Merge pull request #34 from EGaraldi/better-tests

Better tests
EGaraldi · Sep 3, 2024 · 50dec9f · 50dec9f
2 parents 132978c + 9082c05
commit 50dec9f
Show file tree

Hide file tree

Showing 12 changed files with 292 additions and 15 deletions.
diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml
@@ -5,7 +5,7 @@ name: Python package
 
 on:
   push:
-    branches: [ "master" ]
+    branches: [ "master", "better-tests" ]
   pull_request:
     branches: [ "master" ]
 
@@ -16,7 +16,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.8", "3.9", "3.10"]
+        python-version: ["3.8", "3.9", "3.10", "3.11", "3.12"]
 
     steps:
     - uses: actions/checkout@v3
@@ -37,4 +37,4 @@ jobs:
 #        flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
     - name: Test with pytest
       run: |
-        pytest --disable-pytest-warnings
+        pytest --disable-pytest-warnings tests
diff --git a/.gitignore b/.gitignore
@@ -11,4 +11,4 @@ HOWTO
 MANIFEST
 corecon/__init__.py.bkp
 time_of_last_update.dat
-
+tests/__pycache__
diff --git a/corecon/DataEntryClass.py b/corecon/DataEntryClass.py
@@ -50,15 +50,17 @@ def __init__(self,
                 self.extra_data.append(k)
         self.extra_data = np.array(self.extra_data)
 
-        #create named entries
+        #create named entries and dimensions_descriptors_internal
         setattr(self, parent_field, values[:])
+        self.dimensions_descriptors_internal = []
         for k, descr in enumerate(dimensions_descriptors):
             descr = descr.replace(" ", "_")
             descr = ''.join(ch if ch.isalnum() or ch=="_" else '' for ch in descr)
             if ndim==1:
                 setattr(self, descr, axes)
             else:
                 setattr(self, descr, axes[:,k])
+            self.dimensions_descriptors_internal.append(descr)
 
     def __repr__(self):
         """string describing the class

diff --git a/corecon/FieldClass.py b/corecon/FieldClass.py
@@ -17,6 +17,10 @@ def __init__(self, *arg, **kw):
         self.field_description = None
         self.field_units       = None
         self.field_remarks     = None
+
+    # return a copy of the item, to preserve the originally-loaded item
+    #def __getitem__(self, key):
+    #    return copy.deepcopy(super().__getitem__(key))
 
     def __str__(self):
         return super().__str__()
@@ -92,7 +96,7 @@ def filter_by_redshift_range(self, zmin, zmax):
         dict_zslice.field_description = self.field_description 
 
         for k in self.keys():
-            if not 'redshift' in self[k].dimensions_descriptors:
+            if not 'redshift' in self[k].dimensions_descriptors_internal:
                 print("WARNING: missing redshift dimension for entry %s. Skipping it."%(k))
                 continue
 
@@ -106,7 +110,7 @@ def filter_by_redshift_range(self, zmin, zmax):
                 dict_zslice[k].upper_lim = dict_zslice[k].upper_lim[w]
                 dict_zslice[k].lower_lim = dict_zslice[k].lower_lim[w]
                 #variables auto-created from "axes"
-                for dd in self[k].dimensions_descriptors:
+                for dd in self[k].dimensions_descriptors_internal:
                     _temp = getattr(dict_zslice[k], dd)
                     setattr(dict_zslice[k], dd, _temp[w])
                 #variables auto-created from"values"
@@ -160,7 +164,7 @@ def get_lower_limits(self):
                 dict_lls[k].upper_lim = dict_lls[k].upper_lim[self[k].lower_lim]
                 dict_lls[k].lower_lim = dict_lls[k].lower_lim[self[k].lower_lim]
                 #variables auto-created from "axes"
-                for dd in self[k].dimensions_descriptors:
+                for dd in self[k].dimensions_descriptors_internal:
                     _temp = getattr(dict_lls[k], dd)
                     setattr(dict_lls[k], dd, _temp[self[k].lower_lim])
                 #variables auto-created from"values"
@@ -195,7 +199,7 @@ def get_upper_limits(self):
                 dict_uls[k].upper_lim = dict_uls[k].upper_lim[self[k].upper_lim]
                 dict_uls[k].lower_lim = dict_uls[k].lower_lim[self[k].upper_lim]
                 #variables auto-created from "axes"
-                for dd in self[k].dimensions_descriptors:
+                for dd in self[k].dimensions_descriptors_internal:
                     _temp = getattr(dict_uls[k], dd)
                     setattr(dict_uls[k], dd, _temp[self[k].upper_lim])
                 #variables auto-created from"values"
@@ -207,4 +211,3 @@ def get_upper_limits(self):
                     setattr(dict_uls[k], e, _temp[self[k].upper_lim])
 
