LEXO CLI reference

The lexo command drives the same engine as the desktop app. Run lexo --help, or lexo <command> --help, for the built-in help.

Overview

Command	Purpose	Needs Google
lexo extract <pdf>	Extract a digital PDF’s embedded text	No
lexo ocr <pdf|image>	OCR a scanned PDF or an image	Yes
lexo ocr-batch <files-or-folders> -o <dir>	OCR many PDFs to one TXT each	Yes
lexo pdf info|extract|split|crop|rotate|merge|split-spread	PDF operations	No
lexo login / lexo logout	Sign in to or out of Google
lexo gui	Launch the desktop app
lexo info	Show the version and data folders	No
lexo check-update	Check PyPI for a newer release	No
lexo --version (-V)	Print the version	No

Page ranges (--pages, -p) accept all, odd, even, or ranges like 1-3,7,10-.

lexo extract

Extract the embedded text layer of a digital PDF. Works offline, without an account.

lexo extract report.pdf -o report.txt
lexo extract report.pdf --pages 1-10 --format markdown -o report.md

Option	Description
--pages, -p	Limit to a page range
--format, -f	text (default), markdown, or jsonl
--out, -o	Output file. Without it, text goes to standard output

If the PDF has no embedded text, LEXO says so and suggests lexo ocr.

lexo ocr

OCR a scanned PDF or an image (PNG, JPEG, TIFF, BMP) with Google Docs OCR. Pages that already have a usable text layer are extracted instead, unless you pass --force-ocr.

lexo ocr scan.pdf -o scan.txt
lexo ocr scan.pdf --pages odd --format jsonl -o scan.jsonl
lexo ocr page.png --lang my

Option	Description
--pages, -p	Limit to a page range
--lang	OCR language hint. Default my (Burmese)
--format, -f	text (default), markdown, or jsonl
--out, -o	Output file. Default: standard output
--force-ocr	OCR every page, even pages with a text layer

Progress (page N done or page N failed: …) is printed to standard error. If any page fails, the command lists the failed pages and exits with status 1.

lexo ocr-batch

OCR multiple PDFs and write one UTF-8 .txt per PDF. Accepts PDF files and folders; a folder contributes every PDF directly inside it.

lexo ocr-batch ./pdfs --out-dir ./txt
lexo ocr-batch a.pdf b.pdf ./more --out-dir ./txt --overwrite

Option	Description
--out-dir, -o	Output folder (required)
--lang	OCR language hint. Default my
--force-ocr / --no-force-ocr	OCR every page (default), or keep usable embedded text
--overwrite	Replace existing .txt files instead of skipping them

See Batch OCR.

lexo pdf

All PDF commands run locally. Details and examples are in PDF tools.

Command	Usage
info	lexo pdf info book.pdf
extract	lexo pdf extract book.pdf --pages "1-3,7" -o out.pdf
split	lexo pdf split book.pdf --every 10 or --at 4,7, optional --out-dir
crop	lexo pdf crop book.pdf --top 8 --bottom 8 -o out.pdf, also --left, --right, --pages
rotate	lexo pdf rotate book.pdf --degrees 90 -o out.pdf, optional --pages
merge	lexo pdf merge a.pdf b.pdf -o out.pdf
split-spread	lexo pdf split-spread scans.pdf -o out.pdf, optional --ratio 0.5

Crop values are percentages of the page per edge.

lexo login and lexo logout

lexo login
lexo logout

login opens a browser for Google’s consent screen and saves the token in your OS keychain. It needs credentials.json from the Google Cloud setup. logout deletes the stored token.

lexo gui

lexo gui

Opens the desktop app. See the desktop app guide.

lexo info and lexo check-update

lexo info
lexo check-update

info prints the version and the data and config folders. check-update compares your version with the latest on PyPI and prints how to upgrade.

Exit status

Commands exit with 0 on success and 1 on errors, including OCR runs where some pages failed and batch runs where some files failed.