Merge pull request #88 from andrewdelman/ecco_access_fsspec

andrewdelman · web-flow · commit 789a9a4b35ff · 2024-10-08T00:34:34.000-07:00
edits to function headers and arguments order
diff --git a/ecco_access/ecco_access.py b/ecco_access/ecco_access.py
@@ -124,14 +124,14 @@ def ecco_podaac_access(query,version='v4r4',grid=None,time_res='all',\
 
     Returns
     -------
-    download_files: dict, with keys: ShortNames and values:
-                    URLs (if in 'query' mode), or paths of files that can be 
-                    passed directly to xarray (open_dataset or open_mfdataset).
-                    Values are of type str if query finds only one granule/file
-                    for that ShortName; of type list if query finds 
-                    multiple granules in the same dataset; 
-                    or of type fsspec.mapping.FSMap if mode = 's3_open_fsspec'.
-                    Only returned if return_granules=True (default).
+    granule_files: dict with ShortNames as keys; values are URLs or S3 paths
+                   (if in 'query' mode), or paths of files that can be 
+                   passed directly to xarray (open_dataset or open_mfdataset).
+                   Values are of type str if query finds only one granule/file
+                   for that ShortName; of type list if query finds 
+                   multiple granules in the same dataset; 
+                   or of type fsspec.mapping.FSMap if mode = 's3_open_fsspec'.
+                   Only returned if return_granules=True (default).
     
     """
     
@@ -272,6 +272,7 @@ def shortnames_find(query_list,grid,time_res):
 ###================================================================================================================
 
 
+
 def ecco_podaac_to_xrdataset(query,version='v4r4',grid=None,time_res='all',\
                              StartDate=None,EndDate=None,snapshot_interval=None,\
                              mode='download_ifspace',download_root_dir=None,**kwargs):
@@ -325,10 +326,6 @@ def ecco_podaac_to_xrdataset(query,version='v4r4',grid=None,time_res='all',\
                        and 'monthly' otherwise.
     
     mode: str, one of the following:
-          'ls' or 'query': Query dataset ShortNames and variable names/
-                           descriptions only; no downloads.
-          's3_ls' or 's3_query': Query dataset ShortNames and variable names/
-                                 descriptions only; return paths on S3.
           'download': Download datasets using NASA Earthdata URLs
           'download_ifspace': Check storage availability before downloading.
                               Download only if storage footprint of downloads 
@@ -375,16 +372,11 @@ def ecco_podaac_to_xrdataset(query,version='v4r4',grid=None,time_res='all',\
     force_redownload: bool, if True, existing files will be redownloaded and replaced;
                             if False (default), existing files will not be replaced.
 
-    return_granules: bool, if True (default), str or list of queried or 
-                           downloaded granules/files (including ones that 
-                           were already on disk and not replaced) is returned.
-                           if False, the function returns nothing.
-
     Returns
     -------
     ds_out: xarray Dataset or dict of xarray Datasets (with ShortNames as keys), 
             containing all of the accessed datasets.
-            Does not work with the query modes: 'ls','query','s3_ls','s3_query'.
+            This function does not work with the query modes: 'ls','query','s3_ls','s3_query'.
     """
     
     pass
diff --git a/ecco_access/ecco_download.py b/ecco_access/ecco_download.py
@@ -383,8 +383,8 @@ def ecco_podaac_download(ShortName,StartDate,EndDate,snapshot_interval='monthly'
 ###================================================================================================================
 
 
-def ecco_podaac_download_diskaware(ShortNames,StartDate,EndDate,max_avail_frac=0.5,snapshot_interval=None,\
-                                   download_root_dir=None,n_workers=6,force_redownload=False):
+def ecco_podaac_download_diskaware(ShortNames,StartDate,EndDate,snapshot_interval=None,\
+                                   download_root_dir=None,max_avail_frac=0.5,n_workers=6,force_redownload=False):
     
     """
     
