taxcalc 4.3.5__py3-none-any.whl → 4.4.1__py3-none-any.whl

taxcalc/tests/test_compatible_data.py

@@ -12,7 +12,9 @@ plug-in pytest-xdist is able to run all parametrized functions in parallel
 import copy
 import pytest
 import numpy as np
-from taxcalc import Policy, Records, Calculator  # pylint: disable=import-error
+from taxcalc.policy import Policy
+from taxcalc.records import Records
+from taxcalc.calculator import Calculator
 
 
 @pytest.fixture(scope='module', name='allparams')
@@ -49,7 +51,7 @@ def test_compatible_data_presence(allparams):
         return True
 
     # Main logic of test_compatible_data_presence function
-    problem_pnames = list()
+    problem_pnames = []
     for pname in allparams:
         if 'compatible_data' in allparams[pname]:
             compatible_data = allparams[pname]['compatible_data']
@@ -61,7 +63,7 @@ def test_compatible_data_presence(allparams):
         msg = '{} has no or invalid compatible_data field'
         for pname in problem_pnames:
             print(msg.format(pname))
-        assert 'list of problem_pnames' == 'empty list'
+        assert False, 'ERROR: list of problem_pnames is above'
 
 
 XX_YEAR = 2019
@@ -142,7 +144,7 @@ BATCHES = int(np.floor(NPARAMS / BATCHSIZE)) + 1
 
 
 @pytest.fixture(scope='module', name='allparams_batch',
-                params=[i for i in range(0, BATCHES)])
+                params=list(range(0, BATCHES)))
 def fixture_allparams_batch(request, allparams, sorted_param_names):
     """
     Fixture for grouping Tax-Calculator parameters
@@ -229,8 +231,8 @@ def test_compatible_data(cps_subsample, puf_subsample,
     at least one of these reforms when using datasets marked compatible
     and does not differ when using datasets marked as incompatible.
     """
-    # pylint: disable=too-many-arguments,too-many-locals
-    # pylint: disable=too-many-statements,too-many-branches
+    # pylint: disable=too-many-arguments,too-many-positional-arguments
+    # pylint: disable=too-many-statements,too-many-branches,too-many-locals
 
     # Check NPARAMS value
     assert NPARAMS == len(allparams)
@@ -335,4 +337,4 @@ def test_compatible_data(cps_subsample, puf_subsample,
     # test failure if any errors
     if errors:
         print(errors)
-        assert 'compatible_data' == 'invalid'
+        assert False, 'ERROR: compatible_data is invalid; see errors above'

taxcalc/tests/test_consumption.py

@@ -1,19 +1,24 @@
+"""
+Test Consumption class and its methods.
+"""
 # CODING-STYLE CHECKS:
 # pycodestyle test_consumption.py
+# pylint --disable=locally-disabled test_consumption.py
 
+import copy
 import numpy as np
-import paramtools
 import pytest
-import copy
+import paramtools
 from taxcalc import Policy, Records, Calculator, Consumption
 
 
-def test_year_consistency():
+def test_start_year_consistency():
+    """Test docstring"""
     assert Consumption.JSON_START_YEAR == Policy.JSON_START_YEAR
-    assert Consumption.DEFAULT_NUM_YEARS == Policy.DEFAULT_NUM_YEARS
 
 
 def test_validity_of_consumption_vars_set():
+    """Test docstring"""
     records_varinfo = Records(data=None)
     assert Consumption.RESPONSE_VARS.issubset(records_varinfo.USABLE_READ_VARS)
     useable_vars = set(['housing', 'snap', 'tanf', 'vet', 'wic',
@@ -22,6 +27,7 @@ def test_validity_of_consumption_vars_set():
 
