| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
This addresses #9854
|
The debugger, or at least this implementation of it, is going away after version 7 of Node.js and it's been a challenge to write robust debugger tests. So I'd be OK with making an exception to the usual "bug-fixes should have tests" rule on this one. Although it might be worth checking if this fixes some of the broken tests in test/debugger. Those tests are not run during make test or during CI runs. A lot of them are broken. :-| But if this fixes one or more of them, then hey, the test has already been written! :-D Sorry for introducing the bug and thanks for fixing my mistake! |
Sorry, something went wrong.
Sorry, something went wrong.
|
LGTM if CI is ✅ |
Sorry, something went wrong.
There was a problem hiding this comment.
Makes sense given the commit that introduced the regression. LGTM as far as I can tell.
Sorry, something went wrong.
|
Hmm CI is not happy. But I can't work out what the issue is with linting. Locally I've run make lint-ci on several versions of node, but haven't been able to reproduce the problem. The CI logs were not completely clear to me. |
Sorry, something went wrong.
|
Let's try a new CI: https://ci.nodejs.org/job/node-test-pull-request/5150/ |
Sorry, something went wrong.
| this.pause(); | ||
|
|
||
| setImmediate(() => { this.run(); }); | ||
| setImmediate(() => { this.run( () => this.resume() ); }); |
There was a problem hiding this comment.
Extra spaces between ( and (), and () and ).
Sorry, something went wrong.
|
Do we have a test for this? |
Sorry, something went wrong.
|
@indutny no test yet. Existing tests did not catch the regression, but I haven't worked out a new test. I don't think I'll be able to get to it today, but I can devise something in the next couple of days if necessary. Edit Test added |
Sorry, something went wrong.
|
And yet a 3rd CI https://ci.nodejs.org/job/node-test-pull-request/5151/ |
Sorry, something went wrong.
There was a problem hiding this comment.
LGTM. Seemingly unrelated flake on the freebsd10-64 buildbot.
Sorry, something went wrong.
| const common = require('../common'); | ||
| const spawn = require('child_process').spawn; | ||
|
|
||
| const timeoutId = setTimeout(function() { |
There was a problem hiding this comment.
Is the timeout needed? Couldn't we just let the test timeout and fail instead of introducing a timer?
Sorry, something went wrong.
There was a problem hiding this comment.
I suppose it is redundant. I was just modeling off of another test that spawns a node process.
Sorry, something went wrong.
There was a problem hiding this comment.
Yea, if you don't need it, I'd remove it. They add complexity to the code and timers have been a source of annoyance on the CI.
Sorry, something went wrong.
|
This should have had another CI run before landing I think. |
Sorry, something went wrong.
|
@cjihrig sorry about that - I did run another CI but didn't link it. Looks like unrelated flakiness on Windows. https://ci.nodejs.org/job/node-test-pull-request/5210/ |
Sorry, something went wrong.
|
I've added the lts-watch-v6.x label. Should this be back ported to v6.x? |
Sorry, something went wrong.
When the debugger has started we need to call `this.resume` otherwise, the prompt won't appear. Fixes: nodejs#9854 PR-URL: nodejs#10099 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Rich Trott <rtrott@gmail.com>
| Back | FazBrowse Home | New Git URL |
Checklist
Affected core subsystem(s)
debugger
Description of change
This addresses #9854
The issue was introduced here and first appeared in 6.2.2.
I haven't figured out the best way to test this yet. Apparently, existing tests didn't catch the regression though.