diff --git a/a_prep_dat/compile_250m_hok_db.py b/a_prep_dat/compile_250m_hok_db.py
index 7aa60ec19e7fc43f50bfd86caf0ecbc9086f58ca..5345970cfdd78b5a7ad5925d8b6fd2fde58d85de 100644
--- a/a_prep_dat/compile_250m_hok_db.py
+++ b/a_prep_dat/compile_250m_hok_db.py
@@ -43,7 +43,7 @@ sources = ['N1900.pkl', 'N1800.pkl', 'N1706.pkl', 'N1705.pkl', 'N1704.pkl', 'N17
'N0201.pkl', 'OpenDuin.pkl', 'Moeras.pkl', 'Heide.pkl', 'HalfnatuurlijkGrasland.pkl', 'Bos.pkl',
'prov-Zeeland.pkl', 'prov-Noord-Holland.pkl', 'prov-geen.pkl', 'prov-Overijssel.pkl', 'prov-Groningen.pkl',
'prov-Friesland.pkl', 'prov-Limburg.pkl', 'prov-Utrecht.pkl', 'prov-Gelderland.pkl', 'prov-Flevoland.pkl',
- 'prov-Noord-Brabant.pkl', 'prov-Zuid-Holland.pkl', 'prov-Drenthe.pkl', 'twente.pkl']
+ 'prov-Noord-Brabant.pkl', 'prov-Zuid-Holland.pkl', 'prov-Drenthe.pkl', 'twente_hdr.pkl']
pkl_dir = r'd:\hotspot_working\a_broedvogels\SNL_grids\augurken'
for source in sources:
diff --git a/b_analysis/clo_1518_analysis.py b/b_analysis/clo_1518_analysis.py
index 0ba84f3a09b0a7583747cf503b355193357c1545..0cf85e175f2056b42380ffaa0066037f0b226877 100644
--- a/b_analysis/clo_1518_analysis.py
+++ b/b_analysis/clo_1518_analysis.py
@@ -112,6 +112,7 @@ hokken = nl250.query(spatial_query, engine='python') # reduce to hokken complyi
hokken['areaal_m2'] = hokken.loc[:, snl_gebieden_wel].sum(axis=1) # calculate areaal per hok of selected SNL types
hokken['areaal_ha'] = hokken.apply(lambda row: np.divide(row.areaal_m2, 10000), axis=1) # sq m to hectare
hokken.set_index('hok_id')
+hokken['provincie'] = hokken.filter(regex='^prov', axis='columns').idxmax(axis=1) # name of provincie for each hok
if hokken.empty:
raise Exception('No hokken found complying to query.')
else:
@@ -187,17 +188,19 @@ dat_piv[('2002-2009', 'CLO1518', 'score')] = clo1518_2002_2009_score
dat_piv[('2010_2017', 'CLO1518', 'score')] = clo1518_2010_2017_score
'''
-Label each hok in dat_piv with: 1) SNL BeheerType (NXXYY), 2) EcosysteemType relating to the BT 3) area in m2 of (1),
+merge PGO observation data in dat_piv with info on the hokken from hokken. Note multiindex will be removed
+'''
+
+dat_snl = pd.merge(left=dat_piv, right=hokken.loc[:, ['hok_id', 'areaal_ha', 'provincie', 'twente_hdr']],
+ left_index=True, right_on='hok_id', how='right') # how = inner, dus bewaar enkel de overlap.
+
+'''
+Label each hok in dat_snl with: 1) SNL BeheerType (NXXYY), 2) EcosysteemType relating to the BT 3) area in m2 of (1),
4) Provincie to which it is assigned
'''
-dat_piv['snl_bt'] = '_'.join(snl for snl in snl_soortlijst) # beheertypen, possibly > 1
+dat_snl['snl_bt'] = '_'.join(snl for snl in snl_soortlijst) # beheertypen, possibly > 1
ecotype = clo.bt_to_eco(snl_soortlijst[0]) # dit is verschrikkelijk om alleen [0] te gebruiken, maar goed
-dat_piv['ecosystype'] = ecotype
-hokken['provincie'] = hokken.filter(regex='^prov', axis='columns').idxmax(axis=1) # name of provincie for each hok
-
-# merge PGO observation data in dat_piv with info on the hokken from hokken. Note multiindex will be removed
-dat_snl = pd.merge(left=dat_piv, right=hokken.loc[:, ['hok_id', 'areaal_ha', 'provincie', 'twente']], left_index=True,
- right_on='hok_id', how='right') # how = inner, dus bewaar enkel de overlap.
