Skip to content

Commit

Permalink
allixender#11 handle z7string for cells_for_geo_points
Browse files Browse the repository at this point in the history 
    - change output file name of output_conf
      prefix with {output_address_type}
    - supply dtype to pd.read_csv if output_address_type
      contains "STRING" (otherwise it will auto convert to int64)
    - change the cells_id column to "name" in the return df.
      Align with other's functions.
    - update parameters when calling grid_cell_polygons_from_cellids
      with
       a. input_address_type=output_address_type
       b. output_address_type=output_address_type
  • Loading branch information
@tik65536
tik65536 committed Dec 23, 2024
1 parent d8e399d commit c668342
Showing 1 changed file with 26 additions and 25 deletions.
51 changes: 26 additions & 25 deletions dggrid4py/dggrid_runner.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -656,7 +656,7 @@ def dgapi_grid_gen(self, dggs, subset_conf, output_conf):
metafile.append("output_last_seqnum " + str(subset_conf['output_last_seqnum']))
else:
raise ValueError('something is not correct in subset_conf')

if 'input_address_type' in subset_conf.keys() and subset_conf.get('input_address_type', 'NOPE') in input_address_types:
metafile.append("input_address_type " + subset_conf['input_address_type'])

Expand Down Expand Up @@ -951,7 +951,7 @@ def grid_stats_table(self, dggs_type, resolution, mixed_aperture_level=None):
df.rename(columns={0: 'Resolution', 1: "Cells", 2:"Area (km^2)", 3: "CLS (km)"}, inplace=True)
df['Resolution'] = df['Resolution'].astype(int)
df['Cells'] = df['Cells'].astype(np.int64)

return df


Expand Down Expand Up @@ -997,7 +997,7 @@ def grid_cell_polygons_for_extent(self, dggs_type, resolution, mixed_aperture_le
'cell_output_file_name': str( (Path(tmp_dir) / f"temp_{dggs_type}_{resolution}_out_{tmp_id}").resolve())
})
output_conf.pop('cell_output_gdal_format', None)

if not output_address_type is None and output_address_type in output_address_types:
output_conf.update({'output_address_type': output_address_type})
else:
Expand Down Expand Up @@ -1065,7 +1065,7 @@ def grid_cell_centroids_for_extent(self, dggs_type, resolution, mixed_aperture_l
'point_output_file_name': str( (Path(tmp_dir) / f"temp_{dggs_type}_{resolution}_out_{tmp_id}").resolve())
})
output_conf.pop('point_output_gdal_format', None)

if not output_address_type is None and output_address_type in output_address_types:
output_conf.update({'output_address_type': output_address_type})
else:
Expand Down Expand Up @@ -1112,7 +1112,7 @@ def grid_cell_polygons_from_cellids(self, cell_id_list, dggs_type, resolution, m
'clip_subset_type': 'SEQNUMS',
'clip_region_files': str( (Path(tmp_dir) / f"temp_clip_{tmp_id}.txt").resolve()),
})

# TODO, for Z3, Z7, ZORDER can potentially also be COARSE_CELLS / aka parent cells?
# clip_subset_type should INPUT_ADDRESS_TYPE for the equivalent of SEQNUM (tp use input_address_type Z3 ...), or COARSE_CELLS as an actual paent cell type clip (also for Z3 ..)
if (
Expand All @@ -1125,7 +1125,7 @@ def grid_cell_polygons_from_cellids(self, cell_id_list, dggs_type, resolution, m
'input_address_type': input_address_type
}
)

if not clip_subset_type is None and clip_subset_type in ['COARSE_CELLS']:
subset_conf.update(
{
Expand Down Expand Up @@ -1180,7 +1180,7 @@ def grid_cell_polygons_from_cellids(self, cell_id_list, dggs_type, resolution, m
gdf[name_col] = gdf[name_col].astype(np.int64)
# gdf = gdf.join( seq_df, how='inner', left_on=name_col, right_on=input_address_type)
# gdf = gdf.loc[gdf['cell_exists']].drop(columns=['cell_exists'])

if self.debug is False:
try:
os.remove( str( Path(tmp_dir) / f"temp_{dggs_type}_{resolution}_out_{tmp_id}.{self.tmp_geo_out['ext']}") )
Expand Down Expand Up @@ -1218,7 +1218,7 @@ def grid_cell_centroids_from_cellids(self, cell_id_list, dggs_type, resolution,
'clip_subset_type': 'SEQNUMS',
'clip_region_files': str( (Path(tmp_dir) / f"temp_clip_{tmp_id}.txt").resolve()),
})

