kcat

Print the human-readable text extracted from each file, one block per line, regardless of the underlying format. A Word .docx, a JSON catalog and an XLIFF file all print as their plain prose, with the markup and structure stripped.

kcat [flags] [FILE...]

With no file, or when the file is -, standard input is read.

Examples

# See the prose inside a Word document
kcat report.docx

# Number the blocks of a JSON catalog
kcat -n locales/en.json

# Print the French translations of an XLIFF file
kcat --target fr messages.xliff

# Pipe arbitrary text through, treating it as plain text
cat raw.txt | kcat -f plaintext

kcat pairs naturally with the shell tools you already have — pipe its output into the real grep, wc, or sort when you want byte-level line behaviour rather than the block-aware kgrep.

Options

Flag	Meaning
-n, --number	Number the output blocks.
--id	Prefix each block with its source ID.
--target LOCALE	Print the translation for LOCALE instead of the source.
-f, --format	Override format detection (e.g. -f json).
--source-lang	Source language (default en).
--encoding	Input encoding (default UTF-8).
--json	Emit blocks as JSON instead of plain text.