Merge branch 'develop' into hotfix_hfb_pd_lineend

Author: briochh

autotest/utils_tests.py

@@ -1661,7 +1661,7 @@ def fieldgen_dev():
 
 
 if __name__ == "__main__":
-    # fieldgen_dev()
+    #fieldgen_dev()
     # smp_test()
     # smp_dateparser_test()
     # smp_to_ins_test()

pyemu/utils/gw_utils.py

@@ -1855,8 +1855,9 @@ def apply_sfr_obs():
 
 
 def load_sfr_out(sfr_out_file, selection=None):
-    """load an ASCII SFR output file into a dictionary of kper: dataframes.  aggregates
-    flow to aquifer for segments and returns and flow out at downstream end of segment.
+    """load an ASCII SFR output file into a dictionary of kper: dataframes.
+    aggregates flow to aquifer for segments and returns and flow out at
+    downstream end of segment.
 
     Parameters
     ----------
@@ -1877,10 +1878,13 @@ def load_sfr_out(sfr_out_file, selection=None):
     if selection is None:
         pass
     elif isinstance(selection, str):
-        assert selection == 'all', "If string passed as selection only 'all' allowed: {}".format(selection)
+        assert selection == 'all', \
+            "If string passed as selection only 'all' allowed: " \
+            "{}".format(selection)
     else:
-        assert isinstance(
-            selection, pd.DataFrame), "'selection needs to be pandas Dataframe. Type {} passed.".format(type(selection))
+        assert isinstance(selection, pd.DataFrame), \
+            "'selection needs to be pandas Dataframe. " \
+            "Type {} passed.".format(type(selection))
         assert np.all([sr in selection.columns for sr in ['segment', 'reach']]
                       ), "Either 'segment' or 'reach' not in selection columns"
     with open(sfr_out_file) as f:
@@ -1905,34 +1909,47 @@ def load_sfr_out(sfr_out_file, selection=None):
                     dlines.append(draw)
                 df = pd.DataFrame(data=np.array(dlines)).iloc[:, [3, 4, 6, 7]]
                 df.columns = ["segment", "reach", "flaqx", "flout"]
-                df.loc[:, "segment"] = df.segment.apply(np.int)
-                df.loc[:, "reach"] = df.reach.apply(np.int)
-                df.loc[:, "flaqx"] = df.flaqx.apply(np.float)
-                df.loc[:, "flout"] = df.flout.apply(np.float)
-                df.index = df.apply(lambda x: "{0:03d}_{1:03d}".format(int(x.segment), int(x.reach)), axis=1)
+                df["segment"] = df.segment.astype(np.int)
+                df["reach"] = df.reach.astype(np.int)
+                df["flaqx"] = df.flaqx.astype(np.float)
+                df["flout"] = df.flout.astype(np.float)
+                df.index = ["{0:03d}_{1:03d}".format(s, r) for s, r in
+                            np.array([df.segment.values, df.reach.values]).T]
+                # df.index = df.apply(
+                # lambda x: "{0:03d}_{1:03d}".format(
+                # int(x.segment), int(x.reach)), axis=1)
                 if selection is None:  # setup for all segs, aggregate
                     gp = df.groupby(df.segment)
                     bot_reaches = gp[['reach']].max().apply(
-                        lambda x: "{0:03d}_{1:03d}".format(int(x.name), int(x.reach)), axis=1)
-                    df2 = pd.DataFrame(index=gp.groups.keys(), columns=['flaqx', 'flout'])
-                    df2['flaqx'] = gp.flaqx.sum()  # only sum distributed output
-                    df2['flout'] = df.loc[bot_reaches, 'flout'].values  # take flow out of seg
+                        lambda x: "{0:03d}_{1:03d}".format(
+                            int(x.name), int(x.reach)), axis=1)
+                    # only sum distributed output # take flow out of seg
+                    df2 = pd.DataFrame(
+                        {'flaqx':gp.flaqx.sum(),
+                         'flout': df.loc[bot_reaches, 'flout'].values},
+                        index=gp.groups.keys())
                     # df = df.groupby(df.segment).sum()
-                    df2.loc[:,"segment"] = df2.index
+                    df2["segment"] = df2.index
                 elif isinstance(selection, str) and selection == 'all':
                     df2 = df
                 else:
-                    seg_reach_id = selection.apply(lambda x: "{0:03d}_{1:03d}".
-                                                   format(int(x.segment), int(x.reach)), axis=1).values
+                    seg_reach_id = selection.apply(
+                        lambda x: "{0:03d}_{1:03d}".format(
+                            int(x.segment), int(x.reach)), axis=1).values
                     for sr in seg_reach_id:
                         if sr not in df.index:
                             s, r = [x.lstrip('0') for x in sr.split('_')]
-                            warnings.warn("Requested segment reach pair ({0},{1}) is not in sfr output. Dropping...".
-                                          format(int(r), int(s)), PyemuWarning)
-                            seg_reach_id = np.delete(seg_reach_id, np.where(seg_reach_id == sr), axis=0)
+                            warnings.warn(
+                                "Requested segment reach pair ({0},{1}) "
+                                "is not in sfr output. Dropping...".format(
+                                    int(r), int(s)), PyemuWarning)
+                            seg_reach_id = np.delete(
+                                seg_reach_id,
+                                np.where(seg_reach_id == sr), axis=0)
                     df2 = df.loc[seg_reach_id].copy()
                 if kper in sfr_dict.keys():
-                    print("multiple entries found for kper {0}, replacing...".format(kper))
+                    print("multiple entries found for kper {0}, "
+                          "replacing...".format(kper))
                 sfr_dict[kper] = df2
     return sfr_dict
 

pyemu/utils/helpers.py

@@ -3600,8 +3600,7 @@ def apply_list_pars():
                     fmts += " %9G"
         np.savetxt(os.path.join(model_ext_path, fname), df_list.loc[:, names].values, fmt=fmts)
 
-
-def apply_hfb_pars(par_file='hfb6_pars.csv'):
+def apply_hfb_pars():
     """ a function to apply HFB multiplier parameters.  Used to implement
     the parameterization constructed by write_hfb_zone_multipliers_template()
 
@@ -3613,7 +3612,7 @@ def apply_hfb_pars(par_file='hfb6_pars.csv'):
 
     should be added to the forward_run.py script
     """
-    hfb_pars = pd.read_csv(par_file)
+    hfb_pars = pd.read_csv('hfb6_pars.csv')
 
     hfb_mults_contents = open(hfb_pars.mlt_file.values[0], 'r').readlines()
     skiprows = sum([1 if i.strip().startswith('#') else 0 for i in hfb_mults_contents]) + 1
@@ -3636,11 +3635,9 @@ def apply_hfb_pars(par_file='hfb6_pars.csv'):
     # write the results
     with open(hfb_pars.model_file.values[0], 'w') as ofp:
         [ofp.write('{0}\n'.format(line.strip())) for line in header]
-        ofp.flush()
+
         hfb_org[['lay', 'irow1','icol1','irow2','icol2', 'hydchr']].to_csv(ofp, sep=' ',
-                header=None, index=None, mode='a', line_terminator='\n')
-    # hfb_org[['lay', 'irow1', 'icol1', 'irow2', 'icol2', 'hydchr']].to_csv(os.path.join('temp', 'test.csv'), sep=' ',
-    #                header=None, index=None)
+                header=None, index=None)
 
 def write_const_tpl(name, tpl_file, suffix, zn_array=None, shape=None, spatial_reference=None,
                     longnames=False):