|
compiler { |
|
|
|
# fields available per token |
|
allTokenFields = [ |
|
${odinson.index.rawTokenField}, |
|
${odinson.index.wordTokenField}, |
|
${odinson.index.normalizedTokenField}, |
|
${odinson.index.lemmaTokenField}, |
|
${odinson.index.posTagTokenField}, |
|
${odinson.index.chunkTokenField}, |
|
${odinson.index.entityTokenField}, |
|
${odinson.index.incomingTokenField}, |
|
${odinson.index.outgoingTokenField}, |
|
] |
|
|
|
# the token field to be used when none is specified |
|
defaultTokenField = ${odinson.index.normalizedTokenField} |
|
|
|
sentenceLengthField = ${odinson.index.sentenceLengthField} |
|
|
|
dependenciesField = ${odinson.index.dependenciesField} |
|
|
|
incomingTokenField = ${odinson.index.incomingTokenField} |
|
|
|
outgoingTokenField = ${odinson.index.outgoingTokenField} |
|
|
|
# if we are using the normalizedTokenField as the default |
|
# then we should casefold the queries to the default field |
|
# so that they match |
|
aggressiveNormalizationToDefaultField = true |
|
|
|
} |
|
|
|
index { |
|
|
|
# the raw token |
|
rawTokenField = raw |
|
|
|
# the word itself |
|
wordTokenField = word |
|
|
|
# a normalized version of the token |
|
normalizedTokenField = norm |
|
|
|
# the normalized field will include values from the following fields |
|
addToNormalizedField = [ |
|
${odinson.index.rawTokenField}, |
|
${odinson.index.wordTokenField}, |
|
] |
|
|
|
lemmaTokenField = lemma |
|
|
|
posTagTokenField = tag |
|
|
|
chunkTokenField = chunk |
|
|
|
entityTokenField = entity |
|
|
|
incomingTokenField = incoming |
|
|
|
outgoingTokenField = outgoing |
|
|
|
dependenciesField = dependencies |
|
|
|
documentIdField = docId |
|
|
|
sentenceIdField = sentId |
|
|
|
sentenceLengthField = numWords |
|
|
|
maxNumberOfTokensPerSentence = 100 |