def testDiet(self):
def doTheTests(tdf) :
ticDat = tdf.freeze_me(tdf.TicDat(**{t:getattr(dietData(),t) for t in tdf.primary_key_fields}))
filePath = makeCleanPath(os.path.join(_scratchDir, "diet.db"))
tdf.sql.write_db_data(ticDat, filePath)
sqlTicDat = tdf.sql.create_tic_dat(filePath)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
def changeit() :
sqlTicDat.categories["calories"]["minNutrition"]=12
changeit()
self.assertFalse(tdf._same_data(ticDat, sqlTicDat))
self.assertTrue(self.firesException(lambda : tdf.sql.write_db_data(ticDat, filePath)))
tdf.sql.write_db_data(ticDat, filePath, allow_overwrite=True)
sqlTicDat = tdf.sql.create_tic_dat(filePath, freeze_it=True)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
self.assertTrue(self.firesException(changeit))
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
filePath = makeCleanPath(os.path.join(_scratchDir, "diet.sql"))
tdf.sql.write_sql_file(ticDat, filePath)
sqlTicDat = tdf.sql.create_tic_dat_from_sql(filePath)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
changeit()
self.assertFalse(tdf._same_data(ticDat, sqlTicDat))
tdf.sql.write_sql_file(ticDat, filePath, include_schema=True)
sqlTicDat = tdf.sql.create_tic_dat_from_sql(filePath, includes_schema=True, freeze_it=True)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
self.assertTrue(self.firesException(changeit))
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
doTheTests(TicDatFactory(**dietSchema()))
tdf = TicDatFactory(**dietSchema())
self.assertFalse(tdf.foreign_keys)
tdf.set_default_values(categories = {'maxNutrition': float("inf"), 'minNutrition': 0.0},
foods = {'cost': 0.0},
nutritionQuantities = {'qty': 0.0})
addDietForeignKeys(tdf)
ordered = tdf.sql._ordered_tables()
self.assertTrue(ordered.index("categories") < ordered.index("nutritionQuantities"))
self.assertTrue(ordered.index("foods") < ordered.index("nutritionQuantities"))
ticDat = tdf.TicDat(**{t:getattr(dietData(),t) for t in tdf.primary_key_fields})
origTicDat = tdf.copy_tic_dat(ticDat)
self.assertTrue(tdf._same_data(ticDat, origTicDat))
self.assertFalse(tdf.find_foreign_key_failures(ticDat))
ticDat.nutritionQuantities['hot dog', 'boger'] = ticDat.nutritionQuantities['junk', 'protein'] = -12
self.assertTrue(tdf.find_foreign_key_failures(ticDat) ==
{('nutritionQuantities', 'foods', ('food', 'name'), 'many-to-one'): (('junk',), (('junk', 'protein'),)),
('nutritionQuantities', 'categories', ('category', 'name'), 'many-to-one'):
(('boger',), (('hot dog', 'boger'),))})
self.assertFalse(tdf._same_data(ticDat, origTicDat))
tdf.remove_foreign_keys_failures(ticDat)
self.assertFalse(tdf.find_foreign_key_failures(ticDat))
self.assertTrue(tdf._same_data(ticDat, origTicDat))
doTheTests(tdf)
def testDietCleaningFive(self):
tdf = TicDatFactory(**dietSchema())
tdf.add_data_row_predicate("categories",
lambda row: row["maxNutrition"] >= 66)
tdf.set_data_type("categories",
"minNutrition",
max=0,
inclusive_max=True)
addDietForeignKeys(tdf)
ticDat = tdf.copy_tic_dat(dietData())
input_set = create_inputset_mock(tdf, ticDat)
self.assertTrue(
tdf._same_data(
