Page MenuHomePhabricator

Allow ArcanistExternalLinter to specify flags passed to interpreter
AbandonedPublic

Authored by cspeckmim on Jan 20 2016, 6:54 AM.
Tags
None
Referenced Files
F14047688: D15067.id.diff
Thu, Nov 14, 4:55 AM
F14000302: D15067.diff
Thu, Oct 24, 8:59 PM
F13982930: D15067.diff
Sun, Oct 20, 3:01 AM
F13982929: D15067.diff
Sun, Oct 20, 3:01 AM
F13982925: D15067.diff
Sun, Oct 20, 3:00 AM
F13982878: D15067.id.diff
Sun, Oct 20, 2:37 AM
Unknown Object (File)
Oct 9 2024, 3:07 AM
Unknown Object (File)
Sep 26 2024, 8:51 AM

Details

Reviewers
None
Group Reviewers
Blessed Reviewers
Summary

Currently linters are only capable of specifying flags to pass to the binary but not to the interpreter. This change follows the existing framework for specifying binary flags to allow for passing flags to the interpreter.

NOTE: This change is extracted from D14632
Test Plan

TBD

Diff Detail

Repository
rARC Arcanist
Branch
master
Lint
Lint Passed
Unit
Tests Passed
Build Status
Buildable 10220
Build 12422: arc lint + arc unit

Event Timeline

cspeckmim retitled this revision from to Allow ArcanistExternalLinter to specify flags passed to interpreter.
cspeckmim updated this object.
cspeckmim edited the test plan for this revision. (Show Details)
cspeckmim added a subscriber: avivey.

In the least the ECO form probably needs filled with more detail

@avivey Is there a chance to get this change landed? It seems to be unobtrusive and is needed to have the Java lint'ers in D14632 live outside of upstream as an extension. I can imagine there might be some other lint implementations which require this functionality as well. Let me know if you need any changes, I'm more then willing to make them.

I don't really know - looks like @cspeckmim isn't completely happy with this yet.

A possible work-around is to embed a script in the library that will invoke the jar - I'll write something on the other side.

I pulled this out from D14632 as-is. The only thing I think this needs is possibly UT (though similar functionality doesn't appear to be covered by UT already). I haven't moved on this in a while as I don't have any expectation for this going upstream due to the forward-path indicated by @epriestley.