How to Extract Plain Text from a Document Using OfficeCLI

Use the officecli get <file> --format text command (or the -t shorthand) to stream clean plain text from Word, PowerPoint, or Excel documents without loading the Office UI.

The iOfficeAI/OfficeCLI repository provides a cross-platform command-line tool that parses Microsoft Office Open XML files directly. If you need to extract plain text from a document using OfficeCLI, the tool offers a built-in get command that bypasses formatting markup and outputs only the readable content from .docx, .pptx, and .xlsx files.

Command-Line Syntax for Text Extraction

OfficeCLI exposes text extraction through the get verb combined with the --format text option. When invoked, the CLI routes the request to a format-specific handler that parses the underlying XML structure and returns a concatenated string.

The basic invocation pattern is:

officecli get <file-path> --format text

For convenience, use the short alias -t:

officecli get report.docx -t

How OfficeCLI Extracts Text Internally

According to the iOfficeAI/OfficeCLI source code, the extraction logic differs by document type. The tool utilizes the Open XML SDK (referenced in src/officecli/officecli.csproj) to read document parts without loading them into the Office application.

Word Document Processing

For .docx files, the implementation resides in src/officecli/Handlers/Word/WordHandler.View.cs. The handler walks the <w:p> (paragraph) and <w:r> (run) hierarchy defined in the Open XML specification. It aggregates text content from these elements while explicitly discarding drawing objects, styles, field codes, and other non-text nodes.

PowerPoint Presentation Processing

For .pptx files, src/officecli/Handlers/Pptx/PowerPointHandler.Query.cs manages the operation. It targets <a:t> elements—text runs inside shapes—and concatenates them in slide order. When you pass the --format text flag, the query method returns the visible text content of each slide, respecting the original reading order and inserting line breaks between slides.

Practical Code Examples

You can integrate OfficeCLI into shell scripts and automation pipelines to process documents in bulk.

Extract text to stdout:

officecli get contract.docx --format text

Redirect output to a text file:

officecli get presentation.pptx -t > slides.txt

Batch process an entire directory of Word documents:

for file in *.docx; do
    officecli get "$file" -t > "${file%.docx}.txt"
done

The Node.js SDK documented in sdk/node/README.md mirrors this behavior, allowing JavaScript consumers to invoke the same extraction routines programmatically via the wrapper API.

Summary

• Use officecli get <file> -t to extract plain text from Word, PowerPoint, or Excel documents.
• Word extraction walks the <w:p>/<w:r> hierarchy in WordHandler.View.cs.
• PowerPoint extraction collects <a:t> elements via PowerPointHandler.Query.cs.
• The tool leverages the Open XML SDK and supports both .NET and Node.js environments.

Frequently Asked Questions

Does OfficeCLI preserve formatting like bold or italics when extracting text?

No. The plain-text extraction intentionally strips all presentation markup, including styles, fonts, colors, and hyperlinks. It returns only the raw character content in reading order.

Can I extract text from password-protected documents?

No. OfficeCLI cannot decrypt protected files. You must remove document protection or provide decrypted copies before using the get command.

Is there a way to extract text from a specific section only?

Yes. Combine the --path selector with the text format flag. For example, officecli get doc.docx --path "/document/body/sectPr[1]" -t extracts text only from the first section of a Word document.

Does the tool support legacy .doc or .ppt formats?

No. OfficeCLI only supports Open XML formats (.docx, .pptx, .xlsx). Legacy binary formats are not handled by the current handler implementations in the repository.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}