         return dict_uls
-
diff --git a/corecon/data/UV_luminosity_function/Livermore_et_al_2017.py b/corecon/data/UV_luminosity_function/Livermore_et_al_2017.py
@@ -1,4 +1,4 @@
-dictionary_tag         = "Livermore et a. 2017"
+dictionary_tag         = "Livermore et al. 2017"
 
 reference              = "Livermore, Finkelstein, Lotz; ApJ. 835, 113 (2017)"
 

diff --git a/corecon/data/data.zip b/corecon/data/data.zip
diff --git a/corecon/loaders.py b/corecon/loaders.py
@@ -50,7 +50,12 @@ def _expand_field(field, shape):
     extra_data = {}
     for k in local_var_dict.keys():
         extra_data[k] = np.array(local_var_dict[k], dtype=object)
-
+        #convert to float whenever possible
+        try:
+            extra_data[k] = extra_data[k].astype(np.float64)
+        except ValueError:
+            pass
+
     #expand None's, True's, and False's (this will also convert them to array)
     err_up    = _expand_field(err_up   , values.shape)
     err_down  = _expand_field(err_down , values.shape)

diff --git a/test_import.py b/test_import.py
diff --git a/tests/__init__.py b/tests/__init__.py
diff --git a/tests/test_DataEntryClass.py b/tests/test_DataEntryClass.py
@@ -0,0 +1,76 @@
+import corecon as crc
+import numpy as np
+
+
+def test_swap_limits():
+    uvlf = crc.get("UVLF")
+    h23 = uvlf["Harikane et al. 2023"]
+
+    h23.swap_limits()
+    assert np.all( h23.lower_lim == [True, True, False, False, False, False, True, True, False, False, False, False, True, False]),\
+           "Problem detected in DataEntry.swap_limit"
+
+def test_swap_errors():
+    uvlf = crc.get("UVLF")
+    b17 = uvlf["Bouwens et al. 2017"]
+
+    b17.swap_errors()
+    assert np.all(b17.err_down == [0.30103   , 0.15970084, 0.19629465, 0.12493874, 0.09691001, 0.07918125, 0.28546222, 0.03817964, 
+                                   0.29459589, 0.34325599, 0.26787824, 0.35483287, 0.09089366, 0.78861996, 1.14378603, 1.58134989, 2.10047281] ), \
+           "Problem detected in DataEntry.swap_errors"
+
+def test_nan_to_values_all_fields():
+    uvlf = crc.get("UVLF")
+    k22 = uvlf["Kauffmann et al. 2022"]
+
+    k22.nan_to_values("all", 1e-10)
+    assert np.all(k22.err_down == [1.0000000e-10, 3.5139073e-01, 3.4584234e-01, 4.0866387e-01, 5.4104580e-01, 1.0000000e+02]),\
+           "Problem detected in DataEntry.nan_to_vals using 'all' as first argument"
+    assert np.all(k22.err_up == [1.0000000e-10, 1.9165904e-01, 1.9005692e-01, 2.0676008e-01, 2.3357745e-01, 3.4987856e-01]),\
+           "Problem detected in DataEntry.nan_to_vals using 'all' as first argument"
+
+def test_nan_to_values_one_field():
+    uvlf = crc.get("UVLF")
+    k22 = uvlf["Kauffmann et al. 2022"]
+
+    k22.nan_to_values("err_down", 1e-10)
+    assert np.all(k22.err_down == [1.0000000e-10, 3.5139073e-01, 3.4584234e-01, 4.0866387e-01, 5.4104580e-01, 1.0000000e+02]),\
+           "Problem detected in DataEntry.nan_to_vals using a single field as first argument"
+
+def test_nan_to_values_list_fields():
+    uvlf = crc.get("UVLF")
+    k22 = uvlf["Kauffmann et al. 2022"]
+
+    k22.nan_to_values(["err_down", "err_up"], 1e-10)
+    assert np.all(k22.err_down == [1.0000000e-10, 3.5139073e-01, 3.4584234e-01, 4.0866387e-01, 5.4104580e-01, 1.0000000e+02]),\
