load module added

setup.py

@@ -17,6 +17,8 @@
     },
     keywords="glaciology ice lake ESA",
     classifiers=[
+        "Programming Language :: Python :: 3.8",
+        "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "Programming Language :: Python :: 3.11",
         "Programming Language :: Python :: 3.12",
@@ -30,6 +32,6 @@
     package_dir={"": "src"},
     packages=setuptools.find_packages(where="src"),
     package_data={"griml.test": ["*"]},
-    python_requires=">=3.10",
+    python_requires=">=3.8",
     install_requires=['geopandas', 'pandas', 'scipy', 'Shapely', 'rasterio'],
 )

src/griml/convert/convert.py

@@ -5,7 +5,19 @@
 import glob
 from pathlib import Path
 
-def convert(indir, outdir, proj, band_info, startdate, enddate, outfile=None):
+def convert(indir, proj, band_info, startdate, enddate, outdir=None):
+    '''Compile features from multiple processings into one geodataframe
+
+    Parameters
+    ----------
+    inlist : list
+        List of files or geopandas.dataframe.DataFrame objects to merge
+
+    Returns
+    -------
+    all_gdf : geopandas.dataframe.GeoDataFrame
+        Compiled goedataframe
+    '''
 
     # Iterate through files
     converted=[]
@@ -14,18 +26,17 @@ def convert(indir, outdir, proj, band_info, startdate, enddate, outfile=None):
         print('\n'+str(count) + '. Converting ' + str(Path(i).name))
 
         # Convert raster to vector
-        if outfile is not None:
+        if outdir is not None:
             outfile = str(Path(outdir).joinpath(Path(i).stem+'.shp'))
             g = raster_to_vector(str(i), proj, band_info, startdate, enddate, None)
             print('Saved to '+str(Path(outfile).name))
 
         else:
-            g = raster_to_vector(str(i), proj, band_info, startdate, enddate, outfile)
+            g = raster_to_vector(str(i), proj, band_info, startdate, enddate)
 
         converted.append(g)
         count=count+1
 
-    print('Finished')
     return (converted)
 
 if __name__ == "__main__": 

src/griml/filter/filter_vectors.py

@@ -2,38 +2,72 @@
 # -*- coding: utf-8 -*-
 
 from griml.filter import filter_margin, filter_area
+from griml.load import load
 import geopandas as gpd
 from pathlib import Path
-import glob
 
-def filter_vectors(indir, outdir, margin_buff):
+def filter_vectors(inlist, margin_file, outdir=None, min_area=0.05):
+    '''Filter vectors by area and margin proximity
+
+    Parameters
+    ----------
+    inlist : list
+        List of either file paths of GeoDataFrame objects to filter
+    margin_file : str, geopandas.GeoSeries
+        Bufferred margin to perform margin proximity filter
+    outdir : str, optional
+        Output directory to write files to
+    min_area: int, optional
+        Threshold area (sq km) to filter by
+        
+    Returns
+    -------
+    filtered : list
+        List of filtered GeoDataFrame objects
+
+    '''
+
+    # Load margin
+    margin_buff = load(margin_file)
+
+    # Iterate through input list
     count=1
-    for infile in list(glob.glob(indir)):
+    filtered=[]
+    for infile in inlist:
+
+        # Load and define name
+        if type(infile)==str:
+            print('\n'+str(count)+'/'+str(len(inlist)) +
+                  ': Filtering vectors in '+str(Path(infile).name))   
+            name = str(Path(infile).stem)+"_filtered.shp"
+
+        else:
+            print('\n'+str(count)+'/'+str(len(inlist))) 
+            name = 'lakes_' + str(count) + "_filtered.shp"
 
-        print('\n'+str(count)+'. Filtering vectors in '+str(Path(infile).stem))
-        vectors = gpd.read_file(infile)
+        vectors = load(infile)
 
