Remove meaningless metapkg. #39
Conversation
It was added recently sniekum#120 when consolidating two `ar_track_alvar*` repositories into a single repo. What users need is `ar_track_alvar`, which depends on `ar_track_alvar_msgs` that is to be installed altogether, so there's no use of meta package.
|
Triggering Travis CI (in addition to ROS buildfarm CI) by close-reopening. |
|
Pull request testing failed ONLY on buildfarm (similar to the failure in #37), which I asked on answers. |
…s not what is intended. Intead, use a method that evaluates sigfig.
…o far from the measurement. New value is suggested based on the following: Adding all measurements from failed testcases in https://travis-ci.org/ros-perception/ar_track_alvar/jobs/237663084#L933 and http://build.ros.org/job/Lpr__ar_track_alvar__ubuntu_xenial_amd64/11/consoleFull and get the average. z = (0.003014730238554591 + 0.001480243522160433 + 0.006115145186701975 + 0.00505591146182839 + 0.0032586824232304266 + 0.0004646619896828134 + 0.007629782810998689 + 0.002663673148269637) / 7 = 0.00424040439
In the referenced thread I got a solution that I implemented in 7a4dd3f. Then the test started to fail on both Travis and buildfarm, so I needed f5fcaca and 3a7b40c. Basically the testcases were (and still are) a bit strict to require milli meter degree of precision for With them, buildfarm still fails but without obvious reason, which I asked again here. |
|
@ros-pull-request-builder retest this please |
1 similar comment
|
@ros-pull-request-builder retest this please |
|
I opened a new ticket for the issue #39 (comment). Looks like it's an issue of how a testcase that uses particular node tries to write the test result into xml. Not even sure how critical that issue is, but merging this assuming it is not. |
It was added recently sniekum#120 when consolidating two
ar_track_alvar*repositories into a single repo. What users need isar_track_alvar, which depends onar_track_alvar_msgsthat is to be installed altogether, so there's no use of meta package.