def test_remove_unnecessary_parens():
# mult contains a pattern containing a single conc containing a single mult?
# that can be reduced greatly
assert conc.parse("(a){2}b").reduce() == conc.parse("a{2}b")
assert conc.parse("(a?)+b").reduce() == conc.parse("a*b")
assert mult.parse("([ab])*").reduce() == mult.parse("[ab]*")
assert mult.parse("(c{1,2}){3,4}").reduce() == mult.parse("c{3,8}")
def test_mult_gen():
# One term
gen = mult.parse("[ab]").strings()
assert next(gen) == "a"
assert next(gen) == "b"
try:
next(gen)
assert False
except StopIteration:
assert True
# No terms
gen = mult.parse("[ab]{0}").strings()
assert next(gen) == ""
try:
next(gen)
assert False
except StopIteration:
assert True
# Many terms
gen = mult.parse("[ab]*").strings()
assert next(gen) == ""
assert next(gen) == "a"
assert next(gen) == "b"
assert next(gen) == "aa"
assert next(gen) == "ab"
assert next(gen) == "ba"
assert next(gen) == "bb"
assert next(gen) == "aaa"
def test_charclass_multiplication():
# a * 1 = a
assert charclass("a") * one == charclass("a")
# a * {1,3} = a{1,3}
assert charclass("a") * multiplier.parse("{1,3}") == mult.parse("a{1,3}")
# a * {4,} = a{4,}
assert charclass("a") * multiplier.parse("{4,}") == mult.parse("a{4,}")
def test_mult_gen():
# One term
gen = mult.parse("[ab]").strings()
assert next(gen) == "a"
assert next(gen) == "b"
try:
next(gen)
assert False
except StopIteration:
assert True
# No terms
gen = mult.parse("[ab]{0}").strings()
assert next(gen) == ""
try:
next(gen)
assert False
except StopIteration:
assert True
# Many terms
gen = mult.parse("[ab]*").strings()
assert next(gen) == ""
assert next(gen) == "a"
assert next(gen) == "b"
assert next(gen) == "aa"
assert next(gen) == "ab"
assert next(gen) == "ba"
assert next(gen) == "bb"
assert next(gen) == "aaa"
def test_remove_unnecessary_parens():
# mult contains a pattern containing a single conc containing a single mult?
# that can be reduced greatly
assert conc.parse("(a){2}b").reduce() == conc.parse("a{2}b")
assert conc.parse("(a?)+b").reduce() == conc.parse("a*b")
assert mult.parse("([ab])*").reduce() == mult.parse("[ab]*")
assert mult.parse("(c{1,2}){3,4}").reduce() == mult.parse("c{3,8}")
def test_remove_unnecessary_parens():
# mult contains a pattern containing a single conc containing a single mult?
# that can be reduced greatly
# e.g. "([ab])*" -> "[ab]*"
assert mult.parse("([ab])*").reduce() == mult.parse("[ab]*")
# e.g. "(c{1,2}){3,4}" -> "c{3,8}"
assert mult.parse("(c{1,2}){3,4}").reduce() == mult.parse("c{3,8}")
def test_mult_factor_out_qm():
# mult contains a pattern containing an empty conc? Pull the empty
# part out where it's external
assert mult.parse("(a|b*|)").reduce() == mult.parse("(a|b*)?")
assert conc.parse("(a|b*|)c").reduce() == conc.parse("(a|b*)?c")
# This happens even if emptystring is the only thing left inside the mult
assert mult.parse("()").reduce() == conc.parse("")
assert mult.parse("([$%\\^]|){1}").reduce() == mult.parse("[$%\\^]?")
def test_mult_factor_out_qm():
# mult contains a pattern containing an empty conc? Pull the empty
# part out where it's external
assert mult.parse("(a|b*|)").reduce() == mult.parse("(a|b*)?")
assert conc.parse("(a|b*|)c").reduce() == conc.parse("(a|b*)?c")
# This happens even if emptystring is the only thing left inside the mult
assert mult.parse("()").reduce() == conc.parse("")
assert mult.parse("([$%\\^]|){1}").reduce() == mult.parse("[$%\\^]?")