+           "Problem detected in DataEntry.nan_to_vals using a list of fields as first argument"
+    assert np.all(k22.err_up == [1.0000000e-10, 1.9165904e-01, 1.9005692e-01, 2.0676008e-01, 2.3357745e-01, 3.4987856e-01]),\
+           "Problem detected in DataEntry.nan_to_vals using a list of fields as first argument"
+
+def test_set_lim_errors():
+    uvlf = crc.get("UVLF")
+    h23 = uvlf["Harikane et al. 2023"]
+
+    h23.set_lim_errors(5, frac_of_values=False)
+    assert np.all(h23.err_up == [5.0000000e+00, 5.0000000e+00, 5.2569252e-01, 5.2542593e-01, 3.4821760e-02, 3.9757000e-03, 
+                                 5.0000000e+00, 5.0000000e+00, 5.2009033e-01, 3.6845013e-01, 2.9921665e-01, 3.7791135e-01, 
+                                 5.0000000e+00, 3.6835150e-01]),\
+           "Problem detected in DataEntry.set_lim_errors using frac_of_values=False"
+
+def test_set_lim_errors_frac():
+    uvlf = crc.get("UVLF")
+    h23 = uvlf["Harikane et al. 2023"]
+
+    h23.set_lim_errors(0.5, frac_of_values=True)
+    assert np.all(h23.err_up == [-2.0790076 , -2.05760232,  0.52569252,  0.52542593,  0.03482176, 0.0039757 , -2.616422065, 
+                                 -2.596910015,  0.52009033,  0.36845013, 0.29921665,  0.37791135, -2.808092315,  0.3683515 ]),\
+           "Problem detected in DataEntry.set_lim_errors using frac_of_values=True"
+
+def test_list_attributes():
+    uvlf = crc.get("UVLF")
+    h23 = uvlf["Harikane et al. 2023"]
+
+    assert np.all(h23.list_attributes() == ['ndim', 'description','reference','parent_field','url','dimensions_descriptors',
+                                            'extracted','axes','values','err_up','err_down','upper_lim','lower_lim','extra_data',
+                                            'err_right','err_left','UV_luminosity_function', 'dimensions_descriptors_internal','redshift','M_UV']),\
+            "Problem detected in DataEntry.list_attributes"
diff --git a/tests/test_FieldClass.py b/tests/test_FieldClass.py
@@ -0,0 +1,164 @@
+import corecon as crc
+import numpy as np
+
+
+def test_get_all_references():
+    uvlf = crc.get("UVLF")
+
+    #We test that *at least* the following references are returned. In this way, we do not have to update this test every time a 
+    # new constraint is added (or a temporary one updated)
+    references = ['Ishigaki, Kawamata, Ouchi, Oguri, Shimasaku, Ono; ApJ. 854, 73 (2018)',
+                  'McLeod, Donnan, McLure, Dunlop, Magee, Begley, Carnall, et al.; MNRAS 527, 5004 (2024)',
+                  'Atek, Richard, Kneib, Schaerer; MNRAS 479, 5184 (2018)',
+                  'Oesch, Bouwens, Illingworth, Labbe, Smit, Franx, et al.; ApJ. 786, 108 (2014)',
+                  'Morishita, Trenti, Stiavelli, Bradley, Coe, et al.; ApJ. 867, 150 (2018)',
+                  'Perez-Gonzalez, Costantin, Langeroodi, Rinaldi, Annunziatella, et al.; ApJL 951, L1 (2023)',
+                  'Bouwens, Oesch, Labbe, Illingworth, Fazio, et al.; ApJ. 830, 67 (2016)',
+                  'Livermore, Finkelstein, Lotz; ApJ. 835, 113 (2017)',
+                  'Harikane, Ouchi, Oguri, Ono, Nakajima, Isobe, Umeda, Mawatari, Zhang; ApJS 265, 5 (2023)',
