#general
- w
witty-crayon-22786
09/21/2015, 4:45 PMyea, doesn't work for analysis parsing at the moment. - w
witty-crayon-22786
09/21/2015, 4:45 PMand you confirmed that it was slower? - e
enough-analyst-54434
09/21/2015, 4:45 PM@fancy-queen-20734: apropos of? - b
bored-art-40741
09/21/2015, 4:47 PM@witty-crayon-22786: We didn't do heavy perf testing with name hashing since we already had so many axes - w
witty-crayon-22786
09/21/2015, 4:48 PMyea, makes sense. it doesn't look like the parsing is hard to fix - b
bored-art-40741
09/21/2015, 4:48 PMBut my general feel on a couple of runs was that it was slower - w
witty-crayon-22786
09/21/2015, 4:48 PMand certainly not hard post global - b
bored-art-40741
09/21/2015, 4:48 PMSweet - f
fancy-queen-20734
09/21/2015, 4:52 PM@enough-analyst-54434: eg, if two versions eg 1 and 2, of a jar are in a given resolve pants+ivy will fix the version to one that is compatible modulo force, etc—but it doesn’t update the coordinates of the resolved jars that end up in the classpath so the dependency on version 1.0 will contain a coordinate org.example…:1.0, but it’s artifact path will point to 2 - e
enough-analyst-54434
09/21/2015, 4:52 PMold or new bug? - e
enough-analyst-54434
09/21/2015, 4:54 PMI guess I'm asking is this an observation or do you want me to take action or... ? - f
fancy-queen-20734
09/21/2015, 4:54 PMnot sure - e
enough-analyst-54434
09/21/2015, 4:54 PMk - f
fancy-queen-20734
09/21/2015, 4:54 PMI’m going to see if I can fix it up as part of the thing I’m working on now - e
enough-analyst-54434
09/21/2015, 4:54 PMMy guess is pants+ivy is misleading and this is a pants bug - e
enough-analyst-54434
09/21/2015, 4:54 PMsince it hand-jiggers force in particular before the resolve - e
enough-analyst-54434
09/21/2015, 4:55 PMWarning that I have a large change in this space - e
enough-analyst-54434
09/21/2015, 4:55 PMThat trickles from killing JarDependency.artifacts - b
bored-art-40741
09/21/2015, 4:56 PMProbably pants, but never discount ivy's capacity for insane bugs - e
enough-analyst-54434
09/21/2015, 4:57 PMI still now of only 1 - e
enough-analyst-54434
09/21/2015, 4:57 PMI know of much more pants misuse of ivy - w
witty-crayon-22786
09/21/2015, 4:58 PM@enough-analyst-54434: confirmed with the *twadoop guys: one monthish away from consuming idl directly - e
enough-analyst-54434
09/21/2015, 4:58 PMK - e
enough-analyst-54434
09/21/2015, 4:58 PMI have JarPublish mostly sorted keeping the status quo - w
witty-crayon-22786
09/21/2015, 4:58 PMthanks for waiting. - f
full-kilobyte-93429
09/21/2015, 5:17 PMhas anyone encountered coverage test stuck? https://coveralls.io/builds/3623386 it's overnight, all jobs are done though - f
full-kilobyte-93429
09/21/2015, 5:18 PMor if there is a way to restart? - e
enough-analyst-54434
09/21/2015, 5:21 PMI've seen non-responsive coveralls reports for a github PR, but I don;t know how to "restart" or what the cause is - e
enough-analyst-54434
09/21/2015, 5:21 PMWe ship over the coverage data all ready to go to coveralls at the tail of the 2 unit test shards - e
enough-analyst-54434
09/21/2015, 5:21 PMThere is almost no processing needed by coveralls after that