 
 def test_update_consumption():
+    """Test docstring"""
     consump = Consumption()
     consump.update_consumption({})
     revision = {
@@ -31,24 +37,32 @@ def test_update_consumption():
                             2015: 0.80}
     }
     consump.update_consumption(revision)
-    expected_mpc_e20400 = np.full((Consumption.DEFAULT_NUM_YEARS,), 0.06)
+    expected_mpc_e20400 = np.full((consump.num_years,), 0.06)
     expected_mpc_e20400[0] = 0.0
     expected_mpc_e20400[1] = 0.05
-    assert np.allclose(consump._MPC_e20400,
-                       expected_mpc_e20400,
-                       rtol=0.0)
-    assert np.allclose(consump._MPC_e17500,
-                       np.zeros((Consumption.DEFAULT_NUM_YEARS,)),
-                       rtol=0.0)
-    expected_ben_mcare_value = np.full((Consumption.DEFAULT_NUM_YEARS,), 0.80)
+    assert np.allclose(
+        consump._MPC_e20400,  # pylint: disable=protected-access
+        expected_mpc_e20400,
+        rtol=0.0
+    )
+    assert np.allclose(
+        consump._MPC_e17500,  # pylint: disable=protected-access
+        np.zeros((consump.num_years,)),
+        rtol=0.0
+    )
+    expected_ben_mcare_value = np.full((consump.num_years,), 0.80)
     expected_ben_mcare_value[0] = 1.0
     expected_ben_mcare_value[1] = 0.75
-    assert np.allclose(consump._BEN_mcare_value,
-                       expected_ben_mcare_value,
-                       rtol=0.0)
-    assert np.allclose(consump._BEN_snap_value,
-                       np.ones((Consumption.DEFAULT_NUM_YEARS,)),
-                       rtol=0.0)
+    assert np.allclose(
+        consump._BEN_mcare_value,  # pylint: disable=protected-access
+        expected_ben_mcare_value,
+        rtol=0.0
+    )
+    assert np.allclose(
+        consump._BEN_snap_value,  # pylint: disable=protected-access
+        np.ones((consump.num_years,)),
+        rtol=0.0
+    )
     consump.set_year(2015)
     assert consump.current_year == 2015
     assert consump.MPC_e20400 == 0.06
@@ -58,6 +72,7 @@ def test_update_consumption():
 
 
 def test_incorrect_update_consumption():
+    """Test docstring"""
     with pytest.raises(paramtools.ValidationError):
         Consumption().update_consumption([])
     with pytest.raises(paramtools.ValidationError):
@@ -75,6 +90,7 @@ def test_incorrect_update_consumption():
 
 
 def test_future_update_consumption():
+    """Test docstring"""
     consump = Consumption()
     assert consump.current_year == consump.start_year
     assert consump.has_response() is False
@@ -98,6 +114,7 @@ def test_future_update_consumption():
 
 
 def test_consumption_default_data():
+    """Test docstring"""
     consump = Consumption()
     pdata = consump.specification(meta_data=True, ignore_state=True)
     for pname in pdata.keys():
@@ -108,24 +125,26 @@ def test_consumption_default_data():
 
 
 def test_consumption_response(cps_subsample):
+    """Test docstring"""
+    # pylint: disable=too-many-locals
     consump = Consumption()
     mpc = 0.5
     consumption_response = {'MPC_e20400': {2013: mpc}}
     consump.update_consumption(consumption_response)
     # test incorrect call to response method
     with pytest.raises(ValueError):
-        consump.response(list(), 1)
+        consump.response([], 1)
     # test correct call to response method
     rec = Records.cps_constructor(data=cps_subsample)
-    pre = copy.deepcopy(rec.e20400)
+    pre = copy.deepcopy(getattr(rec, 'e20400'))
     consump.response(rec, 1.0)
-    post = rec.e20400
+    post = getattr(rec, 'e20400')
     actual_diff = post - pre
     expected_diff = np.ones(rec.array_length) * mpc
     assert np.allclose(actual_diff, expected_diff)
     # compute earnings mtr with no consumption response
     rec = Records.cps_constructor(data=cps_subsample)
-    ided0 = copy.deepcopy(rec.e20400)
+    ided0 = copy.deepcopy(getattr(rec, 'e20400'))
     calc0 = Calculator(policy=Policy(), records=rec, consumption=None)
     (mtr0_ptax, mtr0_itax, _) = calc0.mtr(variable_str='e00200p',
                                           wrt_full_compensation=False)