+dat_snl['ecosystype'] = ecotype
'''
Wrap up
@@ -208,10 +211,24 @@ if len(set(dat_snl.hok_id)) != dat_snl.shape[0]:
print('Intersection between PGO observation locations {0} and SNL hokken selection {1} reduces to {2} '
'hokken'.format(dat_piv.shape[0], hokken.shape[0], dat_snl.shape[0]))
+# pretty colnames
colnames = dat_snl.columns.tolist()
new_colnames = ['_'.join([x for x in foo if len(x) > 0]) for foo in colnames if isinstance(foo, tuple)]
dat_snl.rename(columns=dict(zip(colnames, new_colnames)), inplace=True) # remove commas from colnames
+# sort columns neatly (dit moet vast makkelijker kunnen, maar daar ben ik nu te moe voor.)
+col_order = ['hok_id', 'snl_bt', 'ecosystype', 'areaal_ha', '1994-2001_vaatplant_SNL', '1994-2001_vaatplant_Bijl1',
+ '1994-2001_vogel_SNL', '1994-2001_vogel_Bijl1', '1994-2001_vlinder_SNL', '1994-2001_vlinder_Bijl1',
+ '1994-2001_Bijl1_tot', '1994-2001_Bijl1_cap', '1994-2001_SNL_tot', '1994-2001_SNL-Bijl1_tot',
+ '1994-2001_CLO1518_score', '2002-2009_vaatplant_SNL', '2002-2009_vaatplant_Bijl1',
+ '2002-2009_vogel_SNL', '2002-2009_vogel_Bijl1', '2002-2009_vlinder_SNL', '2002-2009_vlinder_Bijl1',
+ '2002-2009_Bijl1_tot', '2002-2009_Bijl1_cap', '2002-2009_SNL_tot', '2002-2009_SNL-Bijl1_tot',
+ '2002-2009_CLO1518_score', '2010-2017_vaatplant_SNL', '2010-2017_vogel_SNL', '2010-2017_vogel_Bijl1',
+ '2010-2017_vaatplant_Bijl1', '2010-2017_vlinder_SNL', '2010-2017_vlinder_Bijl1', '2010-2017_Bijl1_tot',
+ '2010-2017_Bijl1_cap', '2010-2017_SNL_tot', '2010-2017_SNL-Bijl1_tot', '2010_2017_CLO1518_score',
+ 'provincie', 'twente_hdr']
+dat_snl = dat_snl.reindex(columns=col_order)
+
# Optional TODO: pivot on CLO scores to calculate area for each category
'''
diff --git a/z_utils/merge_tabs.py b/z_utils/merge_tabs.py
index 3603061783ca4c242057004a87080585966feab8..1096c7542ba73fe02bf99ffdcc867dc2d0acc690 100644
--- a/z_utils/merge_tabs.py
+++ b/z_utils/merge_tabs.py
@@ -7,6 +7,7 @@ Hans Roelofsen, 7 nov 2019
'''
import os
+import pandas as pd
import datetime
in_dir = r'd:\clo1543\CLO1518\bt_tabel'
@@ -17,16 +18,25 @@ out_base_name = 'combined_table_bts'
timestamp_brief = datetime.datetime.now().strftime("%y%m%d-%H%M")
i = 0
-with open(os.path.join(out_dir, '{0}_{1}.txt'.format(out_base_name, timestamp_brief)), 'w') as f:
- for csv_file in in_csv:
+holder = []
+for csv_file in in_csv:
+ df = pd.read_csv(os.path.join(in_dir, csv_file), comment='#', sep=';')
+ df['source_file'] = csv_file
+ holder.append(df)
+
+df_out = pd.concat(holder)
+df_out.to_csv(os.path.join(out_dir, '{0}_{1}.txt'.format(out_base_name, timestamp_brief)), index=False, sep=';')
+'''
+
with open(os.path.join(in_dir, csv_file)) as csv:
if i == 0:
for line in csv.readlines():
- if not (line.startswith('#') or line.endswith('wente')):
+ if not (line.startswith('#')):
f.write('{0};{1}\n'.format(line.rstrip('\n'), csv_file))
else:
for line in csv.readlines():
- if not (line.startswith('#')):
+ if not (line.startswith('#') or line.endswith('wente_hdr')):
f.write('{0};{1}\n'.format(line.rstrip('\n'), csv_file))
i += 1
+'''
\ No newline at end of file