Skip to content

.Net: Fix TextChunker orphan chunk token counting #14013

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

Open
MukundaKatta wants to merge 1 commit into
base: main
Choose a base branch
from
+18 −11
Open

.Net: Fix TextChunker orphan chunk token counting #14013

Show file tree
Hide file tree
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
14 changes: 3 additions & 11 deletions dotnet/src/SemanticKernel.Core/Text/TextChunker.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,6 @@ private sealed class StringListWithTokenCount(TextChunker.TokenCounter? tokenCou
/// <returns>The number of tokens in the input string.</returns>
public delegate int TokenCounter(string input);

private static readonly char[] s_spaceChar = [' '];
private static readonly string?[] s_plaintextSplitOptions = ["\n", ".。．", "?!", ";", ":", ",，、", ")]}", " ", "-", null];
private static readonly string?[] s_markdownSplitOptions = [".\u3002\uFF0E", "?!", ";", ":", ",\uFF0C\u3001", ")]}", " ", "-", "\n\r", null];

Expand Down Expand Up @@ -192,18 +191,11 @@ private static List<string> ProcessParagraphs(List<string> paragraphs, int adjus

if (GetTokenCount(lastParagraph, tokenCounter) < adjustedMaxTokensPerParagraph / 4)
{
var lastParagraphTokens = lastParagraph.Split(s_spaceChar, StringSplitOptions.RemoveEmptyEntries);
var secondLastParagraphTokens = secondLastParagraph.Split(s_spaceChar, StringSplitOptions.RemoveEmptyEntries);
var mergedParagraph = $"{secondLastParagraph} {lastParagraph}";

var lastParagraphTokensCount = lastParagraphTokens.Length;
var secondLastParagraphTokensCount = secondLastParagraphTokens.Length;

if (lastParagraphTokensCount + secondLastParagraphTokensCount <= adjustedMaxTokensPerParagraph)
if (GetTokenCount(mergedParagraph, tokenCounter) <= adjustedMaxTokensPerParagraph)
{
var newSecondLastParagraph = string.Join(" ", secondLastParagraphTokens);
var newLastParagraph = string.Join(" ", lastParagraphTokens);

paragraphs[paragraphs.Count - 2] = $"{newSecondLastParagraph} {newLastParagraph}";
paragraphs[paragraphs.Count - 2] = mergedParagraph;
paragraphs.RemoveAt(paragraphs.Count - 1);
}
}
Expand Down
15 changes: 15 additions & 0 deletions dotnet/src/SemanticKernel.UnitTests/Text/TextChunkerTests.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -819,4 +819,19 @@ public void SplitPlainTextParagraphsSplitsWhenExceedingTokenLimit()
Assert.Contains("Second line", combined);
Assert.Contains("Third line", combined);
}

[Fact]
public void SplitPlainTextParagraphsDoesNotMergeOrphanChunkBeyondTokenLimit()
{
var lines = new[]
{
new string('a', 45),
new string('b', 10),
};

var result = TextChunker.SplitPlainTextParagraphs(lines, 52, tokenCounter: input => input.Length);

Assert.Equal(2, result.Count);
Assert.All(result, paragraph => Assert.True(paragraph.Length <= 52));
}
}
Loading