Package dev.langchain4j.data.document.loader

Class ClassPathDocumentLoader

java.lang.Object

dev.langchain4j.data.document.loader.ClassPathDocumentLoader

public class ClassPathDocumentLoader
extends Object

DocumentLoader implementation for loading documents using a ClassPathSource

Author:

Eric Deandrea

Method Summary

Modifier and Type	Method	Description
static dev.langchain4j.data.document.Document	loadDocument(String pathOnClasspath)	Loads a Document from the specified file path.
static dev.langchain4j.data.document.Document	loadDocument(String pathOnClasspath, dev.langchain4j.data.document.DocumentParser documentParser)	Loads a Document from the specified file path.
static List<dev.langchain4j.data.document.Document>	loadDocuments(String directoryOnClasspath)	Loads Documents from the specified directory.
static List<dev.langchain4j.data.document.Document>	loadDocuments(String directoryOnClasspath, dev.langchain4j.data.document.DocumentParser documentParser)	Loads Documents from the specified directory.
static List<dev.langchain4j.data.document.Document>	loadDocuments(String directoryOnClasspath, PathMatcher pathMatcher)	Loads matching Documents from the specified directory.
static List<dev.langchain4j.data.document.Document>	loadDocuments(String directoryOnClasspath, PathMatcher pathMatcher, dev.langchain4j.data.document.DocumentParser documentParser)	Loads matching Documents from the specified directory.
static List<dev.langchain4j.data.document.Document>	loadDocumentsRecursively(String directoryOnClasspath)	Recursively loads Documents from the specified directory and its subdirectories.
static List<dev.langchain4j.data.document.Document>	loadDocumentsRecursively(String directoryOnClasspath, dev.langchain4j.data.document.DocumentParser documentParser)	Recursively loads Documents from the specified directory and its subdirectories.
static List<dev.langchain4j.data.document.Document>	loadDocumentsRecursively(String directoryOnClasspath, PathMatcher pathMatcher)	Recursively loads matching Documents from the specified directory and its subdirectories.
static List<dev.langchain4j.data.document.Document>	loadDocumentsRecursively(String directoryOnClasspath, PathMatcher pathMatcher, dev.langchain4j.data.document.DocumentParser documentParser)	Recursively loads matching Documents from the specified directory and its subdirectories.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Details

loadDocument

public static dev.langchain4j.data.document.Document loadDocument(String pathOnClasspath)

Loads a Document from the specified file path.
The file is parsed using the default DocumentParser. The default DocumentParser is loaded through SPI (see DocumentParserFactoru). If no DocumentParserFactory is available in the classpath, a TextDocumentParser is used.
Returned Document contains all the textual information from the file.

Parameters:

pathOnClasspath - The path on the classpath to the file.

Returns:

document

Throws:

IllegalArgumentException - If specified path is not a file.

loadDocument

public static dev.langchain4j.data.document.Document loadDocument(String pathOnClasspath,
 dev.langchain4j.data.document.DocumentParser documentParser)

Loads a Document from the specified file path.
The file is parsed using the specified DocumentParser.
Returned Document contains all the textual information from the file.

Parameters:

pathOnClasspath - The path on the classpath to the file.

documentParser - The parser to be used for parsing text from the file.

Returns:

document

Throws:

IllegalArgumentException - If specified path is not a file.

loadDocuments

public static List<dev.langchain4j.data.document.Document> loadDocuments(String directoryOnClasspath)

Loads Documents from the specified directory. Does not use recursion.
The files are parsed using the default DocumentParser. The default DocumentParser is loaded through SPI (see DocumentParserFactoru). If no DocumentParserFactory is available in the classpath, a TextDocumentParser is used.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocuments

public static List<dev.langchain4j.data.document.Document> loadDocuments(String directoryOnClasspath,
 dev.langchain4j.data.document.DocumentParser documentParser)

Loads Documents from the specified directory. Does not use recursion.
The files are parsed using the specified DocumentParser.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

documentParser - The parser to be used for parsing text from each file.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocuments

public static List<dev.langchain4j.data.document.Document> loadDocuments(String directoryOnClasspath,
 PathMatcher pathMatcher)