tdf.opalytics.create_tic_dat(input_set, raw_data=True),
ticDat))
ticDatPurged = tdf.opalytics.create_tic_dat(input_set, raw_data=False)
self.assertFalse(tdf._same_data(ticDatPurged, ticDat))
ticDat.categories.pop("fat")
ticDat.categories.pop("calories")
ticDat.categories.pop("protein")
self.assertFalse(tdf._same_data(ticDatPurged, ticDat))
tdf.remove_foreign_keys_failures(ticDat)
self.assertTrue(tdf._same_data(ticDatPurged, ticDat))
def testDietCleaning(self):
sch = dietSchema()
sch["categories"][-1].append("_active")
tdf1 = TicDatFactory(**dietSchema())
tdf2 = TicDatFactory(**sch)
ticDat2 = tdf2.copy_tic_dat(dietData())
for v in ticDat2.categories.values():
v["_active"] = True
ticDat2.categories["fat"]["_active"] = False
ticDat1 = tdf1.copy_tic_dat(dietData())
input_set = create_inputset_mock_with_active_hack(tdf2, ticDat2)
self.assertTrue(
tdf1._same_data(
tdf1.opalytics.create_tic_dat(input_set, raw_data=True),
ticDat1))
ticDatPurged = tdf1.opalytics.create_tic_dat(input_set)
self.assertFalse(tdf1._same_data(ticDatPurged, ticDat1))
ticDat1.categories.pop("fat")
tdf1.remove_foreign_keys_failures(ticDat1)
self.assertTrue(tdf1._same_data(ticDatPurged, ticDat1))
def testDietCleaningFour(self):
tdf = TicDatFactory(**dietSchema())
addDietForeignKeys(tdf)
ticDat = tdf.copy_tic_dat(dietData())
ticDat.categories.pop("fat")
input_set = create_inputset_mock(tdf, ticDat)
self.assertTrue(
tdf._same_data(
tdf.opalytics.create_tic_dat(input_set, raw_data=True),
ticDat))
ticDatPurged = tdf.opalytics.create_tic_dat(input_set, raw_data=False)
self.assertFalse(tdf._same_data(ticDatPurged, ticDat))
tdf.remove_foreign_keys_failures(ticDat)
self.assertTrue(tdf._same_data(ticDatPurged, ticDat))
def testSix(self):
tdf = TicDatFactory(plants = [["name"], ["stuff", "otherstuff"]],
lines = [["name"], ["plant", "weird stuff"]],
line_descriptor = [["name"], ["booger"]],
products = [["name"],["gover"]],
production = [["line", "product"], ["min", "max"]],
pureTestingTable = [[], ["line", "plant", "product", "something"]],
extraProduction = [["line", "product"], ["extramin", "extramax"]],
weirdProduction = [["line1", "line2", "product"], ["weirdmin", "weirdmax"]])
tdf.add_foreign_key("production", "lines", ("line", "name"))
tdf.add_foreign_key("production", "products", ("product", "name"))
tdf.add_foreign_key("lines", "plants", ("plant", "name"))
tdf.add_foreign_key("line_descriptor", "lines", ("name", "name"))
for f in set(tdf.data_fields["pureTestingTable"]).difference({"something"}):
tdf.add_foreign_key("pureTestingTable", "%ss"%f, (f,"name"))
tdf.add_foreign_key("extraProduction", "production", (("line", "line"), ("product","product")))
tdf.add_foreign_key("weirdProduction", "production", (("line1", "line"), ("product","product")))
tdf.add_foreign_key("weirdProduction", "extraProduction", (("line2","line"), ("product","product")))
goodDat = tdf.TicDat()
goodDat.plants["Cleveland"] = ["this", "that"]
goodDat.plants["Newark"]["otherstuff"] =1
goodDat.products["widgets"] = goodDat.products["gadgets"] = "shizzle"
for i,p in enumerate(goodDat.plants):
goodDat.lines[i]["plant"] = p
for i,(pl, pd) in enumerate(itertools.product(goodDat.lines, goodDat.products)):
goodDat.production[pl, pd] = {"min":1, "max":10+i}
badDat1 = tdf.copy_tic_dat(goodDat)
badDat1.production["notaline", "widgets"] = [0,1]
badDat2 = tdf.copy_tic_dat(badDat1)
fk, fkm = _ForeignKey, _ForeignKeyMapping
self.assertTrue(tdf.find_foreign_key_failures(badDat1) == tdf.find_foreign_key_failures(badDat2) ==
{fk('production', 'lines', fkm('line', 'name'), 'many-to-one'):
(('notaline',), (('notaline', 'widgets'),))})
badDat1.lines["notaline"]["plant"] = badDat2.lines["notaline"]["plant"] = "notnewark"
self.assertTrue(tdf.find_foreign_key_failures(badDat1) == tdf.find_foreign_key_failures(badDat2) ==
{fk('lines', 'plants', fkm('plant', 'name'), 'many-to-one'):
(('notnewark',), ('notaline',))})
tdf.remove_foreign_keys_failures(badDat1, propagate=False)
tdf.remove_foreign_keys_failures(badDat2, propagate=True)
self.assertTrue(tdf._same_data(badDat2, goodDat) and not tdf.find_foreign_key_failures(badDat2))
self.assertTrue(tdf.find_foreign_key_failures(badDat1) ==
{fk('production', 'lines', fkm('line', 'name'), 'many-to-one'):
(('notaline',), (('notaline', 'widgets'),))})
tdf.remove_foreign_keys_failures(badDat1, propagate=False)
self.assertTrue(tdf._same_data(badDat1, goodDat) and not tdf.find_foreign_key_failures(badDat1))
_ = len(goodDat.lines)
for i,p in enumerate(goodDat.plants.keys() + goodDat.plants.keys()):
goodDat.lines[i+_]["plant"] = p
for l in goodDat.lines:
if i%2:
goodDat.line_descriptor[l] = i+10
for i,(l,pl,pdct) in enumerate(sorted(itertools.product(goodDat.lines, goodDat.plants, goodDat.products))):
goodDat.pureTestingTable.append((l,pl,pdct,i))
self.assertFalse(tdf.find_foreign_key_failures(goodDat))
badDat = tdf.copy_tic_dat(goodDat)
badDat.pureTestingTable.append(("j", "u", "nk", "ay"))
l = len(goodDat.pureTestingTable)
self.assertTrue(tdf.find_foreign_key_failures(badDat) ==
{fk('pureTestingTable', 'plants', fkm('plant', 'name'), 'many-to-one'): (('u',),(l,)),
fk('pureTestingTable', 'products', fkm('product', 'name'), 'many-to-one'): (('nk',), (l,)),
fk('pureTestingTable', 'lines', fkm('line', 'name'), 'many-to-one'): (('j',), (l,))})
obfudat = tdf.obfusimplify(goodDat, freeze_it=True)
self.assertTrue(all(len(getattr(obfudat.copy, t)) == len(getattr(goodDat, t))
for t in tdf.all_tables))
for n in goodDat.plants.keys() + goodDat.lines.keys() + goodDat.products.keys() :
self.assertTrue(n in {_[1] for _ in obfudat.renamings.values()})
self.assertFalse(n in obfudat.renamings)
self.assertTrue(obfudat.copy.plants['P2']['otherstuff'] == 1)
self.assertFalse(tdf._same_data(obfudat.copy, goodDat))
for k,r in obfudat.copy.line_descriptor.items():
i = r.values()[0] - 10
self.assertTrue(i%2 and (goodDat.line_descriptor[i].values()[0] == i+10))
obfudat2 = tdf.obfusimplify(goodDat, {"plants": "P", "lines" : "L", "products" :"PR"})
self.assertTrue(tdf._same_data(obfudat.copy, obfudat2.copy))
obfudat3 = tdf.obfusimplify(goodDat, skip_tables=["plants", "lines", "products"])
self.assertTrue(tdf._same_data(obfudat3.copy, goodDat))
obfudat4 = tdf.obfusimplify(goodDat, skip_tables=["lines", "products"])
self.assertFalse(tdf._same_data(obfudat4.copy, goodDat))
self.assertFalse(tdf._same_data(obfudat4.copy, obfudat.copy))
def testDiet(self):
if not self.can_run:
return
def doTheTests(tdf):
ticDat = tdf.freeze_me(
tdf.TicDat(**{
t: getattr(dietData(), t)
for t in tdf.primary_key_fields
}))
filePath = makeCleanPath(os.path.join(_scratchDir, "diet.db"))
tdf.sql.write_db_data(ticDat, filePath)
self.assertFalse(tdf.sql.find_duplicates(filePath))
sqlTicDat = tdf.sql.create_tic_dat(filePath)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
def changeit():
sqlTicDat.categories["calories"]["minNutrition"] = 12
changeit()
self.assertFalse(tdf._same_data(ticDat, sqlTicDat))
self.assertTrue(
self.firesException(
lambda: tdf.sql.write_db_data(ticDat, filePath)))
tdf.sql.write_db_data(ticDat, filePath, allow_overwrite=True)
sqlTicDat = tdf.sql.create_tic_dat(filePath, freeze_it=True)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
self.assertTrue(self.firesException(changeit))
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
filePath = makeCleanPath(os.path.join(_scratchDir, "diet.sql"))
tdf.sql.write_sql_file(ticDat, filePath)
sqlTicDat = tdf.sql.create_tic_dat_from_sql(filePath)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
changeit()
self.assertFalse(tdf._same_data(ticDat, sqlTicDat))
tdf.sql.write_sql_file(ticDat,
filePath,
include_schema=True,
allow_overwrite=True)
sqlTicDat = tdf.sql.create_tic_dat_from_sql(filePath,
includes_schema=True,
freeze_it=True)
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
self.assertTrue(self.firesException(changeit))
self.assertTrue(tdf._same_data(ticDat, sqlTicDat))
doTheTests(TicDatFactory(**dietSchema()))
tdf = TicDatFactory(**dietSchema())
self.assertFalse(tdf.foreign_keys)
tdf.set_default_values(categories={
'maxNutrition': float("inf"),
'minNutrition': 0.0
},
foods={'cost': 0.0},
nutritionQuantities={'qty': 0.0})
addDietForeignKeys(tdf)
ordered = tdf.sql._ordered_tables()
self.assertTrue(
ordered.index("categories") < ordered.index("nutritionQuantities"))
self.assertTrue(
ordered.index("foods") < ordered.index("nutritionQuantities"))
ticDat = tdf.TicDat(
**{t: getattr(dietData(), t)
for t in tdf.primary_key_fields})
self._test_generic_copy(ticDat, tdf)
self._test_generic_copy(ticDat, tdf, ["nutritionQuantities"])
origTicDat = tdf.copy_tic_dat(ticDat)
self.assertTrue(tdf._same_data(ticDat, origTicDat))
self.assertFalse(tdf.find_foreign_key_failures(ticDat))
ticDat.nutritionQuantities[
'hot dog', 'boger'] = ticDat.nutritionQuantities['junk',
'protein'] = -12
self.assertTrue(
tdf.find_foreign_key_failures(ticDat) == {
('nutritionQuantities', 'foods', ('food', 'name'), 'many-to-one'):
(('junk', ), (('junk', 'protein'), )),
('nutritionQuantities', 'categories', ('category', 'name'), 'many-to-one'):
(('boger', ), (('hot dog', 'boger'), ))
})
self.assertFalse(tdf._same_data(ticDat, origTicDat))
tdf.remove_foreign_keys_failures(ticDat)
self.assertFalse(tdf.find_foreign_key_failures(ticDat))
self.assertTrue(tdf._same_data(ticDat, origTicDat))
doTheTests(tdf)
