doc: recompile when testing in CONTRIBUTING.md

[phillipj]

Changes to core modules does not take effect before project gets recompiled. This was not obvious to me when trying to run some tests at first, which made me bother you with an issue a few weeks ago.

Added a line tipping new contributors about this behavior when describing how to to run tests in contribution guide.

Removed jslint from first test command example as jslint are included when running make test.

Fixed wrong path of example stream2-transform test.

[targos]

👍
It may not be obvious to the newcomer (at least it was not to me) that the core library is bundled in the executable.

Maybe you could also fix the command right before from
$ iojs ./test/parallel/test-streams2-transform.js
to
$ ./iojs ./test/parallel/test-streams2-transform.js
The former would use the installed version instead of the newly built one.

[Fishrock123]

+1 to @targos's suggestion.

Also, it may be better to just have ./configure && make -j (-j runs in maximum parallelism i.e. faster), what do you think? :)

[phillipj]

Nice one @targos, I'll fix it shortly.

👍 for make -j. Would running ./configure at every recompile have any benefits compared to prepending it to make jslint test@Fishrock123?

[targos]

I missed something in my suggestion, there is a typo in the file name.
It should be test-stream2-transform.js.

[Fishrock123]

@phillipj not too much, no. Also make test already includes jslint. :)

[phillipj]

Somewhat concerned about advocating maximum parallelism with make as it made my machine come to a grinding halt. make -j8 works perfectly for me, but it depends on individual developer's hardware ofcourse. Any thoughts @Fishrock123?

[Fishrock123]

Hmmm, I've never actually just run make -j, but I think it should run at the number of threads you have?

I'm in favor of just suggesting make -j 8 I think.

[phillipj]

Alright, that should prevent devs getting load of 300 and a non-responding OS as the plain -j did for me.

[phillipj]

I just added PR-URL to the commit.

[bnoordhuis]

LGTM. Is there a reason this hasn't been merged yet?

[Fishrock123]

Is there a reason this hasn't been merged yet?

Not really

[thefourtheye]

LGTM too.

[phillipj]

Just say the word if there's anything I should do to get this landed.

[thefourtheye]

@phillipj Reworded the commit message and landed in c7d8b09. Thanks :-)

[phillipj]

Reworded the commit message

👍 and thanks :)

[targos]

@thefourtheye is there something wrong with your system time? c7d8b09 was committed in the future (today, 5pm UTC)

[targos]

it could also be the reason why your commit wasn't automatically taken for the 2.4.0 changelog

[thefourtheye]

@targos Oh, I am not sure what is wrong. I resynced my machine's time now. Let's see if this happens again. But that commit was landed after 2.4.0 build started, right?

[targos]

@thefourtheye I don't think so, it landed before mine.