@@ -405,11 +405,6 @@ def ecco_podaac_download_diskaware(ShortNames,StartDate,EndDate,max_avail_frac=0
                        ECCOv4r4 date range is '1992-01-01' to '2017-12-31'.
                        For 'SNAPSHOT' datasets, an additional day is added to EndDate to enable closed budgets
                        within the specified date range.
-
-    max_avail_frac: float, maximum fraction of remaining available disk space to use in storing current ECCO datasets.
-                    If storing the datasets exceeds this fraction, an error is returned.
-                    Valid range is [0,0.9]. If number provided is outside this range, it is replaced by the closer 
-                    endpoint of the range.
     
     snapshot_interval: ('monthly', 'daily', or None), if snapshot datasets are included in ShortNames, 
                        this determines whether snapshots are included for only the beginning/end of each month 
@@ -420,6 +415,11 @@ def ecco_podaac_download_diskaware(ShortNames,StartDate,EndDate,max_avail_frac=0
     download_root_dir: str, defines parent directory to download files to.
                        Files will be downloaded to directory download_root_dir/ShortName/.
                        If not specified, parent directory defaults to '~/Downloads/ECCO_V4r4_PODAAC/'.
+
+    max_avail_frac: float, maximum fraction of remaining available disk space to use in storing current ECCO datasets.
+                    If storing the datasets exceeds this fraction, an error is returned.
+                    Valid range is [0,0.9]. If number provided is outside this range, it is replaced by the closer 
+                    endpoint of the range.
     
     n_workers: int, number of workers to use in concurrent downloads. Benefits typically taper off above 5-6.
                Applies only if files are downloaded.
diff --git a/ecco_access/ecco_s3_retrieve.py b/ecco_access/ecco_s3_retrieve.py
@@ -527,8 +527,8 @@ def ecco_podaac_s3_get(ShortName,StartDate,EndDate,snapshot_interval='monthly',d
 ###================================================================================================================
 
 
-def ecco_podaac_s3_get_diskaware(ShortNames,StartDate,EndDate,max_avail_frac=0.5,snapshot_interval=None,\
-                                 download_root_dir=None,n_workers=6,force_redownload=False):
+def ecco_podaac_s3_get_diskaware(ShortNames,StartDate,EndDate,snapshot_interval=None,\
+                                 download_root_dir=None,max_avail_frac=0.5,n_workers=6,force_redownload=False):
     
     """
     
@@ -551,12 +551,7 @@ def ecco_podaac_s3_get_diskaware(ShortNames,StartDate,EndDate,max_avail_frac=0.5
                        ECCOv4r4 date range is '1992-01-01' to '2017-12-31'.
                        For 'SNAPSHOT' datasets, an additional day is added to EndDate to enable closed budgets
                        within the specified date range.
-
-    max_avail_frac: float, maximum fraction of remaining available disk space to use in storing current ECCO datasets.
-                    This determines whether the dataset files are stored on the current instance, or opened on S3.
-                    Valid range is [0,0.9]. If number provided is outside this range, it is replaced by the closer 
-                    endpoint of the range.
-
+    
     snapshot_interval: ('monthly', 'daily', or None), if snapshot datasets are included in ShortNames, 
                        this determines whether snapshots are included for only the beginning/end of each month 
                        ('monthly'), or for every day ('daily').
@@ -567,6 +562,11 @@ def ecco_podaac_s3_get_diskaware(ShortNames,StartDate,EndDate,max_avail_frac=0.5
                        Files will be downloaded to directory download_root_dir/ShortName/.
                        If not specified, parent directory defaults to '~/Downloads/ECCO_V4r4_PODAAC/'.
     
+    max_avail_frac: float, maximum fraction of remaining available disk space to use in storing current ECCO datasets.
+                    This determines whether the dataset files are stored on the current instance, or opened on S3.
+                    Valid range is [0,0.9]. If number provided is outside this range, it is replaced by the closer 
+                    endpoint of the range.
+    
     n_workers: int, number of workers to use in concurrent downloads. Benefits typically taper off above 5-6.
                Applies only if files are downloaded.
     
