WikiWatershed / mmw-geoprocessing

A Spark Job Server job for Model My Watershed geoprocessing.
Apache License 2.0
6 stars 6 forks source link

Use SHA version of GeoTrellis from Bintray #4

Closed hectcastro closed 9 years ago

hectcastro commented 9 years ago

The JARs for SHA cb236ac are now published on Bintray, so after adding the azavea/geotrellis repository to this project's build.scala, the GeoTrellis dependencies resolve from Bintray vs. the local Ivy cache.

/cc @jwalgran

jamesmcclain commented 9 years ago

I had to make the following change in order to be able to build starting with a completely empty .ivy2 folder:

diff --git a/project/build.scala b/project/build.scala
index 362fe62..ee08176 100644
--- a/project/build.scala
+++ b/project/build.scala
@@ -51,7 +51,9 @@ object Geoprocessing extends Build {

   val resolutionRepos = Seq(
     Resolver.bintrayRepo("spark-jobserver", "spark-jobserver"),
-    Resolver.bintrayRepo("azavea", "geotrellis")
+    Resolver.bintrayRepo("azavea", "geotrellis"),
+    "Scalaz Bintray Repo" at "https://dl.bintray.com/scalaz/releases",
+    "OpenGeo" at "http://repo.boundlessgeo.com/main"
   )

   val defaultAssemblySettings =

After making that change, it worked famously.

The first addition is for the scalaz, the second is for the gdal stuff.

jamesmcclain commented 9 years ago

Those dependencies might already be present within the docker image.

hectcastro commented 9 years ago

Good idea to test with an empty Ivy cache. I was able to reproduce your issue, and then resolve it with https://github.com/WikiWatershed/mmw-geoprocessing/commit/0f629ecf799d5e0d349323dd86cd0c7d537d27f3.

jamesmcclain commented 9 years ago

:+1: It works great.