1 /*
2 * Copyright 2012-2025 CodeLibs Project and the Others.
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 * http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
13 * either express or implied. See the License for the specific language
14 * governing permissions and limitations under the License.
15 */
16 package org.codelibs.fess.crawler.helper;
17
18 import java.util.LinkedHashMap;
19 import java.util.Map;
20
21 import org.apache.commons.lang3.StringUtils;
22 import org.apache.logging.log4j.LogManager;
23 import org.apache.logging.log4j.Logger;
24 import org.codelibs.core.lang.StringUtil;
25 import org.codelibs.core.stream.StreamUtil;
26 import org.codelibs.fess.crawler.helper.impl.MimeTypeHelperImpl;
27 import org.codelibs.fess.util.ComponentUtil;
28
29 import jakarta.annotation.PostConstruct;
30
31 /**
32 * FessMimeTypeHelper extends MimeTypeHelperImpl to provide Fess-specific
33 * MIME type detection configuration. It reads extension-to-MIME-type override
34 * mappings from FessConfig to handle cases where content-based detection
35 * produces incorrect results (e.g., SQL files starting with REM comments
36 * being misdetected as batch files).
37 */
38 public class FessMimeTypeHelper extends MimeTypeHelperImpl {
39
40 private static final Logger logger = LogManager.getLogger(FessMimeTypeHelper.class);
41
42 /**
43 * Default constructor for FessMimeTypeHelper.
44 */
45 public FessMimeTypeHelper() {
46 // Default constructor
47 }
48
49 /**
50 * Initializes the extension override map by loading configuration from FessConfig.
51 * The mappings are loaded from the crawler.document.mimetype.extension.overrides property,
52 * where each line contains a mapping in the format ".ext=mime/type".
53 */
54 @PostConstruct
55 public void init() {
56 final String overrides = ComponentUtil.getFessConfig().getCrawlerDocumentMimetypeExtensionOverrides();
57 if (StringUtil.isNotBlank(overrides)) {
58 final Map<String, String> map = new LinkedHashMap<>();
59 StreamUtil.split(overrides, "\n").of(stream -> stream.filter(StringUtil::isNotBlank).forEach(s -> {
60 final String[] values = StringUtils.split(s, "=", 2);
61 if (values.length == 2) {
62 map.put(values[0].trim(), values[1].trim());
63 }
64 }));
65 setExtensionMimeTypeMap(map);
66 if (logger.isDebugEnabled()) {
67 logger.debug("loaded mimetype extension overrides: {}", map);
68 }
69 }
70 }
71 }