View source for Implementation:Deepset ai Haystack TextFileToDocument

{{PageInfo|type=Implementation|title=Deepset_ai_Haystack_TextFileToDocument}}
{{Metadata
| implementation_name = TextFileToDocument
| technology = Haystack
| category = Component Implementation
| domains = Data_Processing, ETL
| related_principle = [[Deepset_ai_Haystack_Text_File_Conversion]]
| source_file = haystack/components/converters/txt.py:L17-97
| import = from haystack.components.converters import TextFileToDocument
}}

== Overview ==

'''TextFileToDocument''' is a Haystack component that converts plain text files into Document objects suitable for pipeline processing. It reads text files or byte streams, decodes their content using a configurable encoding, and produces Document objects with merged metadata.

== Code Reference ==

'''Source file''': <code>haystack/components/converters/txt.py</code>, lines 17-97

'''Import''':
<syntaxhighlight lang="python">
from haystack.components.converters import TextFileToDocument
</syntaxhighlight>

=== Constructor ===

<syntaxhighlight lang="python">
TextFileToDocument(
    encoding: str = "utf-8",
    store_full_path: bool = False
)
</syntaxhighlight>

'''Parameters''':
* <code>encoding</code> ('''str''', default <code>"utf-8"</code>): The character encoding to use when decoding text files. If a source ByteStream specifies an encoding in its metadata, that value overrides this default.
* <code>store_full_path</code> ('''bool''', default <code>False</code>): If <code>True</code>, the full file path is stored in the document metadata. If <code>False</code>, only the file name (basename) is stored.

=== Run Method ===

<syntaxhighlight lang="python">
run(
    sources: list[str | Path | ByteStream],
    meta: dict[str, Any] | list[dict[str, Any]] | None = None
) -> {"documents": list[Document]}
</syntaxhighlight>

'''Parameters''':
* <code>sources</code> ('''list[str | Path | ByteStream]''', required): A list of file paths or ByteStream objects to convert.
* <code>meta</code> ('''dict | list[dict] | None''', default <code>None</code>): Optional metadata to attach. A single dictionary is added to all documents; a list must match the number of sources.

== I/O Contract ==

{| class="wikitable"
! Direction !! Name !! Type !! Description
|-
| Input || sources || list[str {{!}} Path {{!}} ByteStream] || Text file paths or byte streams to convert
|-
| Input || meta || dict {{!}} list[dict] {{!}} None || Optional metadata to attach to documents
|-
| Output || documents || list[Document] || Converted Document objects with text content and metadata
|}

== Usage Examples ==

=== Basic Text File Conversion ===

<syntaxhighlight lang="python">
from haystack.components.converters import TextFileToDocument

converter = TextFileToDocument()
results = converter.run(sources=["sample.txt"])
documents = results["documents"]
print(documents[0].content)
# 'This is the content from the txt file.'
</syntaxhighlight>

=== Custom Encoding ===

<syntaxhighlight lang="python">
from haystack.components.converters import TextFileToDocument

converter = TextFileToDocument(encoding="latin-1")
results = converter.run(sources=["legacy_file.txt"])
</syntaxhighlight>

=== With Metadata ===

<syntaxhighlight lang="python">
from haystack.components.converters import TextFileToDocument

converter = TextFileToDocument(store_full_path=True)
results = converter.run(
    sources=["notes.txt", "readme.txt"],
    meta=[{"author": "Alice"}, {"author": "Bob"}]
)
for doc in results["documents"]:
    print(doc.meta)
</syntaxhighlight>

=== Pipeline Integration ===

<syntaxhighlight lang="python">
from haystack import Pipeline
from haystack.components.routers import FileTypeRouter
from haystack.components.converters import TextFileToDocument
from haystack.components.preprocessors import DocumentCleaner

pipeline = Pipeline()
pipeline.add_component("router", FileTypeRouter(mime_types=["text/plain"]))
pipeline.add_component("converter", TextFileToDocument())
pipeline.add_component("cleaner", DocumentCleaner())

pipeline.connect("router.text/plain", "converter.sources")
pipeline.connect("converter.documents", "cleaner.documents")

result = pipeline.run({"router": {"sources": ["document.txt"]}})
</syntaxhighlight>

== Related Pages ==

=== Implements Principle ===
* [[implements::Principle:Deepset_ai_Haystack_Text_File_Conversion]]

* [[Deepset_ai_Haystack_Text_File_Conversion]] - The principle behind text file conversion
* [[Deepset_ai_Haystack_FileTypeRouter]] - Routes files by MIME type before conversion
* [[Deepset_ai_Haystack_PyPDFToDocument]] - PDF file converter component
* [[Deepset_ai_Haystack_DocumentCleaner]] - Cleans converted documents

[[Category:Implementations]]