-        vectors = filter_area(vectors)
+        # Perform filtering steps
+        vectors = filter_area(vectors, min_area)
         print(f'{vectors.shape[0]} features over 0.05 sq km')
 
         vectors = filter_margin(vectors, margin_buff)
         print(f'{vectors.shape[0]} features within 500 m of margin')    
 
+        # Retain and save if vectors are present after filtering
         if vectors.shape[0]>0:
             if outdir is not None:
-                vectors.to_file(outdir+str(Path(infile).stem)+"_filtered.shp")
+                vectors.to_file(Path(outdir).joinpath(name))
+            filtered.append(vectors)
         else:
         	print('No vectors present after filter. Moving to next file.')
         count=count+1
+
 
 
 if __name__ == "__main__": 
-    indir = "/home/pho/python_workspace/GrIML/other/iml_2017/vectors/*.shp"   
-
-    infile_margin = "/home/pho/python_workspace/GrIML/other/datasets/ice_margin/gimp_icemask_line_polstereo_simple_buffer.shp"
-    print('Preparing ice margin buffer...')
-    margin_buff = gpd.read_file(infile_margin)   
-    # margin_buff = margin.buffer(500)
-    # margin_buff = gpd.GeoDataFrame(geometry=margin_buff, crs=margin.crs)
+    infile1 = '../test/test_icemask.shp'
+    margin_buff = gpd.read_file(infile1)
 
-    filter_vectors(indir, margin_buff)
+    infile2 = ['../test/test_filter.shp']       
+    filter_vectors(infile2, margin_buff)

src/griml/load/__init__.py

@@ -0,0 +1 @@
+from griml.load.load import *

src/griml/load/load.py

@@ -0,0 +1,80 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import geopandas as gpd
+import pandas as pd
+import os, errno
+from pathlib import Path
+
+def load(i): 
+    '''Load vectors into appropriate format for processing
+
+    Parameters
+    ----------
+    i : str, geopandas.geodataframe.GeoDataFrame, pandas.core.series.Series
+        Input vectors (either from file or vector object)
+
+    Returns
+    -------
+    out : geopandas.geodataframe.GeoDataFrame
+        Output vector object
+    '''
+    if is_string(i):
+        if is_filepath(i):
+            out = load_vector_from_file(i)
+            return out
+
+        else:
+            ValueError('Expected valid vector filepath, but instead got '+i)
+
+    elif is_geo_object(i):
+        out = i
+        return out
+
+    else:  
+        TypeError('Expected str, geopandas.geodataframe.GeoDataFrame or pandas.core.series.Series object, but instead got '+str(type(i)))
+
+def load_vector_from_file(infile):
+    '''Load vector object from file'''
+    gdf = gpd.read_file(infile)
+    return gdf
+
+def is_string(n):
+    '''Check if input for loading is string'''
+    if type(n)==str:
+        return True
+    else:
+        return False
+
+def is_filepath(n):
+    '''Check if string is valid file'''
+    if os.path.isfile(n):
+        return True
+    else:
+        raise FileNotFoundError(
+            errno.ENOENT, os.strerror(errno.ENOENT), n)
+
+def is_dir(n):
+    '''Check if string is valid directory'''
+    if os.path.isdir(str(Path(n).parent)):
+        return True
+    else:
+        raise FileNotFoundError(
+            errno.ENOENT, os.strerror(errno.ENOENT), n)
+
+def is_geo_object(typ):
+    '''Check if object is valid vector object'''
+    if type(typ)==gpd.geodataframe.GeoDataFrame or type(typ)==pd.core.series.Series:
+        return True
+    else:
+        return False
+
+if __name__ == "__main__": 
+    infile1 = '../test/test_icemask.shp'
+    test1 = load(infile1)
+
+    g = gpd.read_file(infile1)
+    test2 = load(g)
+
+    infile2 = '../test/test_icemsak.shp'
+    test2 = load(infile2)

src/griml/merge/merge_vectors.py

@@ -4,29 +4,24 @@
 import geopandas as gpd
 import pandas as pd
 from glob import glob
+from griml.load import load
 
