add crs to geodataframes to address #43

emptyframe as geodataframe with crs
save crs as attribute to netcdf and hdf5 files
add file load and save widgets

Create postBuild

match requirements.txt with environment.yml

move dockerfile to github workflow

move polygon routines from ipysliderule to io

use lat and lon keys for io

observe load/save labels to set filename

add sliderule commit and version to attributes

Author: tsutterley

Dockerfile renamed to .github/workflows/Dockerfile

@@ -51,5 +51,6 @@ jobs:
         # Build a docker container and
         # push it to ECR so that it can
         # be deployed to ECS.
-        docker build -t $ECR_REGISTRY/$ECR_REPOSITORY:$IMAGE_TAG .
+        docker build -t $ECR_REGISTRY/$ECR_REPOSITORY:$IMAGE_TAG \
+          --file $GITHUB_ACTION_PATH/Dockerfile
         docker push $ECR_REGISTRY/$ECR_REPOSITORY:$IMAGE_TAG

.github/workflows/aws-container.yml

@@ -9,6 +9,7 @@ dependencies:
   - ipykernel
   - ipywidgets
   - ipyleaflet
+  - tk
   - matplotlib
   - numpy
   - pandas

environment.yml

@@ -26,7 +26,6 @@
    "execution_count": null,
    "source": [
     "from sliderule import icesat2, ipysliderule, io\n",
-    "import time\n",
     "import logging\n",
     "import numpy as np\n",
     "import matplotlib.pyplot as plt\n",
@@ -107,8 +106,6 @@
     "# sliderule asset and data release\n",
     "asset = SRwidgets.asset.value\n",
     "release = SRwidgets.release.value\n",
-    "# get sliderule submission time\n",
-    "submission_time = time.strftime('%Y%m%d%H%M%S',time.gmtime())\n",
     "\n",
     "# create an empty dataframe\n",
     "gdf = icesat2.__emptyframe()\n",
@@ -163,7 +160,7 @@
     "# output map of resource location\n",
     "f1, ax1 = plt.subplots(num=1, nrows=1, ncols=1, figsize=(10,6),\n",
     "    subplot_kw=dict(projection=cartopy.crs.PlateCarree()))\n",
-    "ax1.plot(gdf.geometry.values.x,gdf.geometry.values.y,'r.',\n",
+    "ax1.plot(gdf.geometry.x,gdf.geometry.y,'r.',\n",
     "    ms=1,transform=cartopy.crs.Geodetic())\n",
     "ax1.set_extent((-180,180,-90,90),crs=cartopy.crs.PlateCarree())\n",
     "# add coastlines with filled land and lakes\n",
@@ -182,12 +179,12 @@
     "f2, ax2 = plt.subplots(num=2, nrows=1, ncols=1, figsize=(10,6),\n",
     "    subplot_kw=dict(projection=cartopy.crs.PlateCarree()))\n",
     "# create scatter plot of elevations\n",
-    "sc = ax2.scatter(gdf.geometry.values.x,gdf.geometry.values.y,\n",
+    "sc = ax2.scatter(gdf.geometry.x,gdf.geometry.y,\n",
     "    c=gdf.h_mean,s=1,edgecolor='none',cmap=plt.cm.viridis,\n",
     "    transform=cartopy.crs.PlateCarree())\n",
     "# extract latitude and longitude of polygon\n",
     "for poly in m.regions:\n",
-    "    lon,lat = ipysliderule.coordinates(poly)\n",
+    "    lon,lat = io.from_region(poly)\n",
     "    ax2.plot(lon, lat, 'r', lw=1.5, transform=cartopy.crs.PlateCarree())\n",
     "# add coastlines with filled land and lakes\n",
     "ax2.add_feature(cartopy.feature.LAND, zorder=0, edgecolor='black')\n",
@@ -218,36 +215,66 @@
   {
    "cell_type": "markdown",
    "source": [
-    "#### Save GeoDataFrame to HDF5\n",
-    "Outputs as a [pytables](https://www.pytables.org/) HDF5 which is easily read back as a Geopandas GeoDataFrame"
+    "#### Save GeoDataFrame to output tile\n",
+    "- [pytables HDF5](https://www.pytables.org/): easily read back as a Geopandas GeoDataFrame\n",
+    "- [netCDF](https://www.unidata.ucar.edu/software/netcdf): interoperable with other programs"
    ],
    "metadata": {}
   },
   {
    "cell_type": "code",
    "execution_count": null,
    "source": [
-    "args = (submission_time,release)\n",
-    "io.to_file(gdf,\"ATL06-SR_{0}_{1}.h5\".format(*args),\n",
-    "    format='hdf', parameters=parms, regions=m.regions, verbose=True)"
+    "display(SRwidgets.filesaver)"
+   ],
+   "outputs": [],
+   "metadata": {}
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "source": [
+    "# append sliderule api version to attributes\n",
+    "version = icesat2.get_version()\n",
+    "parms['version'] = version['icesat2']['version']\n",
+    "parms['commit'] = version['icesat2']['commit']\n",
+    "# save to file in format (HDF5 or netCDF)\n",
+    "io.to_file(gdf, SRwidgets.file,\n",
+    "    format=SRwidgets.format,\n",
+    "    driver='pytables',\n",
+    "    parameters=parms,\n",
+    "    regions=m.regions,\n",
+    "    verbose=True)"
    ],
    "outputs": [],
    "metadata": {}
   },
   {
    "cell_type": "markdown",
    "source": [
-    "#### Read GeoDataFrame from HDF5\n",
-    "Read the HDF5 back as a Geopandas GeoDataFrame for later use"
+    "#### Read GeoDataFrame from input file\n",
+    "- [pytables HDF5](https://www.pytables.org/)\n",
+    "- [netCDF](https://www.unidata.ucar.edu/software/netcdf)"
    ],
    "metadata": {}
   },
   {
    "cell_type": "code",
    "execution_count": null,
    "source": [
-    "args = (submission_time,release)\n",
-    "gdf = io.from_file(\"ATL06-SR_{0}_{1}.h5\".format(*args),format='hdf')"
+    "display(SRwidgets.fileloader)"
+   ],
+   "outputs": [],
+   "metadata": {}
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "source": [
+    "# read from file in format (HDF5 or netCDF)\n",
+    "gdf = io.from_file(SRwidgets.file,\n",
+    "    format=SRwidgets.format,\n",
+    "    driver='pytables')"
    ],
    "outputs": [],
    "metadata": {}