taxcalc/tests/test_cpscsv.py

@@ -17,8 +17,11 @@ import json
 import pytest
 import numpy as np
 import pandas as pd
-# pylint: disable=import-error
-from taxcalc import Policy, Records, Calculator
+from taxcalc.growfactors import GrowFactors
+from taxcalc.growdiff import GrowDiff
+from taxcalc.policy import Policy
+from taxcalc.records import Records
+from taxcalc.calculator import Calculator
 
 
 START_YEAR = 2017
@@ -52,7 +55,7 @@ def test_agg(tests_path, cps_fullsample):
         if not np.allclose(adt[icol], edt[str(icol)]):
             diffs = True
     if diffs:
-        new_filename = '{}{}'.format(aggres_path[:-10], 'actual.csv')
+        new_filename = f'{aggres_path[:-10]}actual.csv'
         adt.to_csv(new_filename, float_format='%.1f')
         msg = 'CPSCSV AGG RESULTS DIFFER\n'
         msg += '-------------------------------------------------\n'
@@ -65,7 +68,7 @@ def test_agg(tests_path, cps_fullsample):
         raise ValueError(msg)
     # create aggregate diagnostic table using unweighted sub-sample of records
     rn_seed = 180  # to ensure sub-sample is always the same
-    subfrac = 0.07# 0.03  # sub-sample fraction
+    subfrac = 0.07  # sub-sample fraction
     subsample = cps_fullsample.sample(frac=subfrac, random_state=rn_seed)
     recs_subsample = Records.cps_constructor(data=subsample)
     calc_subsample = Calculator(policy=baseline_policy, records=recs_subsample)
@@ -84,14 +87,17 @@ def test_agg(tests_path, cps_fullsample):
         if not np.allclose(taxes_subsample[cyr], taxes_fullsample[cyr],
                            atol=0.0, rtol=reltol):
             reldiff = (taxes_subsample[cyr] / taxes_fullsample[cyr]) - 1.
-            line1 = '\nCPSCSV AGG SUB-vs-FULL RESULTS DIFFER IN {}'
-            line2 = '\n  when subfrac={:.3f}, rtol={:.4f}, seed={}'
-            line3 = '\n  with sub={:.3f}, full={:.3f}, rdiff={:.4f}'
-            msg += line1.format(cyr)
-            msg += line2.format(subfrac, reltol, rn_seed)
-            msg += line3.format(taxes_subsample[cyr],
-                                taxes_fullsample[cyr],
-                                reldiff)
+            line1 = f'\nCPSCSV AGG SUB-vs-FULL RESULTS DIFFER IN {cyr}'
+            line2 = (
+                f'\n  when subfrac={subfrac:.3f}, rtol={reltol:.4f}, '
+                f'seed={rn_seed}'
+            )
+            line3 = (
+                f'\n  with sub={taxes_subsample[cyr]:.3f}, '
+                f'full={taxes_fullsample[cyr]:.3f}, '
+                f'rdiff={reldiff:.4f}'
+            )
+            msg += line1 + line2 + line3
     if msg:
         raise ValueError(msg)
 