def test_main_bug():
assert parse("a*").reduce() == mult.parse("a*")
assert parse("a|a*").reduce() == mult.parse("a*")
assert parse("a{1,2}|a{3,4}|bc").reduce() == pattern.parse("a{1,4}|bc")
assert parse("a{1,2}|bc|a{3,4}").reduce() == pattern.parse("a{1,4}|bc")
assert parse("a{1,2}|a{3,4}|a{5,6}|bc").reduce() == pattern.parse("a{1,6}|bc")
assert parse("a{1,2}|a{3}|a{5,6}").reduce() == conc.parse("a{1,2}(a?|a{4})")
assert parse("a{1,2}|a{3}|a{5,6}|bc").reduce() == pattern.parse("a{1,3}|a{5,6}|bc")
assert parse("a{1,2}|a{4}|a{5,6}").reduce() == conc.parse("a{1,2}(a{3,4})?")
assert parse("a{1,2}|a{4}|a{5,6}|bc").reduce() == pattern.parse("a{1,2}|a{4,6}|bc")
assert (parse("a") | parse("a*")).reduce() == mult.parse("a*")
def test_main_bug():
assert parse("a*").reduce() == mult.parse("a*")
assert parse("a|a*").reduce() == mult.parse("a*")
assert parse("a{1,2}|a{3,4}|bc").reduce() == pattern.parse("a{1,4}|bc")
assert parse("a{1,2}|bc|a{3,4}").reduce() == pattern.parse("a{1,4}|bc")
assert parse("a{1,2}|a{3,4}|a{5,6}|bc").reduce() == pattern.parse("a{1,6}|bc")
assert parse("a{1,2}|a{3}|a{5,6}").reduce() == conc.parse("a{1,2}(a?|a{4})")
assert parse("a{1,2}|a{3}|a{5,6}|bc").reduce() == pattern.parse("a{1,3}|a{5,6}|bc")
assert parse("a{1,2}|a{4}|a{5,6}").reduce() == conc.parse("a{1,2}(a{3,4})?")
assert parse("a{1,2}|a{4}|a{5,6}|bc").reduce() == pattern.parse("a{1,2}|a{4,6}|bc")
assert (parse("a") | parse("a*")).reduce() == mult.parse("a*")
def test_cardinality():
assert charclass.parse("[]").cardinality() == 0
assert mult.parse("[]?").cardinality() == 1
assert mult.parse("[]{0,6}").cardinality() == 1
assert mult.parse("[ab]{3}").cardinality() == 8
assert mult.parse("[ab]{2,3}").cardinality() == 12
assert len(pattern.parse("abc|def(ghi|jkl)")) == 3
try:
len(pattern.parse(".*"))
assert False
except OverflowError:
assert True
def test_empty():
assert nothing.empty()
assert charclass().empty()
assert not dot.empty()
assert not mult.parse("a{0}").empty()
assert mult.parse("[]").empty()
assert not mult.parse("[]?").empty()
assert conc.parse("a[]").empty()
assert not conc.parse("a[]?").empty()
assert pattern().empty()
assert not pattern.parse("a{0}").empty()
assert not pattern.parse("[]?").empty()
def test_empty():
assert nothing.empty()
assert charclass().empty()
assert not dot.empty()
assert not mult.parse("a{0}").empty()
assert mult.parse("[]").empty()
assert not mult.parse("[]?").empty()
assert conc.parse("a[]").empty()
assert not conc.parse("a[]?").empty()
assert pattern().empty()
assert not pattern.parse("a{0}").empty()
assert not pattern.parse("[]?").empty()
def test_cardinality():
assert charclass.parse("[]").cardinality() == 0
assert mult.parse("[]?").cardinality() == 1
assert mult.parse("[]{0,6}").cardinality() == 1
assert mult.parse("[ab]{3}").cardinality() == 8
assert mult.parse("[ab]{2,3}").cardinality() == 12
assert len(pattern.parse("abc|def(ghi|jkl)")) == 3
try:
len(pattern.parse(".*"))
assert False
except OverflowError:
assert True
def test_conc_reduction_basic():
assert conc.parse("a[]b").reduce() == charclass.parse("[]")
# conc -> conc
assert conc.parse("ab").reduce() == conc.parse("ab")
# conc -> mult
assert conc.parse("a{3,4}").reduce() == mult.parse("a{3,4}")
# conc -> charclass
assert conc.parse("a").reduce() == charclass.parse("a")
def test_pattern_reduce_basic():
# pattern -> pattern
# (ab|cd) -> (ab|cd)
assert pattern.parse("ab|cd").reduce() == pattern.parse("ab|cd")
# pattern -> conc
assert pattern.parse("a{2}b{2}").reduce() == conc.parse("a{2}b{2}")
# pattern -> mult
assert pattern.parse("a{2}").reduce() == mult.parse("a{2}")
# pattern -> charclass
assert pattern.parse("a").reduce() == charclass.parse("a")
def test_mult_parsing():
assert mult.parse("[a-g]+") == mult(charclass("abcdefg"), plus)
assert mult.parse("[a-g0-8$%]+") == mult(charclass("abcdefg012345678$%"), plus)
assert mult.parse("[a-g0-8$%\\^]+") == mult(charclass("abcdefg012345678$%^"), plus)
assert mult.match("abcde[^fg]*", 5) == (
mult(~charclass("fg"), star),
11
)
assert mult.match("abcde[^fg]*h{5}[a-z]+", 11) == (
mult(charclass("h"), multiplier(bound(5), bound(5))),
15
)
assert mult.match("abcde[^fg]*h{5}[a-z]+T{1,}", 15) == (
mult(charclass("abcdefghijklmnopqrstuvwxyz"), plus),
21
)
assert mult.match("abcde[^fg]*h{5}[a-z]+T{2,}", 21) == (
mult(charclass("T"), multiplier(bound(2), inf)),
26
)
def test_mult_reduction_easy():
assert mult.parse("a").reduce() == charclass.parse("a")
assert mult.parse("a").reduce() == charclass("a")
assert mult.parse("a?").reduce() == mult(charclass("a"), qm)
assert mult.parse("a{0}").reduce() == emptystring
assert mult.parse("[]").reduce() == nothing
assert mult.parse("[]?").reduce() == emptystring
assert mult.parse("[]{0}").reduce() == emptystring
assert mult.parse("[]{0,5}").reduce() == emptystring
assert mult(pattern(), one).reduce() == nothing
assert mult(pattern(), qm).reduce() == emptystring
assert mult(pattern(), zero).reduce() == emptystring
assert mult(pattern(), multiplier.parse("{0,5}")).reduce() == emptystring
def test_mult_reduction_easy():
assert mult.parse("a").reduce() == charclass.parse("a")
assert mult.parse("a").reduce() == charclass("a")
assert mult.parse("a?").reduce() == mult(charclass("a"), qm)
assert mult.parse("a{0}").reduce() == emptystring
assert mult.parse("[]").reduce() == nothing
assert mult.parse("[]?").reduce() == emptystring
assert mult.parse("[]{0}").reduce() == emptystring
assert mult.parse("[]{0,5}").reduce() == emptystring
assert mult(pattern(), one).reduce() == nothing
assert mult(pattern(), qm).reduce() == emptystring
assert mult(pattern(), zero).reduce() == emptystring
assert mult(pattern(), multiplier.parse("{0,5}")).reduce() == emptystring
def test_mult_parsing():
assert mult.parse("[a-g]+") == mult(charclass("abcdefg"), plus)
assert mult.parse("[a-g0-8$%]+") == mult(charclass("abcdefg012345678$%"), plus)
assert mult.parse("[a-g0-8$%\\^]+") == mult(charclass("abcdefg012345678$%^"), plus)
assert mult.match("abcde[^fg]*", 5) == (
mult(~charclass("fg"), star),
11
)
assert mult.match("abcde[^fg]*h{5}[a-z]+", 11) == (
mult(charclass("h"), multiplier(bound(5), bound(5))),
15
)
assert mult.match("abcde[^fg]*h{5}[a-z]+T{1,}", 15) == (
mult(charclass("abcdefghijklmnopqrstuvwxyz"), plus),
21
)
assert mult.match("abcde[^fg]*h{5}[a-z]+T{2,}", 21) == (
mult(charclass("T"), multiplier(bound(2), inf)),
26
)
def test_mult_multiplication():
# a{2,3} * 1 = a{2,3}
assert mult.parse("a{2,3}") * one == mult.parse("a{2,3}")
# a{2,3} * {4,5} = a{8,15}
assert mult.parse("a{2,3}") * multiplier.parse("{4,5}") == mult.parse("a{8,15}")
# a{2,} * {2,} = a{4,}
assert mult.parse("a{2,}") * multiplier.parse("{2,}") == mult.parse("a{4,}")
def test_concatenation():
assert charclass.parse("a") + charclass.parse("b") == conc.parse("ab")
assert charclass.parse("a") + mult.parse("b{0,8}") == conc.parse("ab{0,8}")
assert charclass.parse("a") + conc.parse("bc") == conc.parse("abc")
assert charclass.parse("a") + pattern.parse("b|cd") == conc.parse("a(b|cd)")
assert mult.parse("b{0,8}") + charclass.parse("c") == conc.parse("b{0,8}c")
assert mult.parse("a{3,4}") + mult.parse("b?") == conc.parse("a{3,4}b?")
assert mult.parse("a{2}") + conc.parse("bc") == conc.parse("a{2}bc")
assert mult.parse("a{2,3}") + pattern.parse("b|cd") == conc.parse("a{2,3}(b|cd)")
assert conc.parse("ab") + charclass.parse("c") == conc.parse("abc")
assert conc.parse("ab") + mult.parse("c*") == conc.parse("abc*")
assert conc.parse("") + conc.parse("") == conc.parse("")
assert conc.parse("ab") + conc.parse("cd") == conc.parse("abcd")
assert conc.parse("za{2,3}") + pattern.parse("b|cd") == conc.parse("za{2,3}(b|cd)")
assert pattern.parse("a|bd") + charclass.parse("c") == conc.parse("(a|bd)c")
assert pattern.parse("b|cd") + mult.parse("a{2,3}") == conc.parse("(b|cd)a{2,3}")
assert pattern.parse("b|cd") + conc.parse("za{2,3}") == conc.parse("(b|cd)za{2,3}")
assert pattern.parse("a|bc") + pattern.parse("c|de") == conc.parse("(a|bc)(c|de)")
def test_concatenation():
assert charclass.parse("a") + charclass.parse("b") == conc.parse("ab")
assert charclass.parse("a") + mult.parse("b{0,8}") == conc.parse("ab{0,8}")
assert charclass.parse("a") + conc.parse("bc") == conc.parse("abc")
assert charclass.parse("a") + pattern.parse("b|cd") == conc.parse("a(b|cd)")
assert mult.parse("b{0,8}") + charclass.parse("c") == conc.parse("b{0,8}c")
assert mult.parse("a{3,4}") + mult.parse("b?") == conc.parse("a{3,4}b?")
assert mult.parse("a{2}") + conc.parse("bc") == conc.parse("a{2}bc")
assert mult.parse("a{2,3}") + pattern.parse("b|cd") == conc.parse("a{2,3}(b|cd)")
assert conc.parse("ab") + charclass.parse("c") == conc.parse("abc")
assert conc.parse("ab") + mult.parse("c*") == conc.parse("abc*")
assert conc.parse("") + conc.parse("") == conc.parse("")
assert conc.parse("ab") + conc.parse("cd") == conc.parse("abcd")
assert conc.parse("za{2,3}") + pattern.parse("b|cd") == conc.parse("za{2,3}(b|cd)")
assert pattern.parse("a|bd") + charclass.parse("c") == conc.parse("(a|bd)c")
assert pattern.parse("b|cd") + mult.parse("a{2,3}") == conc.parse("(b|cd)a{2,3}")
assert pattern.parse("b|cd") + conc.parse("za{2,3}") == conc.parse("(b|cd)za{2,3}")
assert pattern.parse("a|bc") + pattern.parse("c|de") == conc.parse("(a|bc)(c|de)")
def test_reduce_concatenations():
assert parse("aa").reduce() == mult.parse("a{2}")
assert parse("bb").reduce() == mult.parse("b{2}")
assert parse("b*b").reduce() == mult.parse("b+")
assert parse("aa{2,}").reduce() == mult.parse("a{3,}")
assert parse("a*a{2}").reduce() == mult.parse("a{2,}")
assert parse("aa{0,8}").reduce() == mult.parse("a{1,9}")
assert parse("b{0,8}b").reduce() == mult.parse("b{1,9}")
assert parse("aab").reduce() == conc.parse("a{2}b")
assert parse("abb").reduce() == conc.parse("ab{2}")
assert parse("abb*").reduce() == conc.parse("ab+")
assert parse("abbc").reduce() == conc.parse("ab{2}c")
assert parse("a?ab").reduce() == conc.parse("a{1,2}b")
assert parse("(ac{2}|bc+)c").reduce() == conc.parse("(ac|bc*)c{2}")
assert parse("a(a{2}b|a+c)").reduce() == conc.parse("a{2}(ab|a*c)")
assert parse("a{2,3}(a{2}b|a+c)").reduce() == conc.parse("a{3,4}(ab|a*c)")
assert parse("(ba{2}|ca+)a{2,3}").reduce() == conc.parse("(ba|ca*)a{3,4}")
assert parse("za{2,3}(a{2}b|a+c)").reduce() == conc.parse("za{3,4}(ab|a*c)")
assert parse("(ba{2}|ca+)a{2,3}z").reduce() == conc.parse("(ba|ca*)a{3,4}z")
assert parse("(a|bc)(a|bc)").reduce() == mult.parse("(a|bc){2}")
def test_reduce_concatenations():
assert parse("aa").reduce() == mult.parse("a{2}")
assert parse("bb").reduce() == mult.parse("b{2}")
assert parse("b*b").reduce() == mult.parse("b+")
assert parse("aa{2,}").reduce() == mult.parse("a{3,}")
assert parse("a*a{2}").reduce() == mult.parse("a{2,}")
assert parse("aa{0,8}").reduce() == mult.parse("a{1,9}")
assert parse("b{0,8}b").reduce() == mult.parse("b{1,9}")
assert parse("aab").reduce() == conc.parse("a{2}b")
assert parse("abb").reduce() == conc.parse("ab{2}")
assert parse("abb*").reduce() == conc.parse("ab+")
assert parse("abbc").reduce() == conc.parse("ab{2}c")
assert parse("a?ab").reduce() == conc.parse("a{1,2}b")
assert parse("(ac{2}|bc+)c").reduce() == conc.parse("(ac|bc*)c{2}")
assert parse("a(a{2}b|a+c)").reduce() == conc.parse("a{2}(ab|a*c)")
assert parse("a{2,3}(a{2}b|a+c)").reduce() == conc.parse("a{3,4}(ab|a*c)")
assert parse("(ba{2}|ca+)a{2,3}").reduce() == conc.parse("(ba|ca*)a{3,4}")
assert parse("za{2,3}(a{2}b|a+c)").reduce() == conc.parse("za{3,4}(ab|a*c)")
assert parse("(ba{2}|ca+)a{2,3}z").reduce() == conc.parse("(ba|ca*)a{3,4}z")
assert parse("(a|bc)(a|bc)").reduce() == mult.parse("(a|bc){2}")
def test_pattern_generator():
gen = pattern.parse("[ab]|[cde]").strings()
assert next(gen) == "a"
assert next(gen) == "b"
assert next(gen) == "c"
assert next(gen) == "d"
assert next(gen) == "e"
try:
next(gen)
assert False
except StopIteration:
assert True
# more complex
gen = pattern.parse("abc|def(ghi|jkl)").strings()
assert next(gen) == "abc"
assert next(gen) == "defghi"
assert next(gen) == "defjkl"
gen = mult.parse("[0-9a-fA-F]{3,10}").strings()
assert next(gen) == "000"
assert next(gen) == "001"
assert next(gen) == "002"
def test_pattern_generator():
gen = pattern.parse("[ab]|[cde]").strings()
assert next(gen) == "a"
assert next(gen) == "b"
assert next(gen) == "c"
assert next(gen) == "d"
assert next(gen) == "e"
try:
next(gen)
assert False
except StopIteration:
assert True
# more complex
gen = pattern.parse("abc|def(ghi|jkl)").strings()
assert next(gen) == "abc"
assert next(gen) == "defghi"
assert next(gen) == "defjkl"
gen = mult.parse("[0-9a-fA-F]{3,10}").strings()
assert next(gen) == "000"
assert next(gen) == "001"
assert next(gen) == "002"
def test_conc_reduction_basic():
assert conc.parse("a").reduce() == charclass.parse("a")
assert conc.parse("a{3,4}").reduce() == mult.parse("a{3,4}")
assert conc.parse("ab").reduce() == conc.parse("ab")
assert conc.parse("a[]b").reduce() == charclass.parse("[]")
def test_even_star_bug():
# Defect: (a{2})* should NOT reduce to a*
a2 = mult.parse("a{2}")
a2star = a2 * star
assert a2star == mult.parse("(a{2})*")
def test_mult_common():
assert mult.parse("a{3,4}").common(mult.parse("a{2,5}")) == mult.parse("a{2,3}")
assert mult.parse("a{2,}").common(mult.parse("a{1,5}")) == mult.parse("a{1,5}")
assert mult.parse("a{3,}").common(mult.parse("a{2,}")) == mult.parse("a{2,}")
def test_charclass_multiplication():
assert charclass("a") * one == charclass("a")
assert charclass("a") * multiplier.parse("{1,3}") == mult.parse("a{1,3}")
assert charclass("a") * multiplier.parse("{4,}") == mult.parse("a{4,}")
def test_mult_multiplication():
assert mult.parse("a{2,3}") * one == mult.parse("a{2,3}")
assert mult.parse("a{2,3}") * multiplier.parse("{4,5}") == mult.parse("a{8,15}")
assert mult.parse("a{2,}") * multiplier.parse("{2,}") == mult.parse("a{4,}")
def test_mult_squoosh():
# sequence squooshing of mults within a conc
assert conc.parse("[0-9a-fA-F][0-9a-fA-F][0-9a-fA-F]").reduce() == mult.parse("[0-9A-Fa-f]{3}")
assert conc.parse("[$%\\^]?[$%\\^]").reduce() == mult.parse("[$%\\^]{1,2}")
assert conc.parse("(|(|(|(|(|(|[$%\\^])[$%\\^])[$%\\^])[$%\\^])[$%\\^])[$%\\^])[$%\\^][$%\\^]").reduce() == mult.parse("[$%\\^]{2,8}")
def test_conc_multiplication():
assert conc.parse("ab?") * qm == mult.parse("(ab?)?")
def test_conc_multiplication():
assert conc.parse("ab?") * qm == mult.parse("(ab?)?")
def test_mult_multiplication():
assert mult.parse("a{2,3}") * one == mult.parse("a{2,3}")
assert mult.parse("a{2,3}") * multiplier.parse("{4,5}") == mult.parse("a{8,15}")
assert mult.parse("a{2,}") * multiplier.parse("{2,}") == mult.parse("a{4,}")
def test_charclass_multiplication():
assert charclass("a") * one == charclass("a")
assert charclass("a") * multiplier.parse("{1,3}") == mult.parse("a{1,3}")
assert charclass("a") * multiplier.parse("{4,}") == mult.parse("a{4,}")
def test_obvious_reduction():
assert mult.parse("(a|b)*").reduce() == mult.parse("[ab]*")
def test_mult_dock():
assert mult.parse("a{4,5}").dock(mult.parse("a{3}")) == mult.parse("a{1,2}")
def test_mult_intersection():
assert mult.parse("a") & mult.parse("b?") == charclass()
assert mult.parse("a") & mult.parse("b?") == nothing
assert mult.parse("a") & mult.parse("a?") == charclass.parse("a")
assert mult.parse("a{2}") & mult.parse("a{2,}") == mult.parse("a{2}")
assert mult.parse("a") & mult.parse("b") == charclass.parse("[]")
assert mult.parse("a") & mult.parse("a") == charclass.parse("a")
assert mult.parse("a*") & mult.parse("a") == charclass.parse("a")
assert mult.parse("a*") & mult.parse("b*") == conc.parse("")
assert mult.parse("a*") & mult.parse("a+") == mult.parse("a+")
assert mult.parse("a{2}") & mult.parse("a{4}") == charclass.parse("[]")
assert mult.parse("a{3,}") & mult.parse("a{3,}") == mult.parse("a{3,}")
def test_pattern_reduce_basic():
assert pattern.parse("ab|cd").reduce() == pattern.parse("ab|cd")
assert pattern.parse("a{2}b{2}").reduce() == conc.parse("a{2}b{2}")
assert pattern.parse("a{2}").reduce() == mult.parse("a{2}")
assert pattern.parse("a").reduce() == charclass.parse("a")
def test_pattern_multiplication():
assert pattern.parse("ab?|ba?") * multiplier.parse("{2,3}") == mult.parse("(ab?|ba?){2,3}")
def test_pattern_multiplication():
assert pattern.parse("ab?|ba?") * multiplier.parse("{2,3}") == mult.parse("(ab?|ba?){2,3}")
def test_obvious_reduction():
assert mult.parse("(a|b)*").reduce() == mult.parse("[ab]*")
def test_even_star_bug():
# Defect: (a{2})* should NOT reduce to a*
a2 = mult.parse("a{2}")
a2star = a2 * star
assert a2star == mult.parse("(a{2})*")
def test_epsilon_reduction():
assert parse("|(ab)*|def").reduce() == pattern.parse("(ab)*|def")
assert parse("|(ab)+|def").reduce() == pattern.parse("(ab)*|def")
assert parse("|.+").reduce() == mult.parse(".*")
assert parse("|a+|b+").reduce() in {pattern.parse("a+|b*"), pattern.parse("a*|b+")}
def test_mult_squoosh():
# sequence squooshing of mults within a conc
assert conc.parse("[0-9a-fA-F][0-9a-fA-F][0-9a-fA-F]").reduce() == mult.parse("[0-9A-Fa-f]{3}")
assert conc.parse("[$%\\^]?[$%\\^]").reduce() == mult.parse("[$%\\^]{1,2}")
assert conc.parse("(|(|(|(|(|(|[$%\^])[$%\^])[$%\^])[$%\^])[$%\^])[$%\^])[$%\^][$%\^]").reduce() == mult.parse("[$%\^]{2,8}")
def test_derive():
assert parse("a+").derive("a") == mult.parse("a*")
assert parse("a+|b+").derive("a") == mult.parse("a*")
assert parse("abc|ade").derive("a") == pattern.parse("bc|de")
assert parse("abc|ade").derive("ab") == charclass.parse("c")
def test_concatenation():
# empty conc + empty conc
assert conc.parse("") + conc.parse("") == conc.parse("")
# charclass + charclass
# a + b = ab
assert charclass.parse("a") + charclass.parse("b") == conc.parse("ab")
# a + a = a{2}
assert (charclass.parse("a") + charclass.parse("a")).reduce() == mult.parse("a{2}")
# charclass + mult
# a + a = a{2}
assert (charclass.parse("a") + mult.parse("a")).reduce() == mult.parse("a{2}")
# a + a{2,} = a{3,}
assert (charclass.parse("a") + mult.parse("a{2,}")).reduce() == mult.parse("a{3,}")
# a + a{,8} = a{1,9}
assert (charclass.parse("a") + mult.parse("a{0,8}")).reduce() == mult.parse("a{1,9}")
# a + b{,8} = ab{,8}
assert charclass.parse("a") + mult.parse("b{0,8}") == conc.parse("ab{0,8}")
# mult + charclass
# b + b = b{2}
assert (mult.parse("b") + charclass.parse("b")).reduce() == mult.parse("b{2}")
# b* + b = b+
assert (mult.parse("b*") + charclass.parse("b")).reduce() == mult.parse("b+")
# b{,8} + b = b{1,9}
assert (mult.parse("b{0,8}") + charclass.parse("b")).reduce() == mult.parse("b{1,9}")
# b{,8} + c = b{,8}c
assert mult.parse("b{0,8}") + charclass.parse("c") == conc.parse("b{0,8}c")
# charclass + conc
# a + nothing = a
assert (charclass.parse("a") + conc.parse("")).reduce() == charclass.parse("a")
# a + bc = abc
assert charclass.parse("a") + conc.parse("bc") == conc.parse("abc")
# a + ab = a{2}b
assert (charclass.parse("a") + conc.parse("ab")).reduce() == conc.parse("a{2}b")
# conc + charclass
# nothing + a = a
assert (conc.parse("") + charclass.parse("a")).reduce() == charclass.parse("a")
# ab + c = abc
assert conc.parse("ab") + charclass.parse("c") == conc.parse("abc")
# ab + b = ab{2}
assert (conc.parse("ab") + charclass.parse("b")).reduce() == conc.parse("ab{2}")
# pattern + charclass
# (a|bd) + c = (a|bd)c
assert pattern.parse("a|bd") + charclass.parse("c") == conc.parse("(a|bd)c")
# (ac{2}|bc+) + c = (ac|bc*)c{2}
assert (pattern.parse("ac{2}|bc+") + charclass.parse("c")).reduce() == conc.parse("(ac|bc*)c{2}")
# charclass + pattern
# a + (b|cd) = a(b|cd)
assert charclass.parse("a") + pattern.parse("b|cd") == conc.parse("a(b|cd)")
# a + (a{2}b|a+c) = a{2}(ab|a*c)
assert (charclass.parse("a") + pattern.parse("(a{2}b|a+c)")).reduce() == conc.parse("a{2}(ab|a*c)")
# mult + mult
# a{3,4} + b? = a{3,4}b?
assert mult.parse("a{3,4}") + mult.parse("b?") == conc.parse("a{3,4}b?")
# a* + a{2} = a{2,}
assert (mult.parse("a*") + mult.parse("a{2}")).reduce() == mult.parse("a{2,}")
# mult + conc
# a{2} + bc = a{2}bc
assert mult.parse("a{2}") + conc.parse("bc") == conc.parse("a{2}bc")
# a? + ab = a{1,2}b
assert (mult.parse("a?") + conc.parse("ab")).reduce() == conc.parse("a{1,2}b")
# conc + mult
# ab + c* = abc*
assert conc.parse("ab") + mult.parse("c*") == conc.parse("abc*")
# ab + b* = ab+
assert (conc.parse("ab") + mult.parse("b*")).reduce() == conc.parse("ab+")
# mult + pattern
# a{2,3} + (b|cd) = a{2,3}(b|cd)
assert mult.parse("a{2,3}") + pattern.parse("b|cd") == conc.parse("a{2,3}(b|cd)")
# a{2,3} + (a{2}b|a+c) = a{3,4}(ab|a*c)
assert (mult.parse("a{2,3}") + pattern.parse("a{2}b|a+c")).reduce() == conc.parse("a{3,4}(ab|a*c)")
# pattern + mult
# (b|cd) + a{2,3} = (b|cd)a{2,3}
assert pattern.parse("b|cd") + mult.parse("a{2,3}") == conc.parse("(b|cd)a{2,3}")
# (ba{2}|ca+) + a{2,3} = (ba|ca*)a{3,4}
assert (pattern.parse("ba{2}|ca+") + mult.parse("a{2,3}")).reduce() == conc.parse("(ba|ca*)a{3,4}")
# conc + conc
# ab + cd = abcd
assert conc.parse("ab") + conc.parse("cd") == conc.parse("abcd")
# ab + bc = ab{2}c
assert (conc.parse("ab") + conc.parse("bc")).reduce() == conc.parse("ab{2}c")
# conc + pattern
# za{2,3} + (b|cd) = za{2,3}(b|cd)
assert conc.parse("za{2,3}") + pattern.parse("b|cd") == conc.parse("za{2,3}(b|cd)")
# za{2,3} + (a{2}b|a+c) = za{3,4}(ab|a*c)
assert (conc.parse("za{2,3}") + pattern.parse("a{2}b|a+c")).reduce() == conc.parse("za{3,4}(ab|a*c)")
# pattern + conc
# (b|cd) + za{2,3} = (b|cd)za{2,3}
assert pattern.parse("b|cd") + conc.parse("za{2,3}") == conc.parse("(b|cd)za{2,3}")
# (ba{2}|ca+) + a{2,3}z = (ba|ca*)a{3,4}z
assert (pattern.parse("ba{2}|ca+") + conc.parse("a{2,3}z")).reduce() == conc.parse("(ba|ca*)a{3,4}z")
# pattern + pattern
# (a|bc) + (c|de) = (a|bc)(c|de)
assert pattern.parse("a|bc") + pattern.parse("c|de") == conc.parse("(a|bc)(c|de)")
# (a|bc) + (a|bc) = (a|bc){2}
assert (pattern.parse("a|bc") + pattern.parse("a|bc")).reduce() == mult.parse("(a|bc){2}")
def test_two_or_more_qm_bug():
assert mult.parse("a{2,}") * qm == mult.parse("(a{2,})?")
def test_pattern_reduce_basic():
assert pattern.parse("ab|cd").reduce() == pattern.parse("ab|cd")
assert pattern.parse("a{2}b{2}").reduce() == conc.parse("a{2}b{2}")
assert pattern.parse("a{2}").reduce() == mult.parse("a{2}")
assert pattern.parse("a").reduce() == charclass.parse("a")
def test_empty_pattern_suppression():
assert mult.parse("[]?").reduce() == conc.parse("")
assert mult(pattern(), qm).reduce() == conc.parse("")
def test_two_two_bug():
assert mult.parse("a{2}") * multiplier.parse("{2}") == mult.parse("a{4}")
def test_mult_intersection():
assert mult.parse("a") & mult.parse("b?") == charclass()
assert mult.parse("a") & mult.parse("b?") == nothing
assert mult.parse("a") & mult.parse("a?") == charclass.parse("a")
assert mult.parse("a{2}") & mult.parse("a{2,}") == mult.parse("a{2}")
assert mult.parse("a") & mult.parse("b") == charclass.parse("[]")
assert mult.parse("a") & mult.parse("a") == charclass.parse("a")
assert mult.parse("a*") & mult.parse("a") == charclass.parse("a")
assert mult.parse("a*") & mult.parse("b*") == conc.parse("")
assert mult.parse("a*") & mult.parse("a+") == mult.parse("a+")
assert mult.parse("a{2}") & mult.parse("a{4}") == charclass.parse("[]")
assert mult.parse("a{3,}") & mult.parse("a{3,}") == mult.parse("a{3,}")
def test_epsilon_reduction():
assert parse("|(ab)*|def").reduce() == pattern.parse("(ab)*|def")
assert parse("|(ab)+|def").reduce() == pattern.parse("(ab)*|def")
assert parse("|.+").reduce() == mult.parse(".*")
assert parse("|a+|b+").reduce() in {pattern.parse("a+|b*"), pattern.parse("a*|b+")}
def test_mult_dock():
assert mult.parse("a{4,5}").dock(mult.parse("a{3}")) == mult.parse("a{1,2}")
def test_mult_common():
assert mult.parse("a{3,4}").common(mult.parse("a{2,5}")) == mult.parse("a{2,3}")
assert mult.parse("a{2,}").common(mult.parse("a{1,5}")) == mult.parse("a{1,5}")
assert mult.parse("a{3,}").common(mult.parse("a{2,}")) == mult.parse("a{2,}")