nzw921rx commented on code in PR #11414: URL: https://github.com/apache/seatunnel/pull/11414#discussion_r3567839005
########## seatunnel-transforms-v2/src/main/java/org/apache/seatunnel/transform/chunk/TextChunkTransformConfig.java: ########## @@ -0,0 +1,119 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.seatunnel.transform.chunk; + +import org.apache.seatunnel.api.configuration.Option; +import org.apache.seatunnel.api.configuration.Options; +import org.apache.seatunnel.api.configuration.ReadonlyConfig; +import org.apache.seatunnel.api.configuration.util.OptionValidationException; + +import lombok.Getter; +import lombok.Setter; + +import java.io.Serializable; +import java.util.Arrays; +import java.util.List; + +@Getter +@Setter +public class TextChunkTransformConfig implements Serializable { + + public static final Option<String> TEXT_FIELD = + Options.key("text_field") + .stringType() + .noDefaultValue() + .withDescription("The source text field to split into chunks"); + + public static final Option<String> OUTPUT_FIELD = + Options.key("output_field") + .stringType() + .defaultValue("chunk") + .withDescription("The output field name holding each chunk (STRING)"); + + public static final Option<String> CHUNK_INDEX_FIELD = + Options.key("chunk_index_field") + .stringType() + .defaultValue("chunk_index") + .withDescription( + "The output field name holding the chunk sequence index within a " + + "document (INT, 0-based)"); + + public static final Option<Integer> CHUNK_SIZE = + Options.key("chunk_size") + .intType() + .defaultValue(1000) + .withDescription("Maximum length of each chunk, measured in characters"); + + public static final Option<Integer> OVERLAP_SIZE = + Options.key("overlap_size") + .intType() + .defaultValue(0) + .withDescription( + "Overlap length between adjacent chunks, measured in characters. " + + "Must be less than chunk_size"); + + public static final Option<List<String>> SEPARATORS = + Options.key("separators") + .listType() + .defaultValue(Arrays.asList("\n\n", "\n", "。", "！", "？", ". ", " ")) + .withDescription( + "Separators tried in priority order to avoid cutting mid-sentence. " + + "If left empty, falls back to fixed-size splitting"); + + private String textField; + private String outputField; + private String chunkIndexField; + private int chunkSize; + private int overlapSize; + private List<String> separators; + + public static TextChunkTransformConfig of(ReadonlyConfig config) { + TextChunkTransformConfig c = new TextChunkTransformConfig(); + c.setTextField(config.get(TEXT_FIELD)); + c.setOutputField(config.get(OUTPUT_FIELD)); + c.setChunkIndexField(config.get(CHUNK_INDEX_FIELD)); + c.setChunkSize(config.get(CHUNK_SIZE)); + c.setOverlapSize(config.get(OVERLAP_SIZE)); + c.setSeparators(config.get(SEPARATORS)); + return c; + } + + public static void validate(TextChunkTransformConfig config) { Review Comment: Thank you for your contribution. This is an essential supplement to RAG. Currently, the `validate (TextChunkTransformConfig config) ` function is used by Factory and has no common use. From the perspective of class self blame, this function is more suitable to be placed in `TextChunkTransformFactory`. If you have any different opinions, please feel free to discuss -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
