From c253c69d8f9b79d1073be919c06ae7cccace8528 Mon Sep 17 00:00:00 2001
From: Timothy Hodson <34148978+thodson-usgs@users.noreply.github.com>
Date: Tue, 21 Nov 2023 16:43:42 -0600
Subject: [PATCH] Open without fsspec

---
 recipes/us-ssebop/recipe.py | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/recipes/us-ssebop/recipe.py b/recipes/us-ssebop/recipe.py
index 67fb5feb71..4598df3274 100644
--- a/recipes/us-ssebop/recipe.py
+++ b/recipes/us-ssebop/recipe.py
@@ -4,12 +4,14 @@
 import pandas as pd
 
 from pangeo_forge_recipes.patterns import ConcatDim, FilePattern
-from pangeo_forge_recipes.transforms import Indexed, OpenURLWithFSSpec, StoreToZarr, T
+from pangeo_forge_recipes.transforms import Indexed, StoreToZarr, T
 
 input_url_pattern = (
+    'zip+'
     'https://edcintl.cr.usgs.gov/downloads/sciweb1/shared/uswem/web/'
     'conus/eta/modis_eta/daily/downloads/'
     'det{yyyyjjj}.modisSSEBopETactual.zip'
+    '!/det{yyyyjjj}.modisSSEBopETactual.tif'
 )
 
 start = date(2001, 1, 1)
@@ -32,14 +34,15 @@ def _preproc(item: Indexed[T]) -> Indexed[T]:
         import numpy as np
         import rioxarray
 
-        index, f = item
+        index, url = item
         time_dim = index.find_concat_dim('time')
         time_index = index[time_dim].value
         time = dates[time_index]
 
-        da = rioxarray.open_rasterio(f.open()).drop('band')
+        da = rioxarray.open_rasterio(url).drop('band')
         da = da.rename({'x': 'lon', 'y': 'lat'})
         ds = da.to_dataset(name='aet')
+        ds = ds['aet'].where(ds['aet'] != 9999)
         ds = ds.expand_dims(time=np.array([time]))
 
         return index, ds
@@ -50,7 +53,6 @@ def expand(self, pcoll: beam.PCollection) -> beam.PCollection:
 
 recipe = (
     beam.Create(pattern.items())
-    | OpenURLWithFSSpec(max_concurrency=10, open_kwargs={'compression': 'zip'})
     | Preprocess()
     | StoreToZarr(
         store_name='us-ssebop.zarr',