{
'module' => {
'name' => 'Latin pre-processing',
'submodules' => [
'pre/markup',
'pre/sent',
'pre/la/toktag',
],
'submodule names' => [
'basic XML markup',
'sentence boundary detection',
'tokenization and POS tagging (TreeTagger)',
],
'stdin' => 'text',
'stdout' => 'text',
},
'description' => 'This is the pre-processing pipe-line for Latin that includes the TreeTagger for tokenization and POS tagging.',
'input' => {
'text' => {
'format' => 'text',
}
},
'output' => {
'text' => {
'format' => 'xml',
'root' => 's',
'write_mode' => 'overwrite',
'status' => 'tag'
}
},
'arguments' => {
'shortcuts' => {
'in' => 'input:text:file',
'out' => 'output:text:file',
'ci' => 'input:text:encoding',
'co' => 'output:text:encoding',
}
},
'widgets' => {
'input' => {
'text' => {
'stream name' => 'stream(format=text,language=de)'
},
},
}
}