output

Documentation

Index

• func CSVHeader(withVerification bool, sep rune) string
• type Kingdom
• type Meta
• type Name
  • func FilterNames(names []Name, genera map[string]struct{}) []Name
• type OddsDatum
• type Output
  • func TokensToOutput(ts []token.TokenSN, text []rune, version string, cfg config.Config) Output
  • func (o *Output) Format(f gnfmt.Format) string
  • func (o *Output) MergeVerification(v map[string]vlib.Name, st stats.Stats, dur float32)
  • func (o *Output) UniqueNameStrings() []string

Functions

func CSVHeader

func CSVHeader(withVerification bool, sep rune) string

CSVHeader returns the header string for CSV output format.

Types

type Kingdom

type Kingdom struct {
	NamesNumber     int     `json:"namesNumber"`
	Kingdom         string  `json:"kingdom"`
	NamesPercentage float32 `json:"namesPercentage"`
}

Kingdom contains names resolved to it and their percentage.

type Meta

type Meta struct {
	// Documentation
	Documentation string `json:"documentation"`

	// Date represents time when output was generated.
	Date time.Time `json:"date"`

	// FinderVersion the version of gnfinder.
	FinderVersion string `json:"gnfinderVersion"`

	// InputFile is the name of the source file.
	InputFile string `json:"inputFile,omitempty"`

	// TextExtractionSec is the time spent on converting the file
	// into UTF8-encoded text.
	TextExtractionSec float32 `json:"textExtractSec,omitempty"`

	// NameFindingSec is the time spent on name-finding.
	NameFindingSec float32 `json:"nameFindingSec"`

	// NameVerifSec is the time spent on name-verification.
	NameVerifSec float32 `json:"nameVerifSec,omitempty"`

	// TotalSec is time spent for the whole process
	TotalSec float32 `json:"totalSec"`

	// WordsAround shows the number of tokens preserved before and after
	// a name-string candidate.
	WordsAround int `json:"wordsAround"`

	// Language setting used by the name-finding algorithm.
	Language string `json:"language"`

	// LanguageDetected automatically for the text.
	LanguageDetected string `json:"languageDetected,omitempty"`

	// WithAllMatches is true if all verifcation results are shown.
	WithAllMatches bool `json:"withAllMatches,omitempty"`

	// WithAmbiguousNames is true if ambiguous uninomials are preserved.
	// Examples of ambiguous uninomial names are `Cancer`, `America`.
	WithAmbiguousNames bool `json:"withAmbiguousNames,omitempty"`

	// WithUniqueNames is true when unique names are returned instead
	// of every occurance of a name.
	WithUniqueNames bool `json:"withUniqueNames,omitempty"`

	// WithBayes use of bayes during name-finding
	WithBayes bool `json:"withBayes,omitempty"`

	// WithOddsAdjustment to adjust prior odds according to the density of
	// scientific names in the text.
	WithOddsAdjustment bool `json:"withOddsAdjustment,omitempty"`

	// WithPositionInBytes names get start/enc positionx in bytes
	// instead of UTF-8 chars.
	WithPositionInBytes bool `json:"withPositionInBytes,omitempty"`

	// WithVerification is true if results are checked by verification service.
	WithVerification bool `json:"withVerification,omitempty"`

	// WithLanguageDetection sets automatic language determination.
	WithLanguageDetection bool `json:"withLanguageDetection,omitempty"`

	// TotalWords is a number of 'normalized' words in the text
	TotalWords int `json:"totalWords"`

	// TotalNameCandidates is a number of words that might be a start of
	// a scientific name
	TotalNameCandidates int `json:"totalNameCandidates"`

	// TotalNames is a number of scientific names found
	TotalNames int `json:"totalNames"`

	// Kingdoms are the kingdoms to which the names resolved by
	// the Catalogue of Life are placed.
	// Kingdoms are sorted by percentage in descending order.
	// The first kingom contains the most number of names.
	Kingdoms []Kingdom `json:"kingdoms,omitempty"`

	// MainTaxon is the taxon containing the majority of resolved by
	// the Catalogue of Life names.
	MainTaxon string `json:"mainTaxon,omitempty"`

	// MainTaxonRank is the rank of the MainTaxon.
	MainTaxonRank string `json:"mainTaxonRank,omitempty"`

	// MainTaxonPercentage is the percentage of names in MainTaxon.
	MainTaxonPercentage float32 `json:"mainTaxonPercentage,omitempty"`

	// StatsNamesNum is the number of names used for calculating statistics.
	// It includes names that are genus and lower and are verified to
	// Catalogue of Life.
	StatsNamesNum int `json:"statsNamesNum,omitempty"`
}

Meta contains meta-information of name-finding result.

type Name

type Name struct {
	// Cardinality depicts number of elements in a name. 0 - Cannot determine
	// cardinality, 1 - Uninomial, 2 - Binomial, 3 - Trinomial.
	Cardinality int `json:"cardinality"`

	// Verbatim shows name the way it was in the text.
	Verbatim string `json:"verbatim,omitempty"`

	// Name is a normalized version of a name.
	Name string `json:"name"`

	// Decision about the quality of name detection.
	Decision token.Decision `json:"-"`

	// Odds show a probability that name detection was correct.
	Odds float64 `json:"-"`

	// OddsLog10 show a Log10 of Odds.
	OddsLog10 float64 `json:"oddsLog10,omitempty"`

	// OddsDetails descibes how Odds were calculated.
	OddsDetails boutput.OddsDetails `json:"oddsDetails,omitempty"`

	// OffsetStart is a start of a name on a page.
	OffsetStart int `json:"start"`

	// OffsetEnd is the end of the name on a page.
	OffsetEnd int `json:"end"`

	// AnnotNomen is a nomenclatural annotation for new species or combination.
	AnnotNomen string `json:"annotationNomen,omitempty"`

	// AnnotNomenType is normalized nomenclatural annotation.
	AnnotNomenType string `json:"annotationNomenType,omitempty"`

	// WordsBefore are words that happened before the name.
	WordsBefore []string `json:"wordsBefore,omitempty"`

	// WordsAfter are words that happened right after the name.
	WordsAfter []string `json:"wordsAfter,omitempty"`

	// Verification gives results of verification process of the name.
	Verification *vlib.Name `json:"verification,omitempty"`
}

Name represents one found name.

func FilterNames

func FilterNames(names []Name, genera map[string]struct{}) []Name

type OddsDatum

type OddsDatum struct {
	Name bool
	Odds float64
}

OddsDatum is a simplified version of a name, that stores boolean decision (Name/NotName), and corresponding odds of the name.

type Output

type Output struct {
	Meta      `json:"metadata"`
	InputText string `json:"inputText,omitempty"`
	Names     []Name `json:"names"`
}

Output type is the result of name-finding.

func TokensToOutput

func TokensToOutput(
	ts []token.TokenSN,
	text []rune,
	version string,
	cfg config.Config) Output

TokensToOutput takes tagged tokens and assembles output out of them.

func (*Output) Format

func (o *Output) Format(f gnfmt.Format) string

func (*Output) MergeVerification

func (o *Output) MergeVerification(
	v map[string]vlib.Name,
	st stats.Stats,
	dur float32,
)

MergeVerification takes a map with verified names and incorporates into output.

func (*Output) UniqueNameStrings

func (o *Output) UniqueNameStrings() []string

UniqueNameStrings takes a list of names, and returns a list of unique name-strings