Merge pull request #542 from jbouffard/feature/temporal-resolution

Better Temporal Resolution Control When Writing Layer
locationtech-labs · Nov 16, 2017 · dcdab7e · dcdab7e
2 parents 151b6c3 + 0984003
commit dcdab7e
Show file tree

Hide file tree

Showing 3 changed files with 36 additions and 10 deletions.
diff --git a/...spark-backend/geotrellis/src/main/scala/geopyspark/geotrellis/io/LayerWriterWrapper.scala b/...spark-backend/geotrellis/src/main/scala/geopyspark/geotrellis/io/LayerWriterWrapper.scala
@@ -43,17 +43,27 @@ class LayerWriterWrapper(attributeStore: AttributeStore, uri: String) {
 
   private def getTemporalIndexMethod(
     timeString: String,
+    timeResolution: String,
     indexStrategy: String
   ) =
-    (indexStrategy, timeString) match {
-      case ("zorder", "millis") => ZCurveKeyIndexMethod.byMilliseconds(1)
-      case ("zorder", "seconds") => ZCurveKeyIndexMethod.bySecond
-      case ("zorder", "minutes") => ZCurveKeyIndexMethod.byMinute
-      case ("zorder", "hours") => ZCurveKeyIndexMethod.byHour
-      case ("zorder", "days") => ZCurveKeyIndexMethod.byDay
-      case ("zorder", "months") => ZCurveKeyIndexMethod.byMonth
-      case ("zorder", "years") => ZCurveKeyIndexMethod.byYear
-      case ("hilbert", _) => {
+    (indexStrategy, timeString, timeResolution) match {
+      case ("zorder", "millis", null) => ZCurveKeyIndexMethod.byMilliseconds(1)
+      case ("zorder", "millis", r) => ZCurveKeyIndexMethod.byMilliseconds(r.toLong)
+      case ("zorder", "seconds", null) => ZCurveKeyIndexMethod.bySecond
+      case ("zorder", "seconds", r) => ZCurveKeyIndexMethod.bySeconds(r.toInt)
+      case ("zorder", "minutes", null) => ZCurveKeyIndexMethod.byMinute
+      case ("zorder", "minutes", r) => ZCurveKeyIndexMethod.byMinutes(r.toInt)
+      case ("zorder", "hours", null) => ZCurveKeyIndexMethod.byHour
+      case ("zorder", "hours", r) => ZCurveKeyIndexMethod.byHours(r.toInt)
+      case ("zorder", "days", null) => ZCurveKeyIndexMethod.byDay
+      case ("zorder", "days", r) => ZCurveKeyIndexMethod.byDays(r.toInt)
+      case ("zorder", "weeks", null) => ZCurveKeyIndexMethod.byDays(7)
+      case ("zorder", "weeks", r) => ZCurveKeyIndexMethod.byDays(r.toInt * 7)
+      case ("zorder", "months", null) => ZCurveKeyIndexMethod.byMonth
+      case ("zorder", "months", r) => ZCurveKeyIndexMethod.byMonths(r.toInt)
+      case ("zorder", "years", null) => ZCurveKeyIndexMethod.byYear
+      case ("zorder", "years", r) => ZCurveKeyIndexMethod.byYears(r.toInt)
+      case ("hilbert", _, _) => {
         timeString.split(",") match {
           case Array(minDate, maxDate, resolution) =>
             HilbertKeyIndexMethod(ZonedDateTime.parse(minDate),
@@ -85,14 +95,15 @@ class LayerWriterWrapper(attributeStore: AttributeStore, uri: String) {
     layerName: String,
     temporalRDD: TiledRasterLayer[SpaceTimeKey],
     timeString: String,
+    timeResolution: String,
     indexStrategy: String
   ): Unit = {
     val id =
       temporalRDD.zoomLevel match {
         case Some(zoom) => LayerId(layerName, zoom)
         case None => LayerId(layerName, 0)
       }
-    val indexMethod = getTemporalIndexMethod(timeString, indexStrategy)
+    val indexMethod = getTemporalIndexMethod(timeString, timeResolution, indexStrategy)
     layerWriter.write(id, temporalRDD.rdd, indexMethod)
   }
 }
diff --git a/geopyspark/geotrellis/catalog.py b/geopyspark/geotrellis/catalog.py
@@ -231,6 +231,7 @@ def write(uri,
           tiled_raster_layer,
           index_strategy=IndexingMethod.ZORDER,
           time_unit=None,
+          time_resolution=None,
           store=None):
     """Writes a tile layer to a specified destination.
 
@@ -250,6 +251,15 @@ def write(uri,
             each index. Meaning, what time intervals are used to seperate each record. While this is
             set to ``None`` as default, it must be set if saving spatial-temporal data.
             Depending on the indexing method chosen, different time units are used.
+        time_resolution (str or int, optional): Determines how data for each ``time_unit`` should be
+            grouped together. By default, no grouping will occur.
+
+            As an example, having a ``time_unit`` of ``WEEKS`` and a ``time_resolution`` of 5 will
+            cause the data to be grouped and stored together in units of 5 weeks. If however
+            ``time_resolution`` is not specified, then the data will be grouped and stored in units
+            of single weeks.
+
+            This value can either be an ``int`` or a string representation of an ``int``.
         store (str or :class:`~geopyspark.geotrellis.catalog.AttributeStore`, optional):
             ``AttributeStore`` instance or URI for layer metadata lookup.
     """
@@ -274,9 +284,13 @@ def write(uri,
                             IndexingMethod(index_strategy).value)
 
     elif tiled_raster_layer.layer_type == LayerType.SPACETIME:
+        if time_resolution:
+            time_resolution = str(time_resolution)
+
         writer.writeTemporal(layer_name,
                              tiled_raster_layer.srdd,
                              TimeUnit(time_unit).value,
+                             time_resolution,
                              IndexingMethod(index_strategy).value)
     else:
         raise ValueError("Cannot write {} layer".format(tiled_raster_layer.layer_type))

diff --git a/geopyspark/geotrellis/constants.py b/geopyspark/geotrellis/constants.py
@@ -116,6 +116,7 @@ class TimeUnit(Enum):
     MINUTES = 'minutes'
     HOURS = 'hours'
     DAYS = 'days'
+    WEEKS = 'weeks'
     MONTHS = 'months'
     YEARS = 'years'