bleve

package module
v0.8.0 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Jul 29, 2019 License: Apache-2.0 Imports: 31 Imported by: 0

README

bleve bleve

Build Status Coverage Status GoDoc codebeat Go Report Card Sourcegraph License

modern text indexing in go - blevesearch.com

Try out bleve live by searching the bleve website.

Features

  • Index any go data structure (including JSON)
  • Intelligent defaults backed up by powerful configuration
  • Supported field types:
    • Text, Numeric, Date
  • Supported query types:
    • Term, Phrase, Match, Match Phrase, Prefix
    • Conjunction, Disjunction, Boolean
    • Numeric Range, Date Range
    • Simple query syntax for human entry
  • tf-idf Scoring
  • Search result match highlighting
  • Supports Aggregating Facets:
    • Terms Facet
    • Numeric Range Facet
    • Date Range Facet

Discussion

Discuss usage and development of bleve in the google group.

Indexing

message := struct{
	Id   string
	From string
	Body string
}{
	Id:   "example",
	From: "marty.schoch@gmail.com",
	Body: "bleve indexing is easy",
}

mapping := bleve.NewIndexMapping()
index, err := bleve.New("example.bleve", mapping)
if err != nil {
	panic(err)
}
index.Index(message.Id, message)

Querying

index, _ := bleve.Open("example.bleve")
query := bleve.NewQueryStringQuery("bleve")
searchRequest := bleve.NewSearchRequest(query)
searchResult, _ := index.Search(searchRequest)

License

Apache License Version 2.0

Documentation

Overview

Package bleve is a library for indexing and searching text.

Example Opening New Index, Indexing Data

message := struct{
    Id:   "example"
    From: "marty.schoch@gmail.com",
    Body: "bleve indexing is easy",
}

mapping := bleve.NewIndexMapping()
index, _ := bleve.New("example.bleve", mapping)
index.Index(message.Id, message)

Example Opening Existing Index, Searching Data

index, _ := bleve.Open("example.bleve")
query := bleve.NewQueryStringQuery("bleve")
searchRequest := bleve.NewSearchRequest(query)
searchResult, _ := index.Search(searchRequest)

Index

Examples

Constants

View Source
const SearchQueryEndCallbackKey = "_search_query_end_callback_key"
View Source
const SearchQueryStartCallbackKey = "_search_query_start_callback_key"

Variables

View Source
var Config *configuration

Config contains library level configuration

Functions

func LoadAndHighlightFields added in v0.8.0

func LoadAndHighlightFields(hit *search.DocumentMatch, req *SearchRequest,
	indexName string, r index.IndexReader,
	highlighter highlight.Highlighter) error

func MemoryNeededForSearchResult added in v0.8.0

func MemoryNeededForSearchResult(req *SearchRequest) uint64

MemoryNeededForSearchResult is an exported helper function to determine the RAM needed to accommodate the results for a given search request.

func NewBoolFieldQuery

func NewBoolFieldQuery(val bool) *query.BoolFieldQuery

NewBoolFieldQuery creates a new Query for boolean fields

func NewBooleanFieldMapping

func NewBooleanFieldMapping() *mapping.FieldMapping

NewBooleanFieldMapping returns a default field mapping for booleans

func NewBooleanQuery

func NewBooleanQuery() *query.BooleanQuery

NewBooleanQuery creates a compound Query composed of several other Query objects. These other query objects are added using the AddMust() AddShould() and AddMustNot() methods. Result documents must satisfy ALL of the must Queries. Result documents must satisfy NONE of the must not Queries. Result documents that ALSO satisfy any of the should Queries will score higher.

Example
must := NewMatchQuery("one")
mustNot := NewMatchQuery("great")
query := NewBooleanQuery()
query.AddMust(must)
query.AddMustNot(mustNot)
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 1

func NewConjunctionQuery

func NewConjunctionQuery(conjuncts ...query.Query) *query.ConjunctionQuery

NewConjunctionQuery creates a new compound Query. Result documents must satisfy all of the queries.

Example
conjunct1 := NewMatchQuery("great")
conjunct2 := NewMatchQuery("one")
query := NewConjunctionQuery(conjunct1, conjunct2)
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 2

func NewDateRangeInclusiveQuery

func NewDateRangeInclusiveQuery(start, end time.Time, startInclusive, endInclusive *bool) *query.DateRangeQuery

NewDateRangeInclusiveQuery creates a new Query for ranges of date values. Date strings are parsed using the DateTimeParser configured in the

top-level config.QueryDateTimeParser

Either, but not both endpoints can be nil. startInclusive and endInclusive control inclusion of the endpoints.

func NewDateRangeQuery

func NewDateRangeQuery(start, end time.Time) *query.DateRangeQuery

NewDateRangeQuery creates a new Query for ranges of date values. Date strings are parsed using the DateTimeParser configured in the

top-level config.QueryDateTimeParser

Either, but not both endpoints can be nil.

func NewDateTimeFieldMapping

func NewDateTimeFieldMapping() *mapping.FieldMapping

NewDateTimeFieldMapping returns a default field mapping for dates

func NewDisjunctionQuery

func NewDisjunctionQuery(disjuncts ...query.Query) *query.DisjunctionQuery

NewDisjunctionQuery creates a new compound Query. Result documents satisfy at least one Query.

