Skip to content

CNDB-13408 Improve BM25 test with better logic (#1963) #1977

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 1 commit into from
Sep 4, 2025
Merged

CNDB-13408 Improve BM25 test with better logic (#1963) #1977

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
88 changes: 47 additions & 41 deletions test/unit/org/apache/cassandra/index/sai/cql/BM25Test.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -870,26 +870,26 @@ public void testOrderingSeveralSegments() throws Throwable

public final static Object[][] DATASET =
{
{ 0, "Climate", 5, "Climate change is a pressing issue. Climate patterns are shifting globally. Scientists study climate data daily.", 1 },
{ 1, "Technology", 3, "Technology is advancing. New technology in AI and robotics is groundbreaking.", 1 },
{ 2, "Economy", 4, "The economy is recovering. Economy experts are optimistic. However, the global economy still faces risks.", 1 },
{ 3, "Health", 3, "Health is wealth. Health policies need to be improved to ensure better public health outcomes.", 1 },
{ 4, "Education", 2, "Education is the foundation of success. Online education is booming.", 4 },
{ 5, "Climate", 4, "Climate and health are closely linked. Climate affects air quality and health outcomes.", 2 },
{ 6, "Education", 3, "Technology and education go hand in hand. EdTech is revolutionizing education through technology.", 3 },
{ 7, "Economy", 3, "The global economy is influenced by technology. Fintech is a key part of the economy today.", 2 },
{ 8, "Health", 3, "Education and health programs must be prioritized. Health education is vital in schools.", 2 },
{ 9, "Mixed", 3, "Technology, economy, and education are pillars of development.", 2 },
{ 10, "Climate", 5, "Climate climate climate. It's everywhere. Climate drives political and economic decisions.", 1 },
{ 11, "Health", 2, "Health concerns rise with climate issues. Health organizations are sounding the alarm.", 2 },
{ 12, "Economy", 3, "The economy is fluctuating. Uncertainty looms over the economy.", 1 },
{ 13, "Health", 3, "Cutting-edge technology is transforming healthcare. Healthtech merges health and technology.", 1 },
{ 14, "Education", 2, "Education reforms are underway. Education experts suggest holistic changes.", 1 },
{ 15, "Climate", 4, "Climate affects the economy and health. Climate events cost billions annually.", 1 },
{ 16, "Technology", 3, "Technology is the backbone of the modern economy. Without technology, economic growth stagnates.", 2 },
{ 17, "Health", 2, "Health is discussed less than economy or climate or technology, but health matters deeply.", 1 },
{ 18, "Climate", 5, "Climate change, climate policies, climate research—climate is the buzzword of our time.", 2 },
{ 19, "Mixed", 3, "Investments in education and technology will shape the future of the global economy.", 1 }
{ 0, "Climate", 5, "Climate change is a pressing issue. Climate patterns are shifting globally. Scientists study climate data daily." },
{ 1, "Technology", 3, "Technology is advancing. New technology in AI and robotics is groundbreaking." },
{ 2, "Economy", 4, "The economy is recovering. Economy experts are optimistic. However, the global economy still faces risks." },
{ 3, "Health", 3, "Health is wealth. Health policies need to be improved to ensure better public health outcomes." },
{ 4, "Education", 2, "Education is the foundation of success. Online education is booming." },
{ 5, "Climate", 4, "Climate and health are closely linked. Climate affects air quality and health outcomes." },
{ 6, "Education", 3, "Technology and education go hand in hand. EdTech is revolutionizing education through technology." },
{ 7, "Economy", 3, "The global economy is influenced by technology. Fintech is a key part of the economy today." },
{ 8, "Health", 3, "Education and health programs must be prioritized. Health education is vital in schools." },
{ 9, "Mixed", 3, "Technology, economy, and education are pillars of development." },
{ 10, "Climate", 5, "Climate climate climate. It's everywhere. Climate drives political and economic decisions." },
{ 11, "Health", 2, "Health concerns rise with climate issues. Health organizations are sounding the alarm." },
{ 12, "Economy", 3, "The economy is fluctuating. Uncertainty looms over the economy." },
{ 13, "Health", 3, "Cutting-edge technology is transforming healthcare. Healthtech merges health and technology." },
{ 14, "Education", 2, "Education reforms are underway. Education experts suggest holistic changes." },
{ 15, "Climate", 4, "Climate affects the economy and health. Climate events cost billions annually." },
{ 16, "Technology", 3, "Technology is the backbone of the modern economy. Without technology, economic growth stagnates." },
{ 17, "Health", 2, "Health is discussed less than economy or climate or technology, but health matters deeply." },
{ 18, "Climate", 5, "Climate change, climate policies, climate research—climate is the buzzword of our time." },
{ 19, "Mixed", 3, "Investments in education and technology will shape the future of the global economy." }
};

private void analyzeDataset(String term)
Expand Down Expand Up @@ -932,36 +932,42 @@ private void insertPrimitiveData(int start, int end)

public static void insertCollectionData(SAITester tester)
{
int setsize = 1;
for (int row = 0; row < DATASET.length; row++)
{
var set = new HashSet<String>();
for (int j = 0; j < setsize; j++)
set.add((String) DATASET[row - j][3]);
if (setsize >= 3)
setsize -= 2;
else
setsize++;
var map = new HashMap<Integer, String>();
var map_text = new HashMap<String, String>();
for (int j = 0; j <= row && j < 3; j++)
{
map.putIfAbsent((Integer) DATASET[row - j][2], (String) DATASET[row - j][1]);
map_text.putIfAbsent((String) DATASET[row - j][1], (String) DATASET[row - j][3]);
}

tester.execute(
"INSERT INTO %s (id, category, score, body, bodyset, map_category, map_body) " +
"VALUES (?, ?, ?, ?, ?, ?, ?)",
DATASET[row][0],
DATASET[row][1],
DATASET[row][2],
DATASET[row][3],
set,
map,
map_text
generateSetColumnValue(row),
generateMapValueIntKey(row),
generateMapValueTextKey(row)
);
}
}

private static HashSet<String> generateSetColumnValue(int row)
{
HashSet<String> setColumnValue = new HashSet<>();
for (int i = 0; i < row % 3 + 1; i++)
setColumnValue.add((String) DATASET[row - i][3]);
return setColumnValue;
}

private static HashMap<Integer, String> generateMapValueIntKey(int row)
{
HashMap<Integer, String> mapIntKey = new HashMap<>();
for (int j = 0; j <= row && j < 3; j++)
mapIntKey.putIfAbsent((Integer) DATASET[row - j][2], (String) DATASET[row - j][1]);
return mapIntKey;
}

private static HashMap<String, String> generateMapValueTextKey(int row)
{
HashMap<String, String> mapTextKey = new HashMap<>();
for (int j = 0; j <= row && j < 3; j++)
mapTextKey.putIfAbsent((String) DATASET[row - j][1], (String) DATASET[row - j][3]);
return mapTextKey;
}

private void executeQuery(List<Integer> expected, String query, Object... values) throws Throwable
Expand Down