# TODO, for Z3, Z7, ZORDER can potentially also be COARSE_CELLS / aka parent cells?
# clip_subset_type should INPUT_ADDRESS_TYPE for the equivalent of SEQNUM (tp use input_address_type Z3 ...), or COARSE_CELLS as an actual paent cell type clip (also for Z3 ..)
if (
Expand All @@ -1231,7 +1231,7 @@ def grid_cell_centroids_from_cellids(self, cell_id_list, dggs_type, resolution,
'input_address_type': input_address_type
}
)

if not clip_subset_type is None and clip_subset_type in ['COARSE_CELLS']:
subset_conf.update(
{
Expand Down Expand Up @@ -1386,7 +1386,7 @@ def cells_for_geo_points(self, geodf_points_wgs84, cell_ids_only, dggs_type, res
}

output_conf = {
'output_file_name': str( (Path(tmp_dir) / f"seqnums_{tmp_id}.txt").resolve()),
'output_file_name': str( (Path(tmp_dir) / f"{output_address_type}_{tmp_id}.txt").resolve()),
'output_address_type': 'SEQNUM',
'output_delimiter': "\",\""
}
Expand All @@ -1400,8 +1400,8 @@ def cells_for_geo_points(self, geodf_points_wgs84, cell_ids_only, dggs_type, res
dggs_ops = self.dgapi_grid_transform(dggs, subset_conf, output_conf)
if self.debug is True:
print(dggs_ops)

df = pd.read_csv( dggs_ops['output_conf']['output_file_name'] , header=None)
datatype = {0: str} if ('STRING' in output_address_type) else {}
df = pd.read_csv( dggs_ops['output_conf']['output_file_name'] , header=None, dtype=datatype)
df = df.dropna()
cell_id_list = df[0].values

Expand All @@ -1413,14 +1413,15 @@ def cells_for_geo_points(self, geodf_points_wgs84, cell_ids_only, dggs_type, res
pass

if cell_ids_only == True:
geodf_points_wgs84['seqnums'] = cell_id_list
geodf_points_wgs84['name'] = cell_id_list
return geodf_points_wgs84
else:
# grid_gen from seqnums
gdf = self.grid_cell_polygons_from_cellids(cell_id_list=cell_id_list,
dggs_type=dggs_type,
resolution=resolution,
mixed_aperture_level=mixed_aperture_level)
mixed_aperture_level=mixed_aperture_level, input_address_type=output_address_type,
output_address_type=output_address_types)
try:
for col in cols_ordered:
gdf[col] = geodf_points_wgs84[col].values
Expand All @@ -1429,25 +1430,25 @@ def cells_for_geo_points(self, geodf_points_wgs84, cell_ids_only, dggs_type, res

if split_dateline == True:
return self.post_process_split_dateline(gdf)

return gdf


def address_transform(self, cell_id_list, dggs_type, resolution, mixed_aperture_level=None, input_address_type='SEQNUM', output_address_type='SEQNUM'):
"""
generates the DGGS for the input cell_ids and returns all the transformed cell_ids
cell_id_list is a list/numpy array, takes this list as seqnums ids (potentially also Z3, Z7, or ZORDER .. TODO)
cell_id_list is a list/numpy array, takes this list as seqnums ids (potentially also Z3, Z7, or ZORDER .. TODO)
"""
tmp_id = uuid.uuid4()
tmp_dir = self.working_dir
dggs = dgselect(dggs_type = dggs_type, res= resolution, mixed_aperture_level=mixed_aperture_level)

if cell_id_list is None or len(cell_id_list) <= 0:
raise ValueError("Expecting cell_id_list to transform.")

if not input_address_type in input_address_types:
raise ValueError(f"unknown input_address_type: {input_address_type}")

if not output_address_type in output_address_types:
raise ValueError(f"unknown output_address_type: {output_address_type}")

Expand All @@ -1466,7 +1467,7 @@ def address_transform(self, cell_id_list, dggs_type, resolution, mixed_aperture_
'output_delimiter': "\" \""
}


dggs_ops = self.dgapi_grid_transform(dggs, subset_conf, output_conf)
if self.debug is True:
print(dggs_ops)
Expand All @@ -1481,28 +1482,28 @@ def address_transform(self, cell_id_list, dggs_type, resolution, mixed_aperture_
os.remove( str( Path(tmp_dir) / f"temp_out_{output_address_type}_{tmp_id}.txt") )
except Exception:
pass

return seq_df


def guess_zstr_resolution(self, cell_id_list, dggs_type, input_address_type='Z7_STRING'):
if cell_id_list is None or len(cell_id_list) <= 0:
raise ValueError("Expecting cell_id_list to transform.")

if not input_address_type in ['Z3_STRING', 'Z7_STRING']:
raise ValueError(f"this will likely not work for this input_address_type: {input_address_type} | only Z3 and Z7 verified")

if not dggs_type in ['ISEA3H', 'ISEA7H', 'IGEO7']:
raise ValueError(f"this will likely not work for this dggs_type: {dggs_type} | only Z3 and Z7 compatible")

df = pd.DataFrame({ input_address_type: cell_id_list})

# df = self.address_transform(cell_id_list, dggs_type, input_address_type=input_address_type,
# output_address_type=input_address_type + '_STRING')
df['resolution'] = df[input_address_type].apply(lambda s: len(s) - 2)

return df




Expand Down

0 comments on commit c668342

Please sign in to comment.