Example
disjunct1 := NewMatchQuery("great")
disjunct2 := NewMatchQuery("named")
query := NewDisjunctionQuery(disjunct1, disjunct2)
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(len(searchResults.Hits))
Output:

2

func NewDocIDQuery

func NewDocIDQuery(ids []string) *query.DocIDQuery

NewDocIDQuery creates a new Query object returning indexed documents among the specified set. Combine it with ConjunctionQuery to restrict the scope of other queries output.

func NewDocumentDisabledMapping

func NewDocumentDisabledMapping() *mapping.DocumentMapping

NewDocumentDisabledMapping returns a new document mapping that will not perform any indexing.

func NewDocumentMapping

func NewDocumentMapping() *mapping.DocumentMapping

NewDocumentMapping returns a new document mapping with all the default values.

func NewDocumentStaticMapping

func NewDocumentStaticMapping() *mapping.DocumentMapping

NewDocumentStaticMapping returns a new document mapping that will not automatically index parts of a document without an explicit mapping.

func NewFuzzyQuery

func NewFuzzyQuery(term string) *query.FuzzyQuery

NewFuzzyQuery creates a new Query which finds documents containing terms within a specific fuzziness of the specified term. The default fuzziness is 1.

The current implementation uses Levenshtein edit distance as the fuzziness metric.

func NewGeoBoundingBoxQuery added in v0.6.0

func NewGeoBoundingBoxQuery(topLeftLon, topLeftLat, bottomRightLon, bottomRightLat float64) *query.GeoBoundingBoxQuery

NewGeoBoundingBoxQuery creates a new Query for performing geo bounding box searches. The arguments describe the position of the box and documents which have an indexed geo point inside the box will be returned.

func NewGeoDistanceQuery added in v0.6.0

func NewGeoDistanceQuery(lon, lat float64, distance string) *query.GeoDistanceQuery

NewGeoDistanceQuery creates a new Query for performing geo distance searches. The arguments describe a position and a distance. Documents which have an indexed geo point which is less than or equal to the provided distance from the given position will be returned.

func NewGeoPointFieldMapping added in v0.6.0

func NewGeoPointFieldMapping() *mapping.FieldMapping

func NewIndexAlias

func NewIndexAlias(indexes ...Index) *indexAliasImpl

NewIndexAlias creates a new IndexAlias over the provided Index objects.

func NewIndexMapping

func NewIndexMapping() *mapping.IndexMappingImpl

NewIndexMapping creates a new IndexMapping that will use all the default indexing rules

func NewMatchAllQuery

func NewMatchAllQuery() *query.MatchAllQuery

NewMatchAllQuery creates a Query which will match all documents in the index.

Example
// finds all documents in the index
query := NewMatchAllQuery()
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(len(searchResults.Hits))
Output:

2

func NewMatchNoneQuery

func NewMatchNoneQuery() *query.MatchNoneQuery

NewMatchNoneQuery creates a Query which will not match any documents in the index.

Example
// matches no documents in the index
query := NewMatchNoneQuery()
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(len(searchResults.Hits))
Output:

0

func NewMatchPhraseQuery

func NewMatchPhraseQuery(matchPhrase string) *query.MatchPhraseQuery

NewMatchPhraseQuery creates a new Query object for matching phrases in the index. An Analyzer is chosen based on the field. Input text is analyzed using this analyzer. Token terms resulting from this analysis are used to build a search phrase. Result documents must match this phrase. Queried field must have been indexed with IncludeTermVectors set to true.

Example
// finds all documents with the given phrase in the index
query := NewMatchPhraseQuery("nameless one")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 2

func NewMatchQuery

func NewMatchQuery(match string) *query.MatchQuery

NewMatchQuery creates a Query for matching text. An Analyzer is chosen based on the field. Input text is analyzed using this analyzer. Token terms resulting from this analysis are used to perform term searches. Result documents must satisfy at least one of these term searches.

Example
// finds documents with fields fully matching the given query text
query := NewMatchQuery("named one")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 1

func NewNumericFieldMapping

func NewNumericFieldMapping() *mapping.FieldMapping

NewNumericFieldMapping returns a default field mapping for numbers

func NewNumericRangeInclusiveQuery

func NewNumericRangeInclusiveQuery(min, max *float64, minInclusive, maxInclusive *bool) *query.NumericRangeQuery

NewNumericRangeInclusiveQuery creates a new Query for ranges of numeric values. Either, but not both endpoints can be nil. Control endpoint inclusion with inclusiveMin, inclusiveMax.

Example
value1 := float64(10)
value2 := float64(100)
v1incl := false
v2incl := false

query := NewNumericRangeInclusiveQuery(&value1, &value2, &v1incl, &v2incl)
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 3

func NewNumericRangeQuery

func NewNumericRangeQuery(min, max *float64) *query.NumericRangeQuery

NewNumericRangeQuery creates a new Query for ranges of numeric values. Either, but not both endpoints can be nil. The minimum value is inclusive. The maximum value is exclusive.

Example
value1 := float64(11)
value2 := float64(100)
data := struct{ Priority float64 }{Priority: float64(15)}
data2 := struct{ Priority float64 }{Priority: float64(10)}

err = exampleIndex.Index("document id 3", data)
if err != nil {
	panic(err)
}
err = exampleIndex.Index("document id 4", data2)
if err != nil {
	panic(err)
}

query := NewNumericRangeQuery(&value1, &value2)
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 3

func NewPhraseQuery

func NewPhraseQuery(terms []string, field string) *query.PhraseQuery

NewPhraseQuery creates a new Query for finding exact term phrases in the index. The provided terms must exist in the correct order, at the correct index offsets, in the specified field. Queried field must have been indexed with IncludeTermVectors set to true.

Example
// finds all documents with the given phrases in the given field in the index
query := NewPhraseQuery([]string{"nameless", "one"}, "Name")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 2

func NewPrefixQuery

func NewPrefixQuery(prefix string) *query.PrefixQuery

NewPrefixQuery creates a new Query which finds documents containing terms that start with the specified prefix.

Example
// finds all documents with terms having the given prefix in the index
query := NewPrefixQuery("name")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(len(searchResults.Hits))
Output:

2

func NewQueryStringQuery

func NewQueryStringQuery(q string) *query.QueryStringQuery

NewQueryStringQuery creates a new Query used for finding documents that satisfy a query string. The query string is a small query language for humans.

Example
query := NewQueryStringQuery("+one -great")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 1

func NewRegexpQuery

func NewRegexpQuery(regexp string) *query.RegexpQuery

NewRegexpQuery creates a new Query which finds documents containing terms that match the specified regular expression.

func NewTermQuery

func NewTermQuery(term string) *query.TermQuery

NewTermQuery creates a new Query for finding an exact term match in the index.

Example
query := NewTermQuery("great")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 2

func NewTermRangeInclusiveQuery added in v0.6.0

func NewTermRangeInclusiveQuery(min, max string, minInclusive, maxInclusive *bool) *query.TermRangeQuery

NewTermRangeInclusiveQuery creates a new Query for ranges of text terms. Either, but not both endpoints can be "". Control endpoint inclusion with inclusiveMin, inclusiveMax.

func NewTermRangeQuery added in v0.6.0

func NewTermRangeQuery(min, max string) *query.TermRangeQuery

NewTermRangeQuery creates a new Query for ranges of text terms. Either, but not both endpoints can be "". The minimum value is inclusive. The maximum value is exclusive.

func NewTextFieldMapping

func NewTextFieldMapping() *mapping.FieldMapping

NewTextFieldMapping returns a default field mapping for text

func NewWildcardQuery

func NewWildcardQuery(wildcard string) *query.WildcardQuery

NewWildcardQuery creates a new Query which finds documents containing terms that match the specified wildcard. In the wildcard pattern '*' will match any sequence of 0 or more characters, and '?' will match any single character.

func SetLog

func SetLog(l *log.Logger)

SetLog sets the logger used for logging by default log messages are sent to ioutil.Discard

Types

type Batch

type Batch struct {
	// contains filtered or unexported fields
}

A Batch groups together multiple Index and Delete operations you would like performed at the same time. The Batch structure is NOT thread-safe. You should only perform operations on a batch from a single thread at a time. Once batch execution has started, you may not modify it.

func (*Batch) Delete

func (b *Batch) Delete(id string)

Delete adds the specified delete operation to the batch. NOTE: the bleve Index is not updated until the batch is executed.

func (*Batch) DeleteInternal

func (b *Batch) DeleteInternal(key []byte)

DeleteInternal adds the specified delete internal operation to the batch. NOTE: the bleve Index is not updated until the batch is executed.

func (*Batch) Index

func (b *Batch) Index(id string, data interface{}) error

Index adds the specified index operation to the batch