@@ -104,7 +110,7 @@ def test_cps_availability(tests_path, cps_path):
     cpsvars = set(list(cpsdf))
     # make set of variable names that are marked as cps.csv available
     rvpath = os.path.join(tests_path, '..', 'records_variables.json')
-    with open(rvpath, 'r') as rvfile:
+    with open(rvpath, 'r', encoding='utf-8') as rvfile:
         rvdict = json.load(rvfile)
     recvars = set()
     for vname, vdict in rvdict['read'].items():
@@ -142,22 +148,66 @@ def nonsmall_diffs(linelist1, linelist2, small=0.0):
     for line1, line2 in zip(linelist1, linelist2):
         if line1 == line2:
             continue
-        else:
-            tokens1 = line1.replace(',', '').split()
-            tokens2 = line2.replace(',', '').split()
-            for tok1, tok2 in zip(tokens1, tokens2):
-                tok1_isfloat = isfloat(tok1)
-                tok2_isfloat = isfloat(tok2)
-                if tok1_isfloat and tok2_isfloat:
-                    if abs(float(tok1) - float(tok2)) <= smallamt:
-                        continue
-                    else:
-                        return True
-                elif not tok1_isfloat and not tok2_isfloat:
-                    if tok1 == tok2:
-                        continue
-                    else:
-                        return True
-                else:
-                    return True
+        tokens1 = line1.replace(',', '').split()
+        tokens2 = line2.replace(',', '').split()
+        for tok1, tok2 in zip(tokens1, tokens2):
+            tok1_isfloat = isfloat(tok1)
+            tok2_isfloat = isfloat(tok2)
+            if tok1_isfloat and tok2_isfloat:
+                if abs(float(tok1) - float(tok2)) <= smallamt:
+                    continue
+                return True
+            if not tok1_isfloat and not tok2_isfloat:
+                if tok1 == tok2:
+                    continue
+                return True
+            return True
         return False
+
+
+def test_flexible_last_budget_year(cps_fullsample):
+    """
+    Test flexible LAST_BUDGET_YEAR logic using cps.csv file.
+    """
+    tax_calc_year = Policy.LAST_BUDGET_YEAR - 1
+    growdiff_year = tax_calc_year - 1
+    growdiff_dict = {'AWAGE': {growdiff_year: 0.01, tax_calc_year: 0.0}}
+
+    def default_calculator(growdiff_dictionary):
+        """
+        Return CPS-based Calculator object using default LAST_BUDGET_YEAR.
+        """
+        g_factors = GrowFactors()
+        gdiff = GrowDiff()
+        gdiff.update_growdiff(growdiff_dictionary)
+        gdiff.apply_to(g_factors)
+        pol = Policy(gfactors=g_factors)
+        rec = Records.cps_constructor(data=cps_fullsample, gfactors=g_factors)
+        calc = Calculator(policy=pol, records=rec)
+        return calc
+
+    def flexible_calculator(growdiff_dictionary, last_b_year):
+        """
+        Return CPS-based Calculator object using custom LAST_BUDGET_YEAR.
+        """
+        g_factors = GrowFactors()
+        gdiff = GrowDiff(last_budget_year=last_b_year)
+        gdiff.update_growdiff(growdiff_dictionary)
+        gdiff.apply_to(g_factors)
+        pol = Policy(gfactors=g_factors, last_budget_year=last_b_year)
+        rec = Records.cps_constructor(data=cps_fullsample, gfactors=g_factors)
+        calc = Calculator(policy=pol, records=rec)
+        return calc
+
+    # begin main test logic
+    cdef = default_calculator(growdiff_dict)
+    cdef.advance_to_year(tax_calc_year)
+    cdef.calc_all()
+    iitax_def = round(cdef.weighted_total('iitax'))
+
+    cflx = flexible_calculator(growdiff_dict, tax_calc_year)
+    cflx.advance_to_year(tax_calc_year)
+    cflx.calc_all()
+    iitax_flx = round(cflx.weighted_total('iitax'))
+
+    assert np.allclose([iitax_flx], [iitax_def])

taxcalc/tests/test_data.py

@@ -1,5 +1,9 @@
+"""
+Test Data class and its methods.
+"""
 # CODING-STYLE CHECKS:
 # pycodestyle test_data.py
+# pylint --disable=locally-disabled test_data.py
 
 import os
 import tempfile
@@ -75,14 +79,17 @@ def test_recs_class(recs_varinfo_file, cps_subsample):
             super().__init__(data, start_year, gfactors, weights)
 
         def _extrapolate(self, year):
-            self.e00300 *= self.gfactors.factor_value('AINTS', year)
+            val = getattr(self, 'e00300')
+            setattr(
+                self, 'e00300', val * self.gfactors.factor_value('AINTS', year)
+            )
 
