala#5700 from retronym/ticket/10154-refactor (13 days ago) <Lukas Rytz>
|\ \ \
| * | | 06eee79 - Refactor implementation of lookupCompanion (2 weeks ago) <Jason Zaugg>
|  / /
* | |   2f1e0c2 - Merge pull request scala#5704 from som-snytt/issue/10190-elide-string (13 days ago) <Lukas Rytz>
|\ \ \
| * | | 6fb3825 - SI-10190 Elide string to empty instead of null (3 weeks ago) <Som Snytt>
|  / /
* | |   13f7b2a - Merge pull request scala#5640 from optimizely/repl-import-handler (2 weeks ago) <Adriaan Moors>
|\ \ \
| * | | aa7e335 - Fix ImportHandler's reporting of importedNames and importedSymbols (8 weeks ago) <Hao Xia>
| * | | c89d821 - Fix SIOOBE in Name#pos for substrings of length 1 (8 weeks ago) <Jason Zaugg>
|  / /
* | |   023a96a - Merge pull request scala#5629 from som-snytt/issue/10120-quote-err (2 weeks ago) <Adriaan Moors>
|\ \ \
| * | | 05cc3e2 - SI-10120 ReplReporter handles message indent (7 weeks ago) <Som Snytt>
| * | | 939abf1 - SI-10120 Extra advice on unclosed char literal (8 weeks ago) <Som Snytt>
| * | | 855492c - SI-10133 Require escaped single quote char lit (8 weeks ago) <Som Snytt>
|  / /
* | |   e21ab42 - Merge pull request scala#5660 from som-snytt/issue/9464-spec (2 weeks ago) <Adriaan Moors>
|\ \ \
| * | | a6dcceb - SI-9464 Clarify spec on no final trait (6 weeks ago) <Som Snytt>
|  / /
* | |   e87a436 - Merge pull request scala#5659 from retronym/ticket/10026 (2 weeks ago) <Adriaan Moors>
|\ \ \
| |/ /
|/| |
| * | 777a0e5 - SI-10026 Fix endless cycle in runtime reflection (2 weeks ago) <Jason Zaugg>
| |/
* |   23e5ed9 - Merge pull request scala#5707 from retronym/topic/java9-prepare (2 weeks ago) <Lukas Rytz>
|\ \
| * | 96e8e97 - Workaround bug in Scala runtime reflection on JDK 9 (3 weeks ago) <Jason Zaugg>
| * | fe2d9a4 - Avoid ambiguous overload on JDK 9 (3 weeks ago) <Jason Zaugg>
| * | 6bba8f7 - Adapt to change in ClassLoader in JDK 9 (3 weeks ago) <Jason Zaugg>
| * | 8136057 - Bump scala-asm version (3 weeks ago) <Jason Zaugg>
|  /
* |   cad3c3d - Merge pull request scala#5709 from adriaanm/sam_wild_bound (2 weeks ago) <Lukas Rytz>
|\ \
| * | c396e96 - Ignore BoundedWildcardType in erasure type map (2 weeks ago) <Adriaan Moors>
* | |   3e9df41 - Merge pull request scala#5711 from retronym/ticket/jrt (2 weeks ago) <Lukas Rytz>
|\ \ \
| * | | 09c7edc - Faster and simpler Java 9 classpath implementation (2 weeks ago) <Jason Zaugg>
|  / /
* | |   7b9d3cc - Merge pull request scala#5713 from janekdb/issue/scalaGH-644/sync-jekyll-README-to-Gemfile (2 weeks ago) <Lukas Rytz>
|\ \ \
| * | | 5e5ec9a - scalaGH-644: Expand note regarding Jekyll versions (2 weeks ago) <Janek Bogucki>
|  / /
* | |   144f7e0 - Merge pull request scala#5714 from dragos/issue/usage-sterr-SI-10178 (2 weeks ago) <Lukas Rytz>
|\ \ \
| |/ /
|/| |
| * | 640c85e - SI-10178 Route reporter.echo to stdout (2 weeks ago) <Iulian Dragos>
|  /
* |   2fec08b - Merge pull request scala#5717 from som-snytt/issue/10148-followup (2 weeks ago) <Adriaan Moors>
|\ \
| |/
|/|
| * f3d271b - SI-10148 Accept verbose zero (2 weeks ago) <Som Snytt>
* 147e5dd - Merge pull request scala#5716 from adriaanm/i296 (2 weeks ago) <Jason Zaugg>
* 12437a0 - Ensure ordering for args to `choose` in DurationTest (2 weeks ago) <Adriaan Moors>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants