hfst-tokenize(1)

Common options:

-h, --help

Print help message

-V, --version

Print version info

-v, --verbose

Print verbosely while processing

-q, --quiet

Only print fatal erros and requested output

-s, --silent

Alias of --quiet

-n, --newline

Newline as input separator (default is blank line)

-a, --print-all

Print nonmatching text

-w, --print-weight

Print weights (overrides earlier -W option)

-W, --no-weights

Don't print weights (default; overrides earlier -w, or -w implied by -g, options)

-m, --tokenize-multichar Tokenize multicharacter symbols

(by default only one utf-8 character is tokenized at a time regardless of what is present in the alphabet)

-b, --beam=B

Output only analyses whose weight is within B from best result

-tS, --time-cutoff=S

Limit search after having used S seconds per input

-lN, --weight-classes=N

Output no more than N best weight classes (where analyses with equal weight constitute a class

-u, --unique

Remove duplicate analyses

-z, --segment

Segmenting / tokenization mode (default)

-i, --space-separated

Tokenization with one sentence per line, space-separated tokens

-x, --xerox

Xerox output

-c, --cg

Constraint Grammar output

-S, --superblanks

Ignore contents of unescaped [] (cf. apertium-destxt); flush on NUL

-g, --giella-cg

CG format used in Giella infrastructure (implies -w and -l2, treats @PMATCH_INPUT_MARK@ as subreading separator, expects tags to be Multichar_symbols, flush on NUL)

-C --conllu

CoNLL-U format

-f, --finnpos

FinnPos output

-L, --visl

VISL input and output (implies -W, handles <s> as blocks and <STYLE> inline)

Use standard streams for input and output (for now).