-def merge_vectors(feature_list, method_list, collection_list, start_date_list,
-                  end_date_list, proj='EPSG:3413'):
+def merge_vectors(inlist, outfile=None, proj='EPSG:3413'):
     '''Compile features from multiple processings into one geodataframe
 
     Parameters
     ----------
-    feature_list : list
-        List of shapely features
-    method_list : list
-        List of strings denoting processing method
-    collection_list : list
-        List of strings denoting dataset collection
-    date_list : list
-        List of start and end date for processing
+    inlist : list
+        List of files or geopandas.dataframe.DataFrame objects to merge
 
     Returns
     -------
-    all_gdf : geopandas.GeoDataFrame
+    all_gdf : geopandas.dataframe.GeoDataFrame
         Compiled goedataframe
     '''
+    feature, method, collection, start_date, end_date = _load_all(inlist)
     dfs=[]
-    for a,b,c,d,e in zip(feature_list, method_list, collection_list, start_date_list, end_date_list):
+    for a,b,c,d,e in zip(feature, method, collection, start_date, end_date):
         if a is not None:
 
             #Construct geodataframe with basic metadata
@@ -48,10 +43,51 @@ def merge_vectors(feature_list, method_list, collection_list, start_date_list,
 
     all_gdf['area_sqkm'] = all_gdf['geometry'].area/10**6
     all_gdf['length_km'] = all_gdf['geometry'].length/1000
-    all_gdf = gpd.GeoDataFrame(all_gdf, geometry=all_gdf.geometry, 
-                                crs=proj)
+    # all_gdf = gpd.GeoDataFrame(all_gdf, geometry=all_gdf.geometry, 
+    #                             crs=proj)
+
+    if outfile is not None:    
+        print('Saving file...')
+        all_gdf.to_file(outfile)
+        print('Saved to '+str(outfile))
+
     return all_gdf
 
+def _load_all(inlist):
+    '''Load info from all features for merging
+
+    Parameters
+    ----------
+    inlist : list
+        List of files or geodataframe.dataframe.DataFrame objects to load info
+        from
+
+    Returns
+    -------
+    feature_list : list
+        List of shapely features
+    method_list : list
+        List of strings denoting processing method
+    collection_list : list
+        List of strings denoting dataset collection
+    date_list : list
+        List of start and end date for processing
+    ''' 
+    features=[]
+    methods=[]
+    sources=[]
+    starts=[]
+    ends=[]
+
+    for f in inlist:
+        i = load(f)
+        features.append(list(i['geometry']))
+        methods.append(list(i['method']))
+        sources.append(list(i['source']))
+        starts.append(list(i['startdate']))
+        ends.append(list(i['enddate']))
+
+    return features, methods, sources, starts, ends
 
 def _dissolve_vectors(gdf):
     '''Dissolve overlapping polygons in a Pandas GeoDataFrame
@@ -72,26 +108,6 @@ def _dissolve_vectors(gdf):
     return gdf2
 
 if __name__ == "__main__": 
-    indir = "/home/pho/python_workspace/GrIML/other/iml_2017/test/*.shp"
-    outfile = "/home/pho/python_workspace/GrIML/other/iml_2017/merged_vectors/griml_2017_inventory.shp"
-    features=[]
-    methods=[]
-    sources=[]
-    starts=[]
-    ends=[]
-
-    for f in list(glob(indir)):
-        i = gpd.read_file(f)
-
-        features.append(list(i['geometry']))
-        methods.append(list(i['method']))
-        sources.append(list(i['source']))
-        starts.append(list(i['startdate']))
-        ends.append(list(i['enddate']))
-    vectors = merge_vectors(features, methods, sources, starts, ends)
-
-    # features.append(i)
-    # vectors = pd.concat(features, ignore_index=True)
-
-
-    vectors.to_file(outfile)
+        infile1 = '../test/test_merge_1.shp'
+        infile2 = '../test/test_merge_2.shp'               
+        vectors = merge_vectors([infile1,infile2])