-    # test Recs class for incorrect instantiation
+    # test Recs class for incorrect instantiation:
     with pytest.raises(ValueError):
-        Recs(data=list(), start_year=2000,
+        Recs(data=[], start_year=2000,
              gfactors=None, weights=None)
     with pytest.raises(ValueError):
-        Recs(data=cps_subsample, start_year=list(),
+        Recs(data=cps_subsample, start_year=[],
              gfactors=None, weights=None)
     with pytest.raises(ValueError):
         Recs(data=cps_subsample, start_year=2000,
@@ -93,19 +100,18 @@ def test_recs_class(recs_varinfo_file, cps_subsample):
     with pytest.raises(ValueError):
         Recs(data=cps_subsample, start_year=2000,
              gfactors='', weights='')
-    # test Recs class for correct instantiation with no aging of data
+    # test Recs class for correct instantiation with no aging of data:
     syr = 2014
     rec = Recs(data=cps_subsample, start_year=syr,
                gfactors=None, weights=None)
-    assert isinstance(rec, Recs)
-    assert np.all(rec.MARS != 0)
-    assert rec.data_year == syr
-    assert rec.current_year == syr
-    sum_e00300_in_syr = rec.e00300.sum()
+    assert np.all(getattr(rec, 'MARS') != 0)
+    assert getattr(rec, 'data_year') == syr
+    assert getattr(rec, 'current_year') == syr
+    sum_e00300_in_syr = getattr(rec, 'e00300').sum()
     rec.increment_year()
-    assert rec.data_year == syr
-    assert rec.current_year == syr + 1
-    sum_e00300_in_syr_plus_one = rec.e00300.sum()
+    assert getattr(rec, 'data_year') == syr
+    assert getattr(rec, 'current_year') == syr + 1
+    sum_e00300_in_syr_plus_one = getattr(rec, 'e00300').sum()
     assert np.allclose([sum_e00300_in_syr], [sum_e00300_in_syr_plus_one])
     del rec
     # test Recs class for correct instantiation with aging of data
@@ -114,20 +120,21 @@ def test_recs_class(recs_varinfo_file, cps_subsample):
     rec = Recs(data=cps_subsample, start_year=syr,
                gfactors=GrowFactors(), weights=wghts_df)
     assert isinstance(rec, Recs)
-    assert np.all(rec.MARS != 0)
-    assert rec.data_year == syr
-    assert rec.current_year == syr
-    sum_s006_in_syr = rec.s006.sum()
-    sum_e00300_in_syr = rec.e00300.sum()
+    assert np.all(getattr(rec, 'MARS') != 0)
+    assert getattr(rec, 'data_year') == syr
+    assert getattr(rec, 'current_year') == syr
+    sum_s006_in_syr = getattr(rec, 's006').sum()
+    sum_e00300_in_syr = getattr(rec, 'e00300').sum()
     rec.increment_year()
-    assert rec.data_year == syr
-    assert rec.current_year == syr + 1
-    sum_s006_in_syr_plus_one = rec.s006.sum()
+    assert getattr(rec, 'data_year') == syr
+    assert getattr(rec, 'current_year') == syr + 1
+    sum_s006_in_syr_plus_one = getattr(rec, 's006').sum()
     assert sum_s006_in_syr_plus_one > sum_s006_in_syr
-    sum_e00300_in_syr_plus_one = rec.e00300.sum()
+    sum_e00300_in_syr_plus_one = getattr(rec, 'e00300').sum()
     assert sum_e00300_in_syr_plus_one > sum_e00300_in_syr
-    # test private methods
+    # test private methods:
+    # pylint: disable=protected-access
     rec._read_data(data=None)
     rec._read_weights(weights=None)
     with pytest.raises(ValueError):
-        rec._read_weights(weights=list())
+        rec._read_weights(weights=[])