Deprecated API
Contents
-
Deprecated ClassesClassDescriptionafter 2.5.0 this functionality was moved to the CompositeDetectorCurrently not suitable for real use, more a demo / prototype!since 4.1.0, removed in 4.2.0: configure the endpoint once under
"engines"and bind it with aTEXT"inference"binding, which embeds the whole document tree in batched requests.since 4.1.0, removed in 4.2.0 withOpenAIImageEmbeddingParser.since 4.1.0, removed in 4.2.0; seeAbstractEmbeddingFilter.since 4.1.0, removed in 4.2.0; seeAbstractEmbeddingFilter.since 4.1.0, removed in 4.2.0. Configure the endpoint as anopenai-embedding-engineunder"engines"and bind it with anIMAGES"inference"binding and theembedtask: a document tree's images go in requests of the engine's batch size instead of one per image, and the vectors land whereChunkTargetputs them.for removal in 4.xsince 4.1.0; configure"pages"(seePagesConfig).Use theAttributeMetadataHandlerandElementMetadataHandlerclasses instead
-
Deprecated Enum ClassesEnum ClassDescription
-
Deprecated MethodsMethodDescriptionThis API is ICU internal only.This API is ICU internal only.Use
MiscOLEDetector.detect(Set, DirectoryEntry)and pass the root entry of the filesystem whose type is to be detected, as a second argument.since 4.1, removal in 5.0: nothing produces the pseudo-types any more.since 4.1, removal planned for 5.0; useEmbeddedDocumentUtil.recordEmbeddedStreamException(Throwable, Metadata, ParseContext)so the configuredExceptionReportingapplies. Unlike in 4.0.0, this now keeps a bareTikaExceptionwrapper rather than stripping it, so the first line of the recorded value may name the wrapper.since 4.1, removal planned for 5.0; useEmbeddedDocumentUtil.recordException(Throwable, Metadata, ParseContext)so the configuredExceptionReportingapplies. Unlike in 4.0.0, this now keeps a bareTikaExceptionwrapper rather than stripping it, so the first line of the recorded value may name the wrapper.UseTika.detect(File)insteadUseTika.detect(String)insteadUseStructuralEncodingRules.detectIso2022(byte[])which distinguishes JP/KR/CN.useUtf16SpecialistEncodingDetector.score(byte[]). Kept for existing tests.use theParser.parse(TikaInputStream, ContentHandler, Metadata, ParseContext)method insteadThis method will be removed in Apache Tika 1.0.TheContentHandlerFactoryoverride is still experimental and the method signature is subject to change before Tika 2.0This has been replaced byFallbackParserno longer called by this parser: stream-backed input is loaded throughPDFParser.getPDDocument(RandomAccessRead, String, RandomAccessStreamCache.StreamCacheCreateFunction, Metadata, ParseContext), which does not copy the document into heap the wayRandomAccessReadBuffer(InputStream)doessince 4.1.0; usePDFParserConfig.setMarkedContent(MarkedContentConfig).trueisMarkedContentConfig.Strategy.TAGS,falseisMarkedContentConfig.Strategy.NONE.since 4.1.0;RAW_IMAGESisPDFParserConfig.setExtractInlineImages(boolean), theRENDER_PAGES_*values are"pages": {"emit": {"enabled": true}}andNONEisfalse. Renders are emitted at page end whichever value is given; when the engine renders is its own business.since 4.1.0; the 4.0"ocr"block. Every field it sets lands onPDFParserConfig.pages():strategyastext,strategyAutoasocr.auto,maxPagesToOcrasocr.maxPages, the image settings asrender,renderingStrategyasPDFParserConfig.setRenderingStrategy(org.apache.tika.parser.pdf.OcrConfig.RenderingStrategy).since 4.1, removal planned for 5.0; seeUnpackConfig.writesMetadata()since 4.1, removal planned for 5.0; seeUnpackConfig.writesMetadata()since 4.1, removal planned for 5.0; useUnpackConfig.setIncludeMetadata(Boolean)since 4.1, removal planned for 5.0; useUnpackConfig.setIncludeMetadata(Boolean)since 4.1, removal planned for 5.0; useFiles.createTempDirectory(prefix)since 4.1, removal planned for 5.0; seePipesConfig.setTempDirectory(String)since 4.1, removal planned for 5.0; set-Djava.io.tmpdirinsteadsince 4.1.0; seePOIMetafileRenderer.setWidth(int):pages.render.imageFormat.since 4.1.0: applies only to a directPOIMetafileRenderer.render(org.apache.tika.io.TikaInputStream, org.apache.tika.metadata.Metadata, org.apache.tika.parser.ParseContext, org.apache.tika.renderer.RenderRequest...)call with noRenderSettingsin the context. The EMF/WMF parsers always scopepages.render(300 dpi of the drawing's physical size, grayscale PNG, unless the block says otherwise), which replaces this. Configure"pages": {"render": {"maxWidth": ...}}.since 4.1.0: applies only to a directPDFBoxRenderer.render(org.apache.tika.io.TikaInputStream, org.apache.tika.metadata.Metadata, org.apache.tika.parser.ParseContext, org.apache.tika.renderer.RenderRequest...)call with noRenderSettingsin the context. A parse always scopespages.render, which replaces every value set here. Configure"pages": {"render": {...}}, or set aRenderSettingson the context when calling the renderer yourself.since 4.1.0; seePDFBoxRenderer.setDPI(int).since 4.1.0; seePDFBoxRenderer.setDPI(int).since 4.1.0; seePDFBoxRenderer.setDPI(int).since 4.1.0; seePDFBoxRenderer.setDPI(int).since 4.1.0: applies only to a directPopplerRenderer.render(org.apache.tika.io.TikaInputStream, org.apache.tika.metadata.Metadata, org.apache.tika.parser.ParseContext, org.apache.tika.renderer.RenderRequest...)call with noRenderSettingsin the context. A parse always scopespages.render, which replaces this. Configure"pages": {"render": {"dpi": ...}}.since 4.1.0; seePopplerRenderer.setDpi(int):pages.render.imageType.since 4.1, removal planned for 5.0; useExceptionUtils.format(Throwable, ParseContext)so the configured policy applies. Unlike this method, it keeps a bareTikaExceptionwrapper.since 4.1, removal planned for 5.0; useExceptionUtils.format(Throwable, ParseContext)so the configured policy appliessince 4.1, removal planned for 5.0; useParserUtils.recordParserFailure(Parser, Throwable, Metadata, ParseContext)so the configuredExceptionReportingapplies
-
Deprecated ConstructorsConstructorDescriptionUse the constructor which is passed all values directly.Use the constructor which is passed all values directly.since 4.1, use the overload taking an
ExceptionReporting