From e40e648557c7891b22f33fe7e90c4c086ca33b42 Mon Sep 17 00:00:00 2001 From: extio-de Date: Fri, 1 Nov 2024 15:48:28 +0100 Subject: [PATCH] splitParagraphs --- src/main/java/de/extio/lmdatasetprep/preparer/Utils.java | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/main/java/de/extio/lmdatasetprep/preparer/Utils.java b/src/main/java/de/extio/lmdatasetprep/preparer/Utils.java index 5112087..77f371d 100644 --- a/src/main/java/de/extio/lmdatasetprep/preparer/Utils.java +++ b/src/main/java/de/extio/lmdatasetprep/preparer/Utils.java @@ -92,8 +92,8 @@ public class Utils { if (slidingWindows && pos > chunks_var) { for (final String delimiter : PARAGRAPH_DELIMITERS) { final int o = text.indexOf(delimiter, pos - chunks_var); - if (o > -1 && o < pos) { - pos = o; + if (o > -1 && o < pos - (PARAGRAPH_DELIMITERS.size() * 2)) { + pos = o + delimiter.length(); break; } }