#general
- e
enough-analyst-54434
09/11/2015, 9:32 PMand think 0-times - b
bright-dinner-81701
09/11/2015, 9:35 PMok - e
enough-analyst-54434
09/11/2015, 9:37 PMSince the contribs are small and self-contained I tend to just./pants test contrib/node:: - b
bright-dinner-81701
09/11/2015, 9:37 PMyeah cool thats kinda what i just concluded as well - f
fancy-queen-20734
09/11/2015, 9:52 PMI want to write an test where I compile a source file, modify it, and recompile. Are there any examples of that pattern? I mostly see things using testprojects, but I feel like I shouldn’t be modifying those directly—assuming they are shared test fixtures - b
bored-art-40741
09/11/2015, 9:54 PMYou can completely generate + touch your source in a scratch buildroot - w
witty-crayon-22786
09/11/2015, 9:54 PMyea. testprojects only as last resort - w
witty-crayon-22786
09/11/2015, 9:55 PM...although integration tests mostly use the existing buildroot and a temporary
directory....pants.d - b
bright-dinner-81701
09/11/2015, 9:55 PMuh so - b
bright-dinner-81701
09/11/2015, 9:56 PMhttps://travis-ci.org/pantsbuild/pants/builds/79928107 passed for my pull request - b
bright-dinner-81701
09/11/2015, 9:56 PMhowever if i remove node from my path locally and run "./pants test contrib/node::” i noticed it actually fails locally - b
bright-dinner-81701
09/11/2015, 9:57 PMbecause pants doesnt put the bootstrapped node on my PATH, like i was saying the other day - b
bright-dinner-81701
09/11/2015, 9:57 PMso it seems it tries to install all my modules in my example dir - b
bright-dinner-81701
09/11/2015, 9:57 PMwhich i didnt really expect since we dont have any test targets that use them yet... - w
witty-crayon-22786
09/11/2015, 9:59 PM"because pants doesnt put the bootstrapped node on my PATH, like i was saying the other day" <- sounds like you know how to fix it? - b
bright-dinner-81701
09/11/2015, 9:59 PMhaha ok - b
bright-dinner-81701
09/11/2015, 9:59 PMso i guess my pull request can include that too - w
witty-crayon-22786
09/11/2015, 10:00 PMwould be better as a small independent fix probably - b
bright-dinner-81701
09/11/2015, 10:00 PMok… i guess it might block merging this branch - e
enough-analyst-54434
09/11/2015, 10:01 PM@fancy-queen-20734: 2 seconds - pretty sure I did this somewhere - e
enough-analyst-54434
09/11/2015, 10:06 PMbad memory, just coached Jin Feng to do a compile here: https://github.com/pantsbuild/pants/blob/master/tests/python/pants_test/backend/jvm/tasks/test_junit_run.py#L118-118 - e
enough-analyst-54434
09/11/2015, 10:06 PMBut not in the context of compile, edit, compile - e
enough-analyst-54434
09/11/2015, 10:11 PM@bright-dinner-81701: it may be clear by now, but
has a plan that includes resolve (resolve <- compile <- test), and the :: includes all the new examples - so those are run thorugh./pants test contrib/node::
which does npm installsresolve.npm - e
enough-analyst-54434
09/11/2015, 10:13 PMDoes anyone have
set in any of their targets on disk out there?: https://github.com/pantsbuild/pants/blob/master/src/python/pants/backend/jvm/targets/jvm_target.py#L35configurations= - e
enough-analyst-54434
09/11/2015, 10:14 PMThat goes back to 2014 in git blame - and the attribute is only used by
fwictbundle - e
enough-analyst-54434
09/11/2015, 10:14 PMAnd the 2014 is Travis while at Twitter, so I guess I'm asking tweeps - b
bored-art-40741
09/11/2015, 10:15 PM@enough-analyst-54434: None at 4sq - w
witty-crayon-22786
09/11/2015, 10:16 PMsec - e
enough-analyst-54434
09/11/2015, 10:16 PMthanks for checking folks - w
witty-crayon-22786
09/11/2015, 10:17 PMnone here