+                  'Atek, Richard, Kneib, Jauzac, Schaerer, Clement, et al.; ApJ 800, 18 (2015)',
+                  'Bouwens, Illingworth, Oesch, Naidu, van Leeuwen, Magee, MNRAS 523, 1009 (2023)',
+                  'McLeod, McLure, Dunlop; MNRAS 459, 3812 (2016)',
+                  'Donnan, McLeod, McLure, Dunlop, Carnall, Cullen, Magee; MNRAS 520, 4554 (2023)',
+                  'Finkelstein, Ryan, Papovich, Dickinson, Song, et al.; ApJ. 810, 71 (2015)',
+                  'Stefanon, Labbe, Bouwens, Oesch, Ashby, Caputi, et al.; ApJ. 883, 99 (2019)',
+                  'Bouwens, Illingworth, Oesch, Trenti, Labbe, et al.; ApJ. 803, 34 (2015)',
+                  'Rojas-Ruiz S., Finkelstein S. L., Bagley M. B., Stevans M., Finkelstein K. D.,  et al., 2020, ApJ, 891, 146',
+                  'McLure, Dunlop, Bowler, Curtis-Lake, Schenker, et al.; MNRAS 432, 2696 (2013)',
+                  'Bouwens, Oesch, Illingworth, Ellis, Stefanon; ApJ. 843, 129 (2017)',
+                  'Bowler, Jarvis, Dunlop, McLure, McLeod, et al.; MNRAS 493, 2059 (2020)',
+                  'Leung, Bagley, Finkelstein, Ferguson, Koekemoer, Perez-Gonzalez, et al.; ApJL 954, L46 (2023)',
+                  'Donnan, McLeod, Dunlop, McLure, Carnall, Begley, Cullen, et al.; MNRAS 518, 6011 (2023)',
+                  'Bouwens, Stefanon, Brammer, Oesch, Herard-Demanche, Illingworth, et al., MNRAS 523, 1936 (2023)',
+                  'Castellano, Dayal, Pentericci, Fontana, Hutter, et al.; ApJL. 818, L3 (2016)',
+                  'Oesch, Bouwens, Illingworth, Labbe, Stefanon; ApJ. 855, 105 (2018)',
+                  'Bouwens, Illingworth, Ellis, Oesch, Stefanon; ApJ 940, 55 (2022)',
+                  'Bowler, Dunlop, McLure, McCracken, Milvang-Jensen, et al.; MNRAS 452, 1817 (2015)'
+                  ]
+
+    uvlf_ref = uvlf.get_all_references()
+
+    assert len(set(references).difference(uvlf_ref))==0, "Problem detected in Field.get_all_references"
+
+
+
+
+
+def test_get_all_urls():
+    uvlf = crc.get("UVLF")
+
+    #We test that *at least* the following URLs are returned. In this way, we do not have to update this test every time a 
+    # new constraint is added (or a temporary one updated)
+    urls = ['https://iopscience.iop.org/article/10.3847/1538-4357/aaa544',
+            'https://academic.oup.com/mnras/article/527/3/5004/7408621',
+            'https://academic.oup.com/mnras/article/479/4/5184/5050078',
+            'https://iopscience.iop.org/article/10.1088/0004-637X/786/2/108',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/aae68c',
+            'https://iopscience.iop.org/article/10.3847/2041-8213/acd9d0/pdf',
+            'https://iopscience.iop.org/article/10.3847/0004-637X/830/2/67',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/835/2/113',
+            'https://iopscience.iop.org/article/10.3847/1538-4365/acaaa9',
+            'https://ui.adsabs.harvard.edu/abs/2015ApJ...800...18A/abstract',
+            'https://ui.adsabs.harvard.edu/abs/2023MNRAS.523.1009B/',
+            'https://academic.oup.com/mnras/article/459/4/3812/2624050',
+            'https://ui.adsabs.harvard.edu/abs/2023MNRAS.520.4554D/abstract',
