LLM/moltbot
1
0
Fork 0
mirror of https://github.com/moltbot/moltbot.git synced 2026-03-29 16:54:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

refactor(agents): dedupe sentence break scanning

Browse Source
This commit is contained in:
Peter Steinberger
2026-02-15 14:12:25 +00:00
parent 0c29ffac09
commit 0f86ee531b
1 changed files with 24 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

50
src/agents/pi-embedded-block-chunker.ts
View File

@@ -24,6 +24,26 @@ type ParagraphBreak = {
length: number;
};
function findSafeSentenceBreakIndex(
text: string,
fenceSpans: FenceSpan[],
minChars: number,
): number {
const matches = text.matchAll(/[.!?](?=\s|$)/g);
let sentenceIdx = -1;
for (const match of matches) {
const at = match.index ?? -1;
if (at < minChars) {
continue;
}
const candidate = at + 1;
if (isSafeFenceBreak(fenceSpans, candidate)) {
sentenceIdx = candidate;
}
}
return sentenceIdx >= minChars ? sentenceIdx : -1;
}
export class EmbeddedBlockChunker {
#buffer = "";
readonly #chunking: BlockReplyChunking;
@@ -211,19 +231,8 @@ export class EmbeddedBlockChunker {
}
if (preference !== "newline") {
const matches = buffer.matchAll(/[.!?](?=\s|$)/g);
let sentenceIdx = -1;
for (const match of matches) {
const at = match.index ?? -1;
if (at < minChars) {
continue;
}
const candidate = at + 1;
if (isSafeFenceBreak(fenceSpans, candidate)) {
sentenceIdx = candidate;
}
}
if (sentenceIdx >= minChars) {
const sentenceIdx = findSafeSentenceBreakIndex(buffer, fenceSpans, minChars);
if (sentenceIdx !== -1) {
return { index: sentenceIdx };
}
}
@@ -271,19 +280,8 @@ export class EmbeddedBlockChunker {
}
if (preference !== "newline") {
const matches = window.matchAll(/[.!?](?=\s|$)/g);
let sentenceIdx = -1;
for (const match of matches) {
const at = match.index ?? -1;
if (at < minChars) {
continue;
}
const candidate = at + 1;
if (isSafeFenceBreak(fenceSpans, candidate)) {
sentenceIdx = candidate;
}
}
if (sentenceIdx >= minChars) {
const sentenceIdx = findSafeSentenceBreakIndex(window, fenceSpans, minChars);
if (sentenceIdx !== -1) {
return { index: sentenceIdx };
}
}