getTableMemoryUsage(inputTable)
getTableMemoryUsage(outputTable)
tableSize = 10000
function repeatString(val, numRows, colname) {
const vector = arrow.vectorFromArray(
Array.from({ length: numRows }).fill(val),
new arrow.Dictionary(new arrow.Float32, new arrow.Uint8),
);
return {
[colname]: vector,
[`${colname}_codes`]: new Uint8Array(Array.from({ length: numRows }).fill(0)),
}
}
inputTable = arrow.makeTable({
...repeatString(32.1, tableSize, "foo")
})
db = {
return tableSize && DuckDBClient.of({});
}
conn = await db._db.connect()
insertionComplete = {
const buffer = arrow.tableToIPC(inputTable, 'stream');
conn.insertArrowFromIPCStream(buffer, {
name: "arrowTable",
schema: "main"
});
return true;
}
outputTable = tableSize && insertionComplete && conn.query("SELECT foo_codes FROM arrowTable")
outputTable.batches[0].data.children[0].valueOffsets
inputTable.getChild("foo")
inputTable.getChild("foo").get(0)
outputTable.getChild("foo_codes")
outputTable.getChild("foo_codes").get(0)
inputTable.batches.length === 1
outputTable.batches.length === 1
outputTable.batches[0].data.children[0]
inputTable.batches[0].data.children[0].values
outputTable.batches[0].data.children[0].values
inputTable.batches[0].data.children[0].dictionary.data[0].values
outputTable.batches[0].data.children[0].values
outputTable.batches[0].data.children[0].valueOffsets
getTableMemoryUsage = (t) => {
const typedArrs = [];
t.batches.forEach(b => {
if(b.data.children[0].dictionary === undefined) {
typedArrs.push(b.data.children[0].values);
if(b.data.children[0].valueOffsets) {
typedArrs.push(b.data.children[0].valueOffsets);
}
} else {
typedArrs.push(b.data.children[0].values);
typedArrs.push(b.data.children[0].dictionary.data[0].values);
}
})
return typedArrs.map(a => a.length * a.BYTES_PER_ELEMENT).reduce((a, h) => a + h, 0);
}
// Reference: https://github.com/observablehq/feedback/issues/623#issuecomment-1999937586
arrow = await import('https://cdn.observableusercontent.com/npm/apache-arrow@11.0.0/+esm')
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.