+            'https://iopscience.iop.org/article/10.1088/0004-637X/810/1/71',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/ab3792',
+            'https://iopscience.iop.org/article/10.1088/0004-637X/803/1/34',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/ab7659',
+            'https://academic.oup.com/mnras/article/432/4/2696/2907730',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/aa70a4',
+            'https://academic.oup.com/mnras/article/493/2/2059/5721544',
+            'https://iopscience.iop.org/article/10.3847/2041-8213/acf365/pdf',
+            'https://academic.oup.com/mnras/article/518/4/6011/6849970',
+            'https://ui.adsabs.harvard.edu/abs/2023MNRAS.523.1036B/abstract',
+            'https://iopscience.iop.org/article/10.3847/2041-8205/818/1/L3',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/aab03f',
+            'https://iopscience.iop.org/article/10.3847/1538-4357/ac86d1',
+            'https://academic.oup.com/mnras/article/452/2/1817/1068199']
+
+    uvlf_urls = uvlf.get_all_urls()
+
+    assert len(set(urls).difference(uvlf_urls))==0, "Problem detected in Field.get_all_urls"
+
+
+
+
+
+def test_filter_by_redshift_range():
+    uvlf = crc.get("UVLF")
+
+    #We test that *at least* the following entries are returned. In this way, we do not have to update this test every time a 
+    # new constraint is added (or a temporary one updated)
+    entries = ['Ishigaki et al. 2018', 'McLeod et al. 2024', 'Oesch et al. 2014',
+               'Morishita et al. 2018', 'Perez-Gonzalez et al. 2023',
+               'Bouwens et al. 2016', 'Harikane et al. 2023',
+               'Bouwens et al. 2023a', 'McLeod et al. 2016',
+               'Donnan et al. 2023b', 'Stefanon et al. 2019',
+               'Bouwens et al. 2015', 'Rojas-Ruiz et al. 2020',
+               'Bowler et al. 2020', 'Leung et al. 2023', 'Donnan et al. 2023a',
+               'Bouwens et al. 2023b', 'Oesch et al. 2018', 'Bouwens et al. 2022']
+
+    uvlf_zrange = uvlf.filter_by_redshift_range(9,13)
+
+    assert len(set(entries).difference(uvlf_zrange))==0, "Problem detected in Field.filter_by_redshift_range (not all entries are returned)"
+
+    #then test the slicing is correct
+    assert np.all(uvlf_zrange['Ishigaki et al. 2018'].redshift == [9.0, 9.0, 9.0, 9.0, 9.0, 9.0, 9.0, 9.0, 9.0, 9.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0]), \
+           "Problem detected in Field.filter_by_redshift_range (slicing of constraints is wrong)"
+
+
+
+
+
+def test_filter_by_extracted():
+
+    uvlf = crc.get("UVLF")
+
+    #We test that *at least* the following entries are returned. In this way, we do not have to update this test every time a 
+    # new constraint is added (or a temporary one updated)
+    entries = ['Ishigaki et al. 2018',  'Atek et al. 2018', 'Atek et al. 2015', 'McLeod et al. 2016',  'Castellano et al. 2015']
+
+    uvlf_ex = uvlf.filter_by_extracted(True)
+
+    assert len(set(entries).difference(uvlf_ex))==0, "Problem detected in Field.filter_by_extracted (not all entries are returned)"
+
+    for e in entries:
+        assert uvlf_ex[e].extracted == True, "Problem detected in Field.filter_by_extracted (returned wrong entry/slice)"
+
+
+
+