Loads matching Documents from the specified directory. Does not use recursion.
The files are parsed using the default DocumentParser. The default DocumentParser is loaded through SPI (see DocumentParserFactoru). If no DocumentParserFactory is available in the classpath, a TextDocumentParser is used.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

pathMatcher - Only files whose paths match the provided PathMatcher will be loaded. For example, using FileSystems.getDefault().getPathMatcher("glob:*.txt") will load all files from directoryPath with a txt extension. When traversing the directory, each file path is converted from absolute to relative (relative to directoryPath) before being matched by a pathMatcher. Thus, pathMatcher should use relative patterns.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocuments

public static List<dev.langchain4j.data.document.Document> loadDocuments(String directoryOnClasspath,
 PathMatcher pathMatcher,
 dev.langchain4j.data.document.DocumentParser documentParser)

Loads matching Documents from the specified directory. Does not use recursion.
The files are parsed using the specified DocumentParser.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

pathMatcher - Only files whose paths match the provided PathMatcher will be loaded. For example, using FileSystems.getDefault().getPathMatcher("glob:*.txt") will load all files from directoryPath with a txt extension. When traversing the directory, each file path is converted from absolute to relative (relative to directoryPath) before being matched by a pathMatcher. Thus, pathMatcher should use relative patterns.

documentParser - The parser to be used for parsing text from each file.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocumentsRecursively

public static List<dev.langchain4j.data.document.Document> loadDocumentsRecursively(String directoryOnClasspath)

Recursively loads Documents from the specified directory and its subdirectories.
The files are parsed using the default DocumentParser. The default DocumentParser is loaded through SPI (see DocumentParserFactoru). If no DocumentParserFactory is available in the classpath, a TextDocumentParser is used.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocumentsRecursively

public static List<dev.langchain4j.data.document.Document> loadDocumentsRecursively(String directoryOnClasspath,
 dev.langchain4j.data.document.DocumentParser documentParser)

Recursively loads Documents from the specified directory and its subdirectories.
The files are parsed using the specified DocumentParser.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

documentParser - The parser to be used for parsing text from each file.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocumentsRecursively

public static List<dev.langchain4j.data.document.Document> loadDocumentsRecursively(String directoryOnClasspath,
 PathMatcher pathMatcher)

Recursively loads matching Documents from the specified directory and its subdirectories.
The files are parsed using the default DocumentParser. The default DocumentParser is loaded through SPI (see DocumentParserFactoru). If no DocumentParserFactory is available in the classpath, a TextDocumentParser is used.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

pathMatcher - Only files whose paths match the provided PathMatcher will be loaded. For example, using FileSystems.getDefault().getPathMatcher("glob:**.txt") will load all files from directoryPath and its subdirectories with a txt extension. When traversing the directory tree, each file path is converted from absolute to relative (relative to directoryPath) before being matched by a pathMatcher. Thus, pathMatcher should use relative patterns. Please be aware that *.txt pattern (with a single asterisk) will match files only in the directoryPath, but it will not match files from the subdirectories of directoryPath.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.

loadDocumentsRecursively

public static List<dev.langchain4j.data.document.Document> loadDocumentsRecursively(String directoryOnClasspath,
 PathMatcher pathMatcher,
 dev.langchain4j.data.document.DocumentParser documentParser)

Recursively loads matching Documents from the specified directory and its subdirectories.
The files are parsed using the specified DocumentParser.
Skips any Documents that fail to load.

Parameters:

directoryOnClasspath - The path to the directory on the classpath with files.

pathMatcher - Only files whose paths match the provided PathMatcher will be loaded. For example, using FileSystems.getDefault().getPathMatcher("glob:**.txt") will load all files from directoryPath and its subdirectories with a txt extension. When traversing the directory tree, each file path is converted from absolute to relative (relative to directoryPath) before being matched by a pathMatcher. Thus, pathMatcher should use relative patterns. Please be aware that *.txt pattern (with a single asterisk) will match files only in the directoryPath, but it will not match files from the subdirectories of directoryPath.

documentParser - The parser to be used for parsing text from each file.

Returns:

list of documents

Throws:

IllegalArgumentException - If specified path is not a directory.