examples/api_widgets_demo.ipynb

@@ -0,0 +1,2 @@
+jupyter labextension install --clean @jupyter-widgets/jupyterlab-manager \
+   jupyter-matplotlib jupyter-leaflet

postBuild

@@ -1,4 +1,15 @@
 requests
 setuptools_scm
 numpy
+ipykernel
+ipywidgets
+ipyleaflet
+tk
+matplotlib
+cartopy
+pandas
 geopandas
+h5py
+pyproj
+scipy
+shapely

requirements.txt

@@ -278,17 +278,22 @@ def __query_servers(max_workers):
 #
 #  __emptyframe
 #
-def __emptyframe():
-    return geopandas.pd.DataFrame()
+def __emptyframe(**kwargs):
+    # set default keyword arguments
+    kwargs['crs'] = "EPSG:4326"
+    return geopandas.GeoDataFrame(crs=kwargs['crs'])
 
 #
 #  __todataframe
 #
-def __todataframe(columns, delta_time_key="delta_time", lon_key="lon", lat_key="lat", index_key="time"):
+def __todataframe(columns, delta_time_key="delta_time", lon_key="lon", lat_key="lat", **kwargs):
+    # set default keyword arguments
+    kwargs['index_key'] = "time"
+    kwargs['crs'] = "EPSG:4326"
 
     # Check Empty Columns
     if len(columns) <= 0:
-        return __emptyframe()
+        return __emptyframe(**kwargs)
 
     # Generate Time Column
     delta_time = (columns[delta_time_key]*1e9).astype('timedelta64[ns]')
@@ -304,11 +309,11 @@ def __todataframe(columns, delta_time_key="delta_time", lon_key="lon", lat_key="
     df = geopandas.pd.DataFrame(columns)
 
     # Build GeoDataFrame (default geometry is crs="EPSG:4326")
-    gdf = geopandas.GeoDataFrame(df, geometry=geometry)
+    gdf = geopandas.GeoDataFrame(df, geometry=geometry, crs=kwargs['crs'])
 
     # Set index (default is Timestamp), can add `verify_integrity=True` to check for duplicates
     # Can do this during DataFrame creation, but this allows input argument for desired column
-    gdf.set_index(index_key, inplace=True)
+    gdf.set_index(kwargs['index_key'], inplace=True)
 
     # Sort values for reproducible output despite async processing
     gdf.sort_index(inplace=True)