+def test_get_upper_limits():
+    xHII = crc.get("x_HII")
+
+    #We test that *at least* the following entries are returned. In this way, we do not have to update this test every time a 
+    # new constraint is added (or a temporary one updated)
+    entries = ['Schenker et al. 2014','Sobacchi & Mesinger 2015', 'Schroeder et al. 2013', 'Mortlock et al. 2011', 'Mesinger et al. 2015', 'Mason et al. 2019', 
+            'Tilvi et al. 2014', 'Pentericci et al. 2014', 'Bosman et al. 2022', 'Yang et al. 2020b', 'Robertson et al. 2013', 'Hoag et al. 2019']
+
+    xHII_ul = xHII.get_upper_limits()
+
+    assert len(set(entries).difference(xHII_ul))==0, "Problem detected in Field.get_upper_limits (not all entries are returned)"
+
+    for e in entries:
+        assert np.all(xHII_ul[e].upper_lim), "Problem detected in Field.get_upper_limits (returned wrong entry/slice)"
+
+
+
+
+def test_get_lower_limits():
+    xHII = crc.get("x_HII")
+
+    #We test that *at least* the following entries are returned. In this way, we do not have to update this test every time a 
+    # new constraint is added (or a temporary one updated)
+    entries = ['Zhu et al. 2022', 'Ouchi et al. 2010', 'Chornock et al. 2013', 'Totani et al. 2006', 'Nakane et al. 2024',
+               'Lu et al. 2020', 'McGreer et al. 2011','McGreer et al. 2015']
+
+    xHII_ll = xHII.get_lower_limits()
+
+    assert len(set(entries).difference(xHII_ll))==0, "Problem detected in Field.get_lower_limits (not all entries are returned)"
+
+    for e in entries:
+        assert np.all(xHII_ll[e].lower_lim), "Problem detected in Field.get_lower_limits (returned wrong entry/slice)"
+
diff --git a/tests/test_corecon.py b/tests/test_corecon.py
@@ -0,0 +1,30 @@
+import corecon as crc
+import numpy as np
+
+def test_get_fields():
+
+    #We test that *at least* the following fields are returned. In this way, we do not have to update this test every time a 
+    # new constraint field is added
+    fields = ['HII_fraction', 'HeIII_fraction', 'Lya_flux_power_spectrum', 'mean_free_path', 'effective_optical_depth_HI_Lya',
+              'effective_optical_depth_HeII_Lya', 'HeII_to_HI_column_density_ratio', 'quasar_luminosity_function',
+              'UV_luminosity_function','IGM_temperature_mean_density','optical_depth_CMB','sfrd',
+              'Lya_spike_galaxy_correlation','mass_stellar_metallicity_relation','mass_gas_metallicity_relation',
+              'galaxy_main_sequence','UV_slope','ionizing_photons_production_efficiency','HI_photoionization_rate',
+              'ionizing_photons_emission_rate','reionization_midpoint','UV_luminosity_density']
+
+
+    crc_fields = crc.get_fields()
+
+    assert len(set(fields).difference(crc_fields))==0, "Problem detected in CoReCon.get_fields"
+
+
+
+def test_get_field_synonyms():
+    #We test that *at least* the following synonyms are returned. In this way, we do not have to update this test every time a 
+    # new one is added
+    synonyms = ['ionized_fraction', 'x_HII', 'f_HII', 'x_ion', 'f_ion', 'ion_frac']
+
+    crc_syn = crc.get_field_synonyms("HII_fraction")
+
+    assert len(set(synonyms).difference(crc_syn))==0, "Problem detected in CoReCon.get_field_synonym"
+