Skip to content

Support multiple columns in fulltext index #24439

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

Merged
merged 4 commits into from
Sep 9, 2025
Merged

Support multiple columns in fulltext index #24439

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
5a188c2
Support multiple columns in fulltext index
kunga Sep 8, 2025
ca1ada5
better structure inside FulltextIndexSettings
kunga Sep 9, 2025
ac8a824
fix typo
kunga Sep 9, 2025
fb8f9d8
cr: add multiple columns tests
kunga Sep 9, 2025
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
12 changes: 12 additions & 0 deletions ydb/core/tx/schemeshard/schemeshard_utils.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -194,6 +194,18 @@ bool CommonCheck(const TTableDesc& tableDesc, const NKikimrSchemeOp::TIndexCreat
error = TStringBuilder() << "fulltext index can only have a single key text column";
return false;
}
if (indexDesc.GetFulltextIndexDescription().GetSettings().Getcolumns().size() != 1) {
status = NKikimrScheme::EStatus::StatusInvalidParameter;
error = TStringBuilder() << "fulltext index should have single '" << indexKeys.KeyColumns.at(0) << "' column settings"
<< " but have " << indexDesc.GetFulltextIndexDescription().GetSettings().Getcolumns().size() << " of them";
return false;
}
if (indexDesc.GetFulltextIndexDescription().GetSettings().Getcolumns().at(0).Getcolumn() != indexKeys.KeyColumns.at(0)) {
status = NKikimrScheme::EStatus::StatusInvalidParameter;
error = TStringBuilder() << "fulltext index should have '" << indexKeys.KeyColumns.at(0) << "' column settings"
<< " but have '" << indexDesc.GetFulltextIndexDescription().GetSettings().Getcolumns().at(0).Getcolumn() << "' column settings";
return false;
}

const TString& indexColumnName = indexKeys.KeyColumns.back();
Y_ABORT_UNLESS(baseColumnTypes.contains(indexColumnName));
Expand Down
42 changes: 30 additions & 12 deletions ydb/core/tx/schemeshard/ut_helpers/ls_checks.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
#include "ls_checks.h"

#include <google/protobuf/text_format.h>
#include <ydb/public/api/protos/ydb_cms.pb.h>
#include <ydb/public/api/protos/ydb_coordination.pb.h>
#include <ydb/public/lib/scheme_types/scheme_type_id.h>
Expand Down Expand Up @@ -914,20 +915,37 @@ TCheckFunc KMeansTreeDescription(Ydb::Table::VectorIndexSettings_Metric metric,

TCheckFunc SpecializedIndexDescription(const TString& proto) {
return [=] (const NKikimrScheme::TEvDescribeSchemeResult& record) {
TString actual;
switch (record.GetPathDescription().GetTableIndex().GetSpecializedIndexDescriptionCase()) {
case NKikimrSchemeOp::TIndexDescription::kVectorIndexKmeansTreeDescription:
actual = record.GetPathDescription().GetTableIndex().GetVectorIndexKmeansTreeDescription().GetSettings().ShortDebugString();
break;
case NKikimrSchemeOp::TIndexDescription::kFulltextIndexDescription:
actual = record.GetPathDescription().GetTableIndex().GetFulltextIndexDescription().GetSettings().ShortDebugString();
break;
case NKikimrSchemeOp::TIndexDescription::SPECIALIZEDINDEXDESCRIPTION_NOT_SET:
actual = "SPECIALIZEDINDEXDESCRIPTION_NOT_SET";
break;
case NKikimrSchemeOp::TIndexDescription::kVectorIndexKmeansTreeDescription: {
auto actual = record.GetPathDescription().GetTableIndex().GetVectorIndexKmeansTreeDescription().GetSettings();
Ydb::Table::KMeansTreeSettings expected;
UNIT_ASSERT(google::protobuf::TextFormat::ParseFromString(proto, &expected));
UNIT_ASSERT_C(google::protobuf::util::MessageDifferencer::Equals(actual, expected),
TStringBuilder() << "Expected"
<< expected.ShortDebugString()
<< " but got "
<< actual.ShortDebugString());
break;
}
case NKikimrSchemeOp::TIndexDescription::kFulltextIndexDescription: {
auto actual = record.GetPathDescription().GetTableIndex().GetFulltextIndexDescription().GetSettings();
Ydb::Table::FulltextIndexSettings expected;
UNIT_ASSERT(google::protobuf::TextFormat::ParseFromString(proto, &expected));
UNIT_ASSERT_C(google::protobuf::util::MessageDifferencer::Equals(actual, expected),
TStringBuilder() << "Expected"
<< expected.ShortDebugString()
<< " but got "
<< actual.ShortDebugString());
break;
}
case NKikimrSchemeOp::TIndexDescription::SPECIALIZEDINDEXDESCRIPTION_NOT_SET: {
UNIT_ASSERT_C(proto == "SPECIALIZEDINDEXDESCRIPTION_NOT_SET",
TStringBuilder() << "Expected"
<< proto
<< " but got SPECIALIZEDINDEXDESCRIPTION_NOT_SET");
break;
}
}

UNIT_ASSERT_VALUES_EQUAL(actual, proto);
};
}

Expand Down
258 changes: 218 additions & 40 deletions ydb/core/tx/schemeshard/ut_index/ut_fulltext_index.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,23 +18,38 @@ Y_UNIT_TEST_SUITE(TFulltextIndexTests) {
TTestEnv env(runtime);
ui64 txId = 100;

TestCreateIndexedTable(runtime, ++txId, "/MyRoot", R"(
TString fulltextSettings = R"(
layout: FLAT
columns: {
column: "text"
analyzers: {
tokenizer: STANDARD
use_filter_ngram: true
filter_ngram_max_length: 42
}
}
)";
TestCreateIndexedTable(runtime, ++txId, "/MyRoot", Sprintf(R"(
TableDescription {
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
}
IndexDescription {
Name: "idx_fulltext"
KeyColumnNames: ["text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: { Settings: { layout: FLAT, tokenizer: STANDARD, use_filter_ngram: true, filter_ngram_max_length: 42 } }
Name: "idx_fulltext"
KeyColumnNames: ["text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: {
Settings: {
%s
}
}
}
)");
)", fulltextSettings.c_str()));
env.TestWaitNotification(runtime, txId);

NKikimrSchemeOp::TDescribeOptions opts;
Expand All @@ -49,7 +64,7 @@ Y_UNIT_TEST_SUITE(TFulltextIndexTests) {
NLs::IndexState(NKikimrSchemeOp::EIndexStateReady),
NLs::IndexKeys({"text"}),
NLs::IndexDataColumns({"covered"}),
NLs::SpecializedIndexDescription("layout: FLAT tokenizer: STANDARD use_filter_ngram: true filter_ngram_max_length: 42"),
NLs::SpecializedIndexDescription(fulltextSettings),
NLs::ChildrenCount(1),
});

Expand All @@ -70,23 +85,91 @@ Y_UNIT_TEST_SUITE(TFulltextIndexTests) {
TTestEnv env(runtime);
ui64 txId = 100;

TestCreateIndexedTable(runtime, ++txId, "/MyRoot", R"(
TString fulltextSettings = R"(
layout: FLAT
columns: {
column: "text"
analyzers: {
tokenizer: STANDARD
use_filter_ngram: true
filter_ngram_max_length: 42
}
}
)";
TestCreateIndexedTable(runtime, ++txId, "/MyRoot", Sprintf(R"(
TableDescription {
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
}
IndexDescription {
Name: "idx_fulltext"
KeyColumnNames: [ "another", "text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: {
Settings: {
%s
}
}
}
)", fulltextSettings.c_str()), {NKikimrScheme::StatusInvalidParameter});
env.TestWaitNotification(runtime, txId);

TestDescribeResult(DescribePrivatePath(runtime, "/MyRoot/texts/idx_fulltext"),{
NLs::PathNotExist,
});
}

Y_UNIT_TEST(CreateTableMultipleColumns) { // not supported for now, maybe later
TTestBasicRuntime runtime;
TTestEnv env(runtime);
ui64 txId = 100;

TString fulltextSettings = R"(
layout: FLAT
columns: {
column: "text1"
analyzers: {
tokenizer: STANDARD
use_filter_ngram: true
filter_ngram_max_length: 42
}
}
columns: {
column: "text2"
analyzers: {
tokenizer: STANDARD
use_filter_ngram: true
filter_ngram_max_length: 42
}
}
)";
TestCreateIndexedTable(runtime, ++txId, "/MyRoot", Sprintf(R"(
TableDescription {
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: [ "id"]
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text1" Type: "String" }
Columns { Name: "text2" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
}
IndexDescription {
Name: "idx_fulltext"
KeyColumnNames: [ "another", "text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: { Settings: { layout: FLAT, tokenizer: STANDARD, use_filter_ngram: true, filter_ngram_max_length: 42 } }
Name: "idx_fulltext"
KeyColumnNames: ["text1", "text2"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: {
Settings: {
%s
}
}
}
)", {NKikimrScheme::StatusInvalidParameter});
)", fulltextSettings.c_str()), {NKikimrScheme::StatusInvalidParameter});
env.TestWaitNotification(runtime, txId);

TestDescribeResult(DescribePrivatePath(runtime, "/MyRoot/texts/idx_fulltext"),{
Expand All @@ -99,23 +182,118 @@ Y_UNIT_TEST_SUITE(TFulltextIndexTests) {
TTestEnv env(runtime);
ui64 txId = 100;

TestCreateIndexedTable(runtime, ++txId, "/MyRoot", R"(
TString fulltextSettings = R"(
layout: FLAT
columns: {
column: "text"
analyzers: {
tokenizer: STANDARD
use_filter_ngram: true
filter_ngram_max_length: 42
}
}
)";
TestCreateIndexedTable(runtime, ++txId, "/MyRoot", Sprintf(R"(
TableDescription {
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "Uint64" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
}
IndexDescription {
Name: "idx_fulltext"
KeyColumnNames: ["text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: {
Settings: {
%s
}
}
}
)", fulltextSettings.c_str()), {NKikimrScheme::StatusInvalidParameter});
env.TestWaitNotification(runtime, txId);

TestDescribeResult(DescribePrivatePath(runtime, "/MyRoot/texts/idx_fulltext"),{
NLs::PathNotExist,
});
}

Y_UNIT_TEST(CreateTableColumnsMismatch) {
TTestBasicRuntime runtime;
TTestEnv env(runtime);
ui64 txId = 100;

TString fulltextSettings = R"(
layout: FLAT
columns: {
column: "text_wrong"
analyzers: {
tokenizer: STANDARD
use_filter_ngram: true
filter_ngram_max_length: 42
}
}
)";
TestCreateIndexedTable(runtime, ++txId, "/MyRoot", Sprintf(R"(
TableDescription {
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
}
IndexDescription {
Name: "idx_fulltext"
KeyColumnNames: ["text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: {
Settings: {
%s
}
}
}
)", fulltextSettings.c_str()), {NKikimrScheme::StatusInvalidParameter});
env.TestWaitNotification(runtime, txId);

TestDescribeResult(DescribePrivatePath(runtime, "/MyRoot/texts/idx_fulltext"),{
NLs::PathNotExist,
});
}

Y_UNIT_TEST(CreateTableNoColumnsSettings) {
TTestBasicRuntime runtime;
TTestEnv env(runtime);
ui64 txId = 100;

TString fulltextSettings = R"(
layout: FLAT
)";
TestCreateIndexedTable(runtime, ++txId, "/MyRoot", Sprintf(R"(
TableDescription {
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "Uint64" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
Name: "texts"
Columns { Name: "id" Type: "Uint64" }
Columns { Name: "text" Type: "String" }
Columns { Name: "covered" Type: "String" }
Columns { Name: "another" Type: "Uint64" }
KeyColumnNames: ["id"]
}
IndexDescription {
Name: "idx_fulltext"
KeyColumnNames: ["text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: { Settings: { layout: FLAT, tokenizer: STANDARD, use_filter_ngram: true, filter_ngram_max_length: 42 } }
Name: "idx_fulltext"
KeyColumnNames: ["text"]
DataColumnNames: ["covered"]
Type: EIndexTypeGlobalFulltext
FulltextIndexDescription: {
Settings: {
%s
}
}
}
)", {NKikimrScheme::StatusInvalidParameter});
)", fulltextSettings.c_str()), {NKikimrScheme::StatusInvalidParameter});
env.TestWaitNotification(runtime, txId);

TestDescribeResult(DescribePrivatePath(runtime, "/MyRoot/texts/idx_fulltext"),{
Expand Down
Loading
Loading