def test_preprocess_taxons():
dbf = ProgramFilter(db)
names = ["X/S/M.*", "O(?!/C).*", "Y/E", "non_matching_pattern"]
taxons = dbf.preprocess_taxons(names)
print(sorted(taxons))
assert taxons == [
"O",
"O/J",
"O/N",
"O/N/P",
"X/S/M",
"X/S/M/L",
"X/S/M/L/R",
"X/S/M/L/R/D",
"X/S/M/L/R/D/A",
"X/S/M/L/V",
"Y/E",
]
names = [
("never mind", "O/J", "Y/E"),
("whatever", r"X/S/M/L/R/D.*", r"O/N.*"),
("who cares", r"X/S/M/L/R.*", r"O/N"),
]
taxons = dbf.preprocess_taxons(names)
print(sorted(taxons))
assert taxons == [
("never mind", "O/J", "Y/E"),
("whatever", "X/S/M/L/R/D", "O/N"),
("whatever", "X/S/M/L/R/D", "O/N/P"),
("whatever", "X/S/M/L/R/D/A", "O/N"),
("whatever", "X/S/M/L/R/D/A", "O/N/P"),
("who cares", "X/S/M/L/R", "O/N"),
("who cares", "X/S/M/L/R/D", "O/N"),
("who cares", "X/S/M/L/R/D/A", "O/N"),
]
def test_programs_of_pattern():
dbf = ProgramFilter(db)
names = [r"prg[1-3]\.py", r"prg[7-9]\.py", "non_matching_pattern"]
programs = set().union(*(dbf.programs_of_pattern(name) for name in names))
print(sorted(programs))
assert sorted(programs) == [
"prg1.py", "prg2.py", "prg3.py", "prg7.py", "prg8.py", "prg9.py"
]
def test_taxons_of_programs():
dbf = ProgramFilter(db)
programs = {"prg8.py", "prg9.py", "non_existing_program"}
taxons = dbf.taxons_of_programs(programs)
prg8_taxons = set(dbf.db_programs["prg8.py"]["taxons"])
prg9_taxons = set(dbf.db_programs["prg9.py"]["taxons"])
print(sorted(taxons))
assert taxons == prg8_taxons | prg9_taxons
def test_exclude_taxons():
dbf = ProgramFilter(db)
taxons = ["O/J", "X/S/M/L", "non_existing_taxon"]
dbf.exclude_taxons(taxons)
print(sorted(dbf.selected_programs))
assert dbf.selected_programs.keys() == {"prg2.py", "prg5.py"}
for taxon in taxons:
assert taxon not in db["programs"]["prg2.py"]["taxons"]
assert taxon not in db["programs"]["prg5.py"]["taxons"]
def test_programs_of_taxons():
dbf = ProgramFilter(db)
taxons = {"X/S/M/L/R/D/A", "X/S/M/L/R/D", "non_existing_taxon"}
programs = dbf.programs_of_taxons(taxons, follow=False)
print(sorted(programs))
for (db_program, db_program_data) in db["programs"].items():
if taxons.intersection(db_program_data["taxons"]):
assert db_program in programs
else:
assert db_program not in programs
def test_impart_programs():
dbf = ProgramFilter(db)
programs = {"prg1.py", "prg2.py", "non_existing_program"}
dbf.impart_programs(programs)
print(sorted(dbf.selected_programs))
assert dbf.selected_programs.keys() == set(db["programs"]) - set(programs)
print(sorted(dbf.imparted_knowledge))
assert set(db["taxons"]) - dbf.imparted_knowledge == {
"O/C/F",
"O/C/F/U",
"X/S/M/L/R/D/A",
"Y/E",
}
def test_impart_taxons():
# Imparting a knowledge decreases the learning cost of the corresponding taxons, but has no
# effect whatsoever on the selected programs.
dbf = ProgramFilter(db)
dbf.impart_taxons({"flow/conditional"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {
"assignment.py",
"collatz_print.py",
"fizzbuzz.py",
"is_even.py",
}
def test_include_taxons():
dbf = ProgramFilter(db)
taxons = ["O/J", "X/S/M/L", "non_existing_taxon"]
dbf.include_taxons(taxons)
print(sorted(dbf.selected_programs))
assert dbf.selected_programs.keys() == {
"prg1.py",
"prg3.py",
"prg4.py",
"prg6.py",
"prg7.py",
"prg8.py",
"prg9.py",
}
def test_impart_taxons():
dbf = ProgramFilter(db)
taxons = ["O/J", "X/S/M/L", "non/existing/taxon"]
dbf.impart_taxons(taxons)
print(sorted(dbf.imparted_knowledge))
assert dbf.imparted_knowledge == {
"O",
"O/J",
"X",
"X/S",
"X/S/M",
"X/S/M/L",
"non",
"non/existing",
"non/existing/taxon",
}
def test_init():
dbf = ProgramFilter(db)
print(dbf.selected_programs)
assert set(dbf.selected_programs.keys()) == {
"assignment.py",
"collatz_print.py",
"fizzbuzz.py",
"is_even.py",
}
def test_taxa_of_pattern():
dbf = ProgramFilter(db)
names = ["X/S/M", "O(?!/C)", "Y/E$", "non_matching_pattern"]
taxa = set().union(*(dbf.taxa_of_pattern(name) for name in names))
print(sorted(taxa))
assert sorted(taxa) == [
"O",
"O/J",
"O/N",
"O/N/P",
"X/S/M",
"X/S/M/L",
"X/S/M/L/R",
"X/S/M/L/R/D",
"X/S/M/L/R/D/A",
"X/S/M/L/V",
"Y/E",
]
def test_taxa_of_programs():
dbf = ProgramFilter(db)
# collatz is imported by fizzbuzz
subset_taxa = dbf.taxa_of_programs({"collatz.py"}, follow=False)
subset_taxa = set(filter(lambda x: not x.startswith("meta/"), subset_taxa))
print(sorted(subset_taxa))
print()
# when following the importations...
superset_taxa = dbf.taxa_of_programs({"fizzbuzz.py"}, follow=True)
superset_taxa = set(
filter(lambda x: not x.startswith("meta/"), superset_taxa))
print(sorted(superset_taxa))
# in addition to its own taxa, fizzbuzz features all those of collatz...
assert superset_taxa.issuperset(subset_taxa)
assert len(superset_taxa - subset_taxa) > 0
# when not following the importations...
own_taxa = dbf.taxa_of_programs({"fizzbuzz.py"}, follow=False)
own_taxa = set(filter(lambda x: not x.startswith("meta/"), own_taxa))
print(sorted(own_taxa))
assert superset_taxa.issuperset(own_taxa)
assert len(superset_taxa - own_taxa) > 0
def test_programs_of_taxons():
dbf = ProgramFilter(db)
taxons = {"variable/assignment/single"}
# The taxon "variable/assignment/single" is featured by assignment.py and collatz_print.py.
# This corresponds to follow=False. It is indirectly featured by fizzbuzz.py (which imports
# collatz_print.py) and by is_even.py (which imports fizzbuzz.py). Their addition corresponds
# to follow=True.
programs = dbf.programs_of_taxons(taxons, follow=False)
print(set(programs))
assert set(programs) == {"assignment.py", "collatz_print.py"}
programs = dbf.programs_of_taxons(taxons, follow=True)
print(set(programs))
assert set(programs) == {
"assignment.py",
"collatz_print.py",
"fizzbuzz.py",
"is_even.py",
}
def test_include_programs():
# The program collatz.py is self-contained. Including it has no other effect on the
# selection.
dbf = ProgramFilter(db)
dbf.include_programs({"collatz.py"})
print(dbf.selected_programs)
assert dbf.selected_programs == {"collatz.py"}
# Including a program does not include the programs it imports.
dbf = ProgramFilter(db)
dbf.include_programs({"fizzbuzz.py"})
print(dbf.selected_programs)
assert dbf.selected_programs == {"fizzbuzz.py"}
def test_impart_taxons():
# Imparting a triple doesn't make much sense. Currently, it comes down to imparting the two
# taxons, and ignoring the predicate.
dbf = ProgramFilter(db)
dbf.impart_taxons({("equals", "operator/arithmetic/modulo", "type/number/integer")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {
"assignment.py",
"collatz_print.py",
"fizzbuzz.py",
"is_even.py",
}
print(dbf.imparted_knowledge)
assert dbf.imparted_knowledge == {
"operator",
"operator/arithmetic",
"operator/arithmetic/modulo",
"type",
"type/number",
"type/number/integer",
}
def test_include_taxons():
# The taxon "variable/assignment/single" is directly featured by assignment.py and
# collatz_print.py, but only indirectly by the other programs, which therefore cannot be
# included in the result. Note that this behavior contrasts with that of exclude_taxons.
dbf = ProgramFilter(db)
dbf.include_taxons({"variable/assignment/single"})
print(set(dbf.selected_programs.keys()))
assert set(
dbf.selected_programs.keys()) == {"assignment.py", "collatz_print.py"}
# "operator/arithmetic/addition" is directly featured by collatz_print.py only. Therefore,
# including this taxon keeps only collatz_print.py.
dbf = ProgramFilter(db)
dbf.include_taxons({"operator/arithmetic/addition"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py"}
# "flow/conditional" is featured by collatz_print.py, fizzbuzz.py, and indirectly by
# is_even.py. Therefore, including this taxon keeps only the former two.
dbf = ProgramFilter(db)
dbf.include_taxons({"flow/conditional"})
print(set(dbf.selected_programs.keys()))
assert set(
dbf.selected_programs.keys()) == {"collatz_print.py", "fizzbuzz.py"}
# "type/sequence/string/literal" is directly featured by fizzbuzz.py only. Therefore, including
# this taxon keeps only fizzbuzz.py
dbf = ProgramFilter(db)
dbf.include_taxons({"type/sequence/string/literal"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"fizzbuzz.py"}
import json
from pathlib import Path
import pytest
import context
from paroxython.filter_programs import ProgramFilter
from paroxython.compare_spans import compare_spans
db = json.loads(Path("examples/mini/programs_db.json").read_text())
dbf = ProgramFilter(db)
triple_data = [
(
(
"var/assignment/explicit/single", # featured by assignment.py and collatz.py
compare_spans["after"], # but after
"call/function/builtin/print", # this taxon in collatz.py only.
),
{"collatz.py"},
),
(
(
"operator/arithmetic/addition", # this taxon is featured by collatz.py only
compare_spans["equals"], # and on the same line
"operator/arithmetic/multiplication", # as that taxon
),
{"collatz.py"},
),
(
(
def test_include_taxons():
# The taxon "variable/assignment/single" is featured by assignment.py and collatz_print.py. In
# collatz_print.py, it appears after a taxon "io/standard/print". Consequently, it should be included
# in the results, but not the programs which import it: fizzbuzz.py and is_even.py.
dbf = ProgramFilter(db)
dbf.include_taxons({("after", "variable/assignment/single", "io/standard/print")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py"}
# "operator/arithmetic/modulo" and "type/number/integer/literal" are both featured on
# the same line in all programs except assignment.py
dbf = ProgramFilter(db)
dbf.include_taxons({("equals", "operator/arithmetic/modulo", "type/number/integer/literal")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py", "is_even.py", "fizzbuzz.py"}
# The same with "x == y" instead of "equals"
dbf = ProgramFilter(db)
dbf.include_taxons({("equals", "operator/arithmetic/modulo", "type/number/integer/literal")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py", "is_even.py", "fizzbuzz.py"}
# "test/equality" is inside "subroutine/function" in is_even.py, which is not imported anywhere.
dbf = ProgramFilter(db)
dbf.include_taxons({("inside", "test/equality", "subroutine/function")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"is_even.py"}
# "test/equality" is inside "subroutine/function" in is_even.py and inside
# "subroutine/procedure" in collatz_print.py. Both will be included.
dbf = ProgramFilter(db)
taxons = dbf.preprocess_taxons([("inside", "test/equality", "subroutine/.*")])
print(taxons)
assert taxons == [
("inside", "test/equality", "subroutine/argument/arg"),
("inside", "test/equality", "subroutine/function"),
("inside", "test/equality", "subroutine/predicate"),
("inside", "test/equality", "subroutine/procedure"),
]
dbf.include_taxons(taxons)
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py", "is_even.py"}
# "call/function/builtin/range" is not inside "flow/conditional" anywhere.
dbf = ProgramFilter(db)
dbf.include_taxons([("inside", "call/function/builtin/range", "flow/conditional")])
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == set()
def test_exclude_taxons():
# The taxon "variable/assignment/single" is featured by assignment.py and collatz_print.py. It
# is indirectly featured by fizzbuzz.py (which imports collatz_print.py) and by is_even.py
# (which imports fizzbuzz.py). Therefore, excluding this taxon excludes all four programs.
dbf = ProgramFilter(db)
dbf.exclude_taxons({"variable/assignment/single"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == set()
# "operator/arithmetic/addition" is featured by collatz_print.py, and indirectly by fizzbuzz.py
# and is_even.py. Therefore, excluding this taxon keeps only assignment.py.
dbf = ProgramFilter(db)
dbf.exclude_taxons({"io/standard/print"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py"}
# "flow/conditional" is featured by collatz_print.py, fizzbuzz.py, and indirectly by
# is_even.py. Therefore, excluding this taxon keeps only assignment.py.
dbf = ProgramFilter(db)
dbf.exclude_taxons({"flow/conditional"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py"}
# "type/sequence/string/literal" is featured by fizzbuzz.py, and indirectly by is_even.py.
# Therefore, excluding this taxon keeps only assignment.py and collatz_print.py
dbf = ProgramFilter(db)
dbf.exclude_taxons({"type/sequence/string/literal"})
print(set(dbf.selected_programs.keys()))
assert set(
dbf.selected_programs.keys()) == {"assignment.py", "collatz_print.py"}
def test_include_programs():
# The program collatz_print.py is self-contained. Including it has no other effect on the
# selection.
dbf = ProgramFilter(db)
dbf.include_programs({"collatz_print.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py"}
# The program fizzbuzz.py imports collatz_print.py. Including the former includes both.
dbf = ProgramFilter(db)
dbf.include_programs({"fizzbuzz.py"})
print(set(dbf.selected_programs.keys()))
assert set(
dbf.selected_programs.keys()) == {"fizzbuzz.py", "collatz_print.py"}
# The program is_even.py import collatz_print.py, which imports collatz_print.py. Including the
# former includes the three of them.
dbf = ProgramFilter(db)
dbf.include_programs({"is_even.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {
"is_even.py",
"collatz_print.py",
"fizzbuzz.py",
}
def test_impart_programs():
# The program collatz_print.py is self-contained. Imparting it excludes it from the selection,
# but has no other effect.
dbf = ProgramFilter(db)
dbf.impart_programs({"collatz_print.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {
"assignment.py",
"fizzbuzz.py",
"is_even.py",
}
# The program fizzbuzz.py imports collatz_print.py. Imparting the former implies that the
# latter has already been studied, and must be excluded from the selection.
dbf = ProgramFilter(db)
dbf.impart_programs({"fizzbuzz.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py", "is_even.py"}
# The program is_even.py imports collatz_print.py, which imports collatz_print.py. Imparting
# the former excludes the three of them from the selection.
dbf = ProgramFilter(db)
dbf.impart_programs({"is_even.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py"}
def test_exclude_programs():
dbf = ProgramFilter(db)
programs = {"prg1.py", "prg2.py", "non_existing_program"}
dbf.exclude_programs(programs, follow=True)
print(sorted(dbf.selected_programs))
assert dbf.selected_programs == set(db["programs"]) - set(programs)
def test_exclude_programs():
# The program collatz_print.py is imported by fizzbuzz.py, and indirectly by is_even.py.
# Excluding the former excludes the two latter too.
dbf = ProgramFilter(db)
dbf.exclude_programs({"collatz_print.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py"}
# The program fizzbuzz.py is imported by is_even.py. Excluding the former excludes both.
dbf = ProgramFilter(db)
dbf.exclude_programs({"fizzbuzz.py"})
print(set(dbf.selected_programs.keys()))
assert set(
dbf.selected_programs.keys()) == {"assignment.py", "collatz_print.py"}
# The program is_even.py is not imported. Excluding it has no other effect on the selection.
dbf = ProgramFilter(db)
dbf.exclude_programs({"is_even.py"})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {
"assignment.py",
"collatz_print.py",
"fizzbuzz.py",
}
def test_init():
dbf = ProgramFilter(db)
print(dbf.selected_programs)
assert dbf.selected_programs.keys() == db["programs"].keys()
assert not dbf.imparted_knowledge
def test_exclude_taxa():
# The taxon "var/assignment/explicit/single" is featured by assignment.py and collatz.py. It
# is indirectly featured by fizzbuzz.py (which imports collatz.py) and by is_even.py
# (which imports fizzbuzz.py). Therefore, excluding this taxon excludes all four programs.
dbf = ProgramFilter(db)
dbf.exclude_programs(dbf.programs_of_taxa(
{"var/assignment/explicit/single"}),
follow=True)
print(dbf.selected_programs)
assert dbf.selected_programs == set()
# "operator/arithmetic/addition" is featured by collatz.py, and indirectly by fizzbuzz.py
# and is_even.py. Therefore, excluding this taxon keeps only assignment.py.
dbf = ProgramFilter(db)
dbf.exclude_programs(dbf.programs_of_taxa({"call/function/builtin/print"}),
follow=True)
print(dbf.selected_programs)
assert dbf.selected_programs == {"assignment.py"}
# "flow/conditional" is featured by collatz.py, fizzbuzz.py, and indirectly by
# is_even.py. Therefore, excluding this taxon keeps only assignment.py.
dbf = ProgramFilter(db)
dbf.exclude_programs(dbf.programs_of_taxa({"flow/conditional"}),
follow=True)
print(dbf.selected_programs)
assert dbf.selected_programs == {"assignment.py"}
# "type/sequence/string/literal" is featured by fizzbuzz.py, and indirectly by is_even.py.
# Therefore, excluding this taxon keeps only assignment.py and collatz.py
dbf = ProgramFilter(db)
dbf.exclude_programs(dbf.programs_of_taxa({"type/sequence/string/literal"
}),
follow=True)
print(dbf.selected_programs)
assert dbf.selected_programs == {"assignment.py", "collatz.py"}
def test_preprocess_programs():
dbf = ProgramFilter(db)
programs = dbf.preprocess_programs([r"prg[1-3]\.py", r"prg[7-9]\.py", "non_matching_pattern"])
print(sorted(programs))
assert programs == ["prg1.py", "prg2.py", "prg3.py", "prg7.py", "prg8.py", "prg9.py"]
def test_include_programs():
dbf = ProgramFilter(db)
programs = {"prg1.py", "prg2.py", "non_existing_program"}
dbf.include_programs(programs)
print(sorted(dbf.selected_programs))
assert dbf.selected_programs.keys() == {"prg1.py", "prg2.py"}
def test_exclude_taxons():
# The taxon "variable/assignment/single" is featured by assignment.py and collatz_print.py. In
# collatz_print.py, it appears after a taxon "io/standard/print". Consequently, it should be excluded
# from the results, along with the programs which import it: fizzbuzz.py and is_even.py.
dbf = ProgramFilter(db)
dbf.exclude_taxons({("after", "variable/assignment/single", "io/standard/print")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py"}
# "operator/arithmetic/addition" and "operator/arithmetic/multiplication" are both featured on
# the same line of collatz_print.py, and indirectly by fizzbuzz.py and is_even.py. Therefore,
# excluding this taxon keeps only assignment.py.
dbf = ProgramFilter(db)
dbf.exclude_taxons(
{("equals", "operator/arithmetic/addition", "operator/arithmetic/multiplication")}
)
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"assignment.py"}
# "test/equality" is inside "subroutine/function" in is_even.py, which is not imported anywhere.
dbf = ProgramFilter(db)
dbf.exclude_taxons({("inside", "test/equality", "subroutine/function")})
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {"collatz_print.py", "assignment.py", "fizzbuzz.py"}
# "call/function/builtin/range" is not inside "flow/conditional" anywhere.
dbf = ProgramFilter(db)
dbf.exclude_taxons([("inside", "call/function/builtin/range", "flow/conditional")])
print(set(dbf.selected_programs.keys()))
assert set(dbf.selected_programs.keys()) == {
"is_even.py",
"fizzbuzz.py",
"assignment.py",
"collatz_print.py",
}