Update README

geotrellis · jpolchlo · Mar 22, 2019 · Mar 1, 2019 · Mar 1, 2019 · Mar 1, 2019
commit 72470aece63f00c76ac5b255b8ad964ee8da9d45
diff --git a/README.md b/README.md
@@ -5,60 +5,25 @@ VectorPipe
 [![Bintray](https://img.shields.io/bintray/v/azavea/maven/vectorpipe.svg)](https://bintray.com/azavea/maven/vectorpipe)
 [![Codacy Badge](https://api.codacy.com/project/badge/Grade/447170921bc94b3fb494bb2b965c2235)](https://www.codacy.com/app/fosskers/vectorpipe?utm_source=github.com&amp;utm_medium=referral&amp;utm_content=geotrellis/vectorpipe&amp;utm_campaign=Badge_Grade)
 
-A pipeline for mass conversion of Vector data (OpenStreetMap, etc.) into
-Mapbox VectorTiles. Powered by [GeoTrellis](http://geotrellis.io) and
-[Apache Spark](http://spark.apache.org/).
-
-Overview
---------
-
-![](docs/pipeline.png)
-
-There are four main stages here which represent the four main shapes of
-data in the pipeline:
-
-1. Unprocessed Vector (geometric) Data
-2. Clipped GeoTrellis `Feature`s organized into a Grid on the earth
-3. VectorTiles organized into a Grid on the earth
-4. Fully processed VectorTiles, output to some target
-
-Of these, Stage 4 is left to the user to leverage GeoTrellis directly in
-their own application. Luckily the `RDD[(SpatialKey, VectorTile)] => Unit`
-operation only requires about 5 lines of code. Stages 1 to 3 then are the
-primary concern of Vectorpipe.
-
-### Processing Raw Data
-
-For each data source that has first-class support, we expose a
-`vectorpipe.*` module with a matching name. Example: `vectorpipe.osm`. These
-modules expose all the types and functions necessary for transforming the
-raw data into the "Middle Ground" types.
-
-No first-class support for your favourite data source? Want to write it
-yourself, and maybe even keep it private? That's okay, just provide the
-function `YourData => RDD[Feature[G, D]]` and VectorPipe can handle the
-rest.
-
-### Clipping Features into a Grid
-
-GeoTrellis has a consistent `RDD[(K, V)]` pattern for handling grids of
-tiled data, where `K` is the grid index and `V` is the actual value type.
-Before `RDD[(SpatialKey, VectorTile)]` can be achieved, we need to convert
-our gridless `RDD[Feature[G, D]]` into such a grid, such that each Feature's
-`Geometry` is reasonably clipped to the size of an individual tile. Depending
-on which clipping function you choose (from the `vectorpipe.Clip` object, or
-even your own custom one) the shape of the clipped Geometry will vary. See
-our Scaladocs for more detail on the available options.
-
-### Collating Feature Groups into a VectorTile
-
-Once clipped and gridded by `VectorPipe.toGrid`, we have a `RDD[(SpatialKey,
-Iterable[Feature[G, D]])]` that represents all the Geometry fragments
-present at each tiled location on the earth. This is the perfect shape to
-turn into a `VectorTile`. To do so, we need to choose a *Collator* function,
-which determines what VectorTile Layer each `Feature` should be placed into,
-and how (if at all) its corresponding metadata (the `D`) should be
-processed.
-
-Want to write your own Collator? The `Collate.generically` function will be
-of interest to you.
+VectorPipe (VP) is a library for working with OpenStreetMap (OSM) vector
+data. Powered by [Geotrellis](http://geotrellis.io) and [Apache
+Spark](http://spark.apache.org/).
+
+OSM provides a wealth of data which has broad coverage and a deep history.
+This comes at the price of very large size which can make accessing the power
+of OSM difficult.  VectorPipe can help by making OSM processing in Apache
+Spark possible, leveraging large computing clusters to churn through the large
+volume of, say, an OSM full history file.
+
+For those cases where an application needs to process incoming changes, VP
+also provides streaming Spark `DataSource`s for changesets, OsmChange files,
+and Augmented diffs generated by Overpass.
+
+For ease of use, the output of VP imports is a Spark DataFrame containing
+columns of JTS `Geometry` objects, enabled by the user-defined types provided
+by [GeoMesa](https://github.com/locationtech/geomesa).  That package also
+provides functions for manipulating those geometries via Spark SQL directives.
+
+The final important contribution is a set of functions for exporting
+geometries to vector tiles.  This leans on the `geotrellis-vectortile`
+package.
diff --git a/project/Version.scala b/project/Version.scala
@@ -1,11 +1,18 @@
 object Version {
-  val vectorpipe = "0.3.1"
-  val geotrellis = "2.0.0-RC1"
-  val geomesa    = "2.0.1"
-  val spark      = "2.2.0"
-  val decline    = "0.4.0-RC1"
-  val monocle   = "1.5.0-cats-M2"
-  val cats       = "1.0.0-RC1"
-  val spire      = "0.13.0"
-  val scalaTest  = "3.0.3"
+  val scala = "2.11.11"
+  val osmesa = "0.1.0"
+  val geotrellis = "2.1.0"
+  val geomesa = "2.1.0"
+  val vectorpipe = "0.4.0"
+  val decline = "0.5.0"
+  val cats = "1.0.0"
+  val scalactic = "3.0.3"
+  val scalatest = "3.0.3"
+  val spark = "2.3.1"
+  val kryo = "4.0.0"
+  val snakeyaml = "1.8"
+  val circe = "0.9.0-M2"
+  val scalaLogging = "3.5.0"
+  val commonsIO = "2.5"
+  val osmosis = "0.46"
 }
diff --git a/project/build.properties b/project/build.properties
@@ -1 +1 @@
-sbt.version=1.0.4
+sbt.version=1.2.8