starred/aryx
1
0
Fork 0
mirror of https://github.com/davidkaya/aryx.git synced 2026-08-04 19:08:40 +02:00
Code Issues 3 Packages Projects Releases 10 Wiki Activity

fix: remove streaming text boundary heuristics that injected spurious spaces

Browse Source 
The StreamingTextMerger (C#) and mergeStreamingText (TypeScript) contained
custom word-boundary heuristics that inserted spaces between streaming deltas
based on incorrect assumptions. Two bugs caused mid-word spaces:

1. Unconditional space insertion when incoming delta started with uppercase or
   digit characters, breaking acronyms (MDA -> M DA, UAG -> UA G)
2. LooksLikeWordBoundary using global token counts that fired on nearly every
   mid-word token split (writable -> wr itable, fragile -> frag ile)

Both upstream projects (Copilot SDK and Agent Framework) use simple string
concatenation for delta accumulation. LLM streaming tokens natively include
whitespace when needed, so no boundary detection is required.

Removed all boundary heuristic methods and replaced with plain concatenation.
Kept overlap detection and snapshot handling for event redelivery scenarios.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This commit is contained in:
David Kaya
2026-04-15 18:55:34 +02:00
co-authored by Copilot
parent 418946b854
commit c7910e6e4b
4 changed files with 79 additions and 166 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/shared/utils/streamingText.ts
+1 -47
View File
@@ -5,52 +5,6 @@ function tokenize(value: string): string[] {
.filter((token) => token.length > 0);
}
function shouldInsertNewlineBoundary(current: string, incoming: string): boolean {
if (current.endsWith('\n')) {
return false;
}
return /^(?:#{1,6}\s|[-*+]\s|\d+\.\s|>\s|```)/.test(incoming.trimStart());
}
function shouldInsertSpaceBoundary(current: string, incoming: string): boolean {
const lastChar = current.at(-1);
const firstChar = incoming[0];
if (!lastChar || !firstChar) {
return false;
}
if (/\s/.test(lastChar) || /\s/.test(firstChar) || /[([{/"'`]/.test(lastChar)) {
return false;
}
if (/^[.,!?;:%)\]}]/.test(firstChar)) {
return false;
}
if (/^[*_`~\[]/.test(firstChar) || /^[A-Z0-9]/.test(firstChar)) {
return true;
}
const currentTokens = tokenize(current);
const incomingTokens = tokenize(incoming);
const firstIncomingToken = incomingTokens[0] ?? '';
return currentTokens.length >= 2 && incomingTokens.length >= 2 && firstIncomingToken.length >= 2;
}
function appendWithNaturalBoundary(current: string, incoming: string): string {
if (shouldInsertNewlineBoundary(current, incoming)) {
return `${current}\n${incoming}`;
}
if (shouldInsertSpaceBoundary(current, incoming)) {
return `${current} ${incoming}`;
}
return current + incoming;
}
function computeSuffixPrefixOverlap(current: string, incoming: string): number {
const maxOverlap = Math.min(current.length, incoming.length);
for (let length = maxOverlap; length > 0; length -= 1) {
@@ -109,5 +63,5 @@ export function mergeStreamingText(current: string, incoming: string): string {
return incoming;
}
return appendWithNaturalBoundary(current, incoming);
return current + incoming;
}