def test_get_nonterminal_productions_all_column_types(self):
# This test is long, but worth it. These are all of the valid actions in the grammar, and
# we want to be sure they are what we expect.
productions = self.language.get_nonterminal_productions()
assert set(productions.keys()) == {
"@start@",
"<List[Row],StringColumn:List[str]>",
"<List[Row],DateColumn:Date>",
"<List[Row],NumberColumn,Number:List[Row]>",
"<List[Row],ComparableColumn:List[Row]>",
"<List[Row],Column:List[Row]>",
"<List[Row],List[Row],NumberColumn:Number>",
"<List[Row],StringColumn,List[str]:List[Row]>",
"<Number,Number,Number:Date>",
"<List[Row],DateColumn,Date:List[Row]>",
"<List[Row],NumberColumn:Number>",
"<List[Row]:List[Row]>",
'<List[Row],StringColumn:List[str]>',
"<List[Row]:Number>",
"List[str]",
"List[Row]",
"Date",
"Number",
"StringColumn",
"NumberColumn",
"ComparableColumn",
"Column",
"DateColumn",
"List[str]",
}
check_productions_match(productions['@start@'],
['Date', 'Number', 'List[str]'])
check_productions_match(
productions['<List[Row],StringColumn:List[str]>'],
['select_string', 'mode_string'])
check_productions_match(
productions['<List[Row],DateColumn:Date>'],
['select_date', 'max_date', 'min_date', 'mode_date'])
check_productions_match(
productions['<List[Row],NumberColumn,Number:List[Row]>'], [
'filter_number_equals', 'filter_number_greater',
'filter_number_greater_equals', 'filter_number_lesser',
'filter_number_lesser_equals', 'filter_number_not_equals'
])
check_productions_match(
productions['<List[Row],ComparableColumn:List[Row]>'],
['argmax', 'argmin'])
check_productions_match(productions['<List[Row],Column:List[Row]>'],
['same_as'])
check_productions_match(
productions['<List[Row],List[Row],NumberColumn:Number>'], ['diff'])
check_productions_match(
productions['<List[Row],StringColumn,List[str]:List[Row]>'],
['filter_in', 'filter_not_in'])
check_productions_match(productions['<Number,Number,Number:Date>'],
['date'])
check_productions_match(
productions['<List[Row],DateColumn,Date:List[Row]>'], [
'filter_date_equals', 'filter_date_greater',
'filter_date_greater_equals', 'filter_date_lesser',
'filter_date_lesser_equals', 'filter_date_not_equals'
])
check_productions_match(productions['<List[Row],NumberColumn:Number>'],
[
'average', 'max_number', 'min_number',
'sum', 'select_number', 'mode_number'
])
check_productions_match(productions['<List[Row]:List[Row]>'],
['first', 'last', 'next', 'previous'])
check_productions_match(productions['<List[Row]:Number>'], ['count'])
check_productions_match(
productions['List[Row]'],
[
'all_rows',
'[<List[Row],Column:List[Row]>, List[Row], Column]',
'[<List[Row],DateColumn,Date:List[Row]>, List[Row], DateColumn, Date]',
'[<List[Row],ComparableColumn:List[Row]>, List[Row], ComparableColumn]',
'[<List[Row],NumberColumn,Number:List[Row]>, List[Row], NumberColumn, Number]',
'[<List[Row],StringColumn,List[str]:List[Row]>, List[Row], StringColumn, List[str]]', # pylint: disable=line-too-long
'[<List[Row]:List[Row]>, List[Row]]'
])
check_productions_match(productions['Date'], [
'[<Number,Number,Number:Date>, Number, Number, Number]',
'[<List[Row],DateColumn:Date>, List[Row], DateColumn]'
])
# Some of the number productions are instance-specific, and some of them are from the
# grammar.
check_productions_match(productions['Number'], [
'2001', '2002', '2005', '2010', '2013', '-1', '1', '2', '23',
'8000',
'[<List[Row],NumberColumn:Number>, List[Row], NumberColumn]',
'[<List[Row],List[Row],NumberColumn:Number>, List[Row], List[Row], NumberColumn]',
'[<List[Row]:Number>, List[Row]]'
])
# These are the columns in table, and are instance specific.
check_productions_match(productions['StringColumn'], [
'string_column:league', 'string_column:playoffs',
'string_column:open_cup', 'string_column:year',
'string_column:division', 'string_column:avg_attendance',
'string_column:regular_season'
])
check_productions_match(productions['DateColumn'],
['date_column:year'])
check_productions_match(productions['NumberColumn'], [
'number_column:avg_attendance', 'number_column:open_cup',
'number_column:regular_season', 'number_column:division',
'number_column:year'
])
check_productions_match(productions['ComparableColumn'], [
'date_column:year', 'number_column:avg_attendance',
'number_column:open_cup', 'number_column:regular_season',
'number_column:division', 'number_column:year'
])
check_productions_match(productions['Column'], [
'string_column:league', 'string_column:playoffs',
'string_column:open_cup', 'string_column:year',
'string_column:division', 'string_column:avg_attendance',
'string_column:regular_season', 'date_column:year',
'number_column:avg_attendance', 'number_column:open_cup',
'number_column:regular_season', 'number_column:division',
'number_column:year'
])
# Strings come from the question - any span in the question that shows up as a cell in the
# table is a valid string production.
check_productions_match(productions['List[str]'], [
'string:quarterfinals', 'string:did_not_qualify',
'string:a_league', 'string:usl_first_division',
'string:usl_a_league', 'string:1', 'string:2', 'string:2005',
'string:2001',
'[<List[Row],StringColumn:List[str]>, List[Row], StringColumn]'
])
def test_get_nonterminal_productions(self):
productions = self.language.get_nonterminal_productions()
assert set(productions.keys()) == {
"QuaRelType",
"@start@",
"World",
"Direction",
"<QuaRelType,QuaRelType:QuaRelType>",
"int",
"<QuaRelType,QuaRelType,QuaRelType:int>",
"<Direction,World:QuaRelType>",
}
check_productions_match(
productions["QuaRelType"],
[
"[<QuaRelType,QuaRelType:QuaRelType>, QuaRelType, QuaRelType]",
"[<Direction,World:QuaRelType>, Direction, World]",
],
)
check_productions_match(productions["@start@"], ["int"])
check_productions_match(productions["World"], ["world1", "world2"])
check_productions_match(productions["Direction"],
["higher", "lower", "high", "low"])
check_productions_match(
productions["<QuaRelType,QuaRelType:QuaRelType>"], ["and"])
check_productions_match(
productions["int"],
[
"[<QuaRelType,QuaRelType,QuaRelType:int>, QuaRelType, QuaRelType, QuaRelType]"
],
)
check_productions_match(
productions["<QuaRelType,QuaRelType,QuaRelType:int>"], ["infer"])
check_productions_match(
productions["<Direction,World:QuaRelType>"],
[
"friction",
"speed",
"distance",
"heat",
"smoothness",
"acceleration",
"amountSweat",
"apparentSize",
"breakability",
"brightness",
"exerciseIntensity",
"flexibility",
"gravity",
"loudness",
"mass",
"strength",
"thickness",
"time",
"weight",
],
)
def test_get_nonterminal_productions(self):
productions = self.language.get_nonterminal_productions()
assert set(productions.keys()) == {
'QuaRelType', '@start@', 'World', 'Direction',
'<QuaRelType,QuaRelType:QuaRelType>', 'int',
'<QuaRelType,QuaRelType,QuaRelType:int>',
'<Direction,World:QuaRelType>'
}
check_productions_match(productions['QuaRelType'], [
'[<QuaRelType,QuaRelType:QuaRelType>, QuaRelType, QuaRelType]',
'[<Direction,World:QuaRelType>, Direction, World]'
])
check_productions_match(productions['@start@'], ['int'])
check_productions_match(productions['World'], ['world1', 'world2'])
check_productions_match(productions['Direction'],
['higher', 'lower', 'high', 'low'])
check_productions_match(
productions['<QuaRelType,QuaRelType:QuaRelType>'], ['and'])
check_productions_match(productions['int'], [
'[<QuaRelType,QuaRelType,QuaRelType:int>, QuaRelType, QuaRelType, QuaRelType]'
])
check_productions_match(
productions['<QuaRelType,QuaRelType,QuaRelType:int>'], ['infer'])
check_productions_match(productions['<Direction,World:QuaRelType>'], [
"friction", "speed", "distance", "heat", "smoothness",
"acceleration", "amountSweat", "apparentSize", "breakability",
"brightness", "exerciseIntensity", "flexibility", "gravity",
"loudness", "mass", "strength", "thickness", "time", "weight"
])
def test_get_nonterminal_productions_all_column_types(self):
# This test is long, but worth it. These are all of the valid actions in the grammar, and
# we want to be sure they are what we expect.
productions = self.language.get_nonterminal_productions()
assert set(productions.keys()) == {
"@start@",
"<List[Row],StringColumn:List[str]>",
"<List[Row],DateColumn:Date>",
"<List[Row],NumberColumn,Number:List[Row]>",
"<List[Row],ComparableColumn:List[Row]>",
"<List[Row],Column:List[Row]>",
"<List[Row],List[Row],NumberColumn:Number>",
"<List[Row],StringColumn,List[str]:List[Row]>",
"<Number,Number,Number:Date>",
"<List[Row],DateColumn,Date:List[Row]>",
"<List[Row],NumberColumn:Number>",
"<List[Row]:List[Row]>",
"<List[Row],StringColumn:List[str]>",
"<List[Row]:Number>",
"List[str]",
"List[Row]",
"Date",
"Number",
"StringColumn",
"NumberColumn",
"ComparableColumn",
"Column",
"DateColumn",
"List[str]",
}
check_productions_match(productions["@start@"], ["Date", "Number", "List[str]"])
check_productions_match(
productions["<List[Row],StringColumn:List[str]>"], ["select_string", "mode_string"]
)
check_productions_match(
productions["<List[Row],DateColumn:Date>"],
["select_date", "max_date", "min_date", "mode_date"],
)
check_productions_match(
productions["<List[Row],NumberColumn,Number:List[Row]>"],
[
"filter_number_equals",
"filter_number_greater",
"filter_number_greater_equals",
"filter_number_lesser",
"filter_number_lesser_equals",
"filter_number_not_equals",
],
)
check_productions_match(
productions["<List[Row],ComparableColumn:List[Row]>"], ["argmax", "argmin"]
)
check_productions_match(productions["<List[Row],Column:List[Row]>"], ["same_as"])
check_productions_match(productions["<List[Row],List[Row],NumberColumn:Number>"], ["diff"])
check_productions_match(
productions["<List[Row],StringColumn,List[str]:List[Row]>"],
["filter_in", "filter_not_in"],
)
check_productions_match(productions["<Number,Number,Number:Date>"], ["date"])
check_productions_match(
productions["<List[Row],DateColumn,Date:List[Row]>"],
[
"filter_date_equals",
"filter_date_greater",
"filter_date_greater_equals",
"filter_date_lesser",
"filter_date_lesser_equals",
"filter_date_not_equals",
],
)
check_productions_match(
productions["<List[Row],NumberColumn:Number>"],
["average", "max_number", "min_number", "sum", "select_number", "mode_number"],
)
check_productions_match(
productions["<List[Row]:List[Row]>"], ["first", "last", "next", "previous"]
)
check_productions_match(productions["<List[Row]:Number>"], ["count"])
check_productions_match(
productions["List[Row]"],
[
"all_rows",
"[<List[Row],Column:List[Row]>, List[Row], Column]",
"[<List[Row],DateColumn,Date:List[Row]>, List[Row], DateColumn, Date]",
"[<List[Row],ComparableColumn:List[Row]>, List[Row], ComparableColumn]",
"[<List[Row],NumberColumn,Number:List[Row]>, List[Row], NumberColumn, Number]",
"[<List[Row],StringColumn,List[str]:List[Row]>, List[Row], StringColumn, List[str]]",
"[<List[Row]:List[Row]>, List[Row]]",
],
)
check_productions_match(
productions["Date"],
[
"[<Number,Number,Number:Date>, Number, Number, Number]",
"[<List[Row],DateColumn:Date>, List[Row], DateColumn]",
],
)
# Some of the number productions are instance-specific, and some of them are from the
# grammar.
check_productions_match(
productions["Number"],
[
"2001",
"2002",
"2005",
"2010",
"2013",
"-1",
"1",
"2",
"23",
"8000",
"[<List[Row],NumberColumn:Number>, List[Row], NumberColumn]",
"[<List[Row],List[Row],NumberColumn:Number>, List[Row], List[Row], NumberColumn]",
"[<List[Row]:Number>, List[Row]]",
],
)
# These are the columns in table, and are instance specific.
check_productions_match(
productions["StringColumn"],
[
"string_column:league",
"string_column:playoffs",
"string_column:open_cup",
"string_column:year",
"string_column:division",
"string_column:avg_attendance",
"string_column:regular_season",
],
)
check_productions_match(productions["DateColumn"], ["date_column:year"])
check_productions_match(
productions["NumberColumn"],
[
"number_column:avg_attendance",
"number_column:open_cup",
"number_column:regular_season",
"number_column:division",
"number_column:year",
],
)
check_productions_match(
productions["ComparableColumn"],
[
"date_column:year",
"number_column:avg_attendance",
"number_column:open_cup",
"number_column:regular_season",
"number_column:division",
"number_column:year",
],
)
check_productions_match(
productions["Column"],
[
"string_column:league",
"string_column:playoffs",
"string_column:open_cup",
"string_column:year",
"string_column:division",
"string_column:avg_attendance",
"string_column:regular_season",
"date_column:year",
"number_column:avg_attendance",
"number_column:open_cup",
"number_column:regular_season",
"number_column:division",
"number_column:year",
],
)
# Strings come from the question - any span in the question that shows up as a cell in the
# table is a valid string production.
check_productions_match(
productions["List[str]"],
[
"string:quarterfinals",
"string:did_not_qualify",
"string:a_league",
"string:usl_first_division",
"string:usl_a_league",
"string:1",
"string:2",
"string:2005",
"string:2001",
"[<List[Row],StringColumn:List[str]>, List[Row], StringColumn]",
],
)