Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix the annotated_domains_for_relation returned by SchemaHelper::make_hirm_schema #195

Merged
merged 2 commits into from
Sep 11, 2024
Merged

Fix the annotated_domains_for_relation returned by SchemaHelper::make_hirm_schema #195

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0f8772a
Fix annotated_domains_for_relation
ThomasColthurst Sep 10, 2024
348c81e
Fix tests
ThomasColthurst Sep 10, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions cxx/pclean/pclean_lib.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ T_observations translate_observations(
const T_relation& trel = schema.at(col_name);
size_t num_domains = std::visit([&](const auto &r) {
return r.domains.size();}, trel);
assert(num_domains == annotated_domains_for_relations.at(col_name).size());

for (size_t i = 0; i < col.second.size(); ++i) {
const std::string& val = col.second[i];
Expand Down
4 changes: 2 additions & 2 deletions cxx/pclean/pclean_lib_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,8 +32,8 @@ BOOST_AUTO_TEST_CASE(test_translate_observations) {
T_noisy_relation{{"dCounty", "dObs"}, true, EmissionSpec("bigram"), "County:state"}}};

std::map<std::string, std::vector<std::string>> annotated_domains_for_relations;
annotated_domains_for_relations["Room Type"] = {"county:County", "Obs"};
annotated_domains_for_relations["Monthly Rent"] = {"county:County", "Obs"};
annotated_domains_for_relations["Room Type"] = {"Obs"};
annotated_domains_for_relations["Monthly Rent"] = {"Obs"};
annotated_domains_for_relations["County"] = {"county:County", "Obs"};
annotated_domains_for_relations["State"] = {"county:County", "Obs"};

Expand Down
51 changes: 12 additions & 39 deletions cxx/pclean/schema_helper.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,15 +47,6 @@ void PCleanSchemaHelper::compute_domains_for(const std::string& name) {
annotated_domains[name] = annotated_ds;
}

std::string make_prefix_path(
std::vector<std::string>& var_names, size_t index) {
std::string s;
for (size_t i = index; i < var_names.size(); ++i) {
s += var_names[i] + ":";
}
return s;
}

void PCleanSchemaHelper::make_relations_for_queryfield(
const QueryField& f, const PCleanClass& record_class, T_schema* tschema,
std::map<std::string, std::vector<std::string>>
Expand Down Expand Up @@ -108,30 +99,34 @@ void PCleanSchemaHelper::make_relations_for_queryfield(
// Handle only_final_emissions == true.
if (only_final_emissions) {
std::vector<std::string> noisy_domains = domains[class_names.back()];
std::vector<std::string> adfr = annotated_domains[class_names.back()];
for (int i = class_names.size() - 2; i >= 0; --i) {
noisy_domains.push_back(class_names[i]);
for (size_t j = 0; j < adfr.size(); ++j) {
adfr[j] = var_names[i] + ":" + adfr[j];
}
adfr.push_back(class_names[i]);
}
T_noisy_relation tnr = get_emission_relation(
std::get<ScalarVar>(last_var.spec),
noisy_domains,
base_relation_name);
tnr.is_observed = true;
(*tschema)[f.name] = tnr;
std::string path_prefix = make_prefix_path(var_names, 0);
std::vector<std::string> reordered_annotated_domains = reorder_domains(
annotated_domains[record_class.name],
annotated_domains[record_class.name],
path_prefix);
(*annotated_domains_for_relation)[f.name] = reordered_annotated_domains;
(*annotated_domains_for_relation)[f.name] = adfr;
return;
}

// Handle only_final_emissions == false.
std::string& previous_relation = base_relation_name;
std::vector<std::string> current_domains = domains[class_names.back()];
std::vector<std::string> adfr = annotated_domains[class_names.back()];
for (int i = f.class_path.size() - 2; i >= 0; --i) {
std::string path_prefix = make_prefix_path(var_names, i);
current_domains.push_back(class_names[i]);
for (size_t j = 0; j < adfr.size(); ++j) {
adfr[j] = var_names[i] + ":" + adfr[j];
}
adfr.push_back(class_names[i]);
T_noisy_relation tnr = get_emission_relation(
std::get<ScalarVar>(last_var.spec),
current_domains,
Expand All @@ -148,30 +143,8 @@ void PCleanSchemaHelper::make_relations_for_queryfield(
}
(*tschema)[rel_name] = tnr;
previous_relation = rel_name;
std::vector<std::string> reordered_annotated_domains = reorder_domains(
annotated_domains[class_names[i]],
annotated_domains[class_names[i]],
path_prefix);
(*annotated_domains_for_relation)[rel_name] = reordered_annotated_domains;
}
}

std::vector<std::string> reorder_domains(
const std::vector<std::string>& original_domains,
const std::vector<std::string>& annotated_ds,
const std::string& prefix) {
std::vector<std::string> output_domains;
for (size_t i = 0; i < original_domains.size(); ++i) {
if (annotated_ds[i].starts_with(prefix)) {
output_domains.push_back(original_domains[i]);
}
}
for (size_t i = 0; i < original_domains.size(); ++i) {
if (!annotated_ds[i].starts_with(prefix)) {
output_domains.push_back(original_domains[i]);
}
(*annotated_domains_for_relation)[rel_name] = adfr;
}
return output_domains;
}

T_schema PCleanSchemaHelper::make_hirm_schema(
Expand Down
7 changes: 0 additions & 7 deletions cxx/pclean/schema_helper.hh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,10 +44,3 @@ class PCleanSchemaHelper {
std::map<std::string, std::vector<std::string>> domains;
std::map<std::string, std::vector<std::string>> annotated_domains;
};

// Returns original_domains, but with the elements corresponding to
// annotated_ds elements that start with prefix moved to the front.
std::vector<std::string> reorder_domains(
const std::vector<std::string>& original_domains,
const std::vector<std::string>& annotated_ds,
const std::string& prefix);
28 changes: 1 addition & 27 deletions cxx/pclean/schema_helper_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -158,8 +158,7 @@ BOOST_AUTO_TEST_CASE(test_make_relations_for_queryfield) {
BOOST_TEST(!std::get<T_noisy_relation>(tschema["Physician::School"]).is_observed);

std::vector<std::string> expected_adfr = {
"physician:school:School", "location:city:City",
"location:Practice", "physician:Physician", "Record"};
"physician:school:School", "physician:Physician", "Record"};
BOOST_TEST(annotated_domains_for_relation["School"] == expected_adfr,
tt::per_element());
}
Expand Down Expand Up @@ -336,31 +335,6 @@ BOOST_AUTO_TEST_CASE(test_make_hirm_schema_only_final_emissions) {
BOOST_TEST(nr5.domains == expected_domains, tt::per_element());
}

BOOST_AUTO_TEST_CASE(test_reorder_domains) {
std::vector<std::string> origs = {"0", "1", "2", "3", "4", "5", "6", "7"};
std::vector<std::string> annotated = {
"000", "001", "010", "011", "100", "101", "110", "111"};

std::vector<std::string> expected = {
"6", "7", "0", "1", "2", "3", "4", "5"};
BOOST_TEST(reorder_domains(origs, annotated, "11") == expected);

expected = {"4", "5", "6", "7", "0", "1", "2", "3"};
BOOST_TEST(reorder_domains(origs, annotated, "1") == expected);

origs = {
"republic_of_ireland", "northern_ireland", "england", "scotland", "wales"};
annotated = {
"ireland:republic_of_ireland",
"ireland:uk:northern_ireland",
"great_britain:uk:england",
"great_britain:uk:scotland",
"great_britain:uk:wales"};
expected = {
"northern_ireland", "republic_of_ireland", "england", "scotland", "wales"};
BOOST_TEST(reorder_domains(origs, annotated, "ireland:uk:") == expected);
}

BOOST_AUTO_TEST_CASE(test_record_class_is_clean) {
std::stringstream ss2(R"""(
class Record
Expand Down