From e7789ae3e673b7136db6fbbf8c5fa55fbd0f2615 Mon Sep 17 00:00:00 2001 From: Shinsuke Sugaya <shinsuke@apache.org> Date: Fri, 31 Dec 2021 22:35:56 +0900 Subject: [PATCH] fix #2618 tika 2.2.1 --- pom.xml | 2 +- .../crawler/transformer/AbstractFessFileTransformer.java | 8 +++----- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/pom.xml b/pom.xml index a1e44cd49..8c384a32f 100644 --- a/pom.xml +++ b/pom.xml @@ -1143,7 +1143,7 @@ </dependency> <dependency> <groupId>org.apache.tika</groupId> - <artifactId>tika-langdetect</artifactId> + <artifactId>tika-langdetect-optimaize</artifactId> <version>${tika.version}</version> <exclusions> <exclusion> diff --git a/src/main/java/org/codelibs/fess/crawler/transformer/AbstractFessFileTransformer.java b/src/main/java/org/codelibs/fess/crawler/transformer/AbstractFessFileTransformer.java index ea990a756..6454f67c6 100644 --- a/src/main/java/org/codelibs/fess/crawler/transformer/AbstractFessFileTransformer.java +++ b/src/main/java/org/codelibs/fess/crawler/transformer/AbstractFessFileTransformer.java @@ -29,8 +29,6 @@ import java.util.stream.Collectors; import org.apache.commons.lang3.StringUtils; import org.apache.logging.log4j.LogManager; import org.apache.logging.log4j.Logger; -import org.apache.tika.metadata.HttpHeaders; -import org.apache.tika.metadata.TikaMetadataKeys; import org.codelibs.core.io.SerializeUtil; import org.codelibs.core.lang.StringUtil; import org.codelibs.core.misc.Tuple3; @@ -371,9 +369,9 @@ public abstract class AbstractFessFileTransformer extends AbstractTransformer im protected Map<String, String> createExtractParams(final ResponseData responseData, final CrawlingConfig crawlingConfig) { final Map<String, String> params = new HashMap<>(crawlingConfig.getConfigParameterMap(ConfigName.CONFIG)); - params.put(TikaMetadataKeys.RESOURCE_NAME_KEY, getResourceName(responseData)); - params.put(HttpHeaders.CONTENT_TYPE, responseData.getMimeType()); - params.put(HttpHeaders.CONTENT_ENCODING, responseData.getCharSet()); + params.put(ExtractData.RESOURCE_NAME_KEY, getResourceName(responseData)); + params.put(ExtractData.CONTENT_TYPE, responseData.getMimeType()); + params.put(ExtractData.CONTENT_ENCODING, responseData.getCharSet()); params.put(ExtractData.URL, responseData.getUrl()); final Map<String, String> configParam = crawlingConfig.getConfigParameterMap(ConfigName.CONFIG); if (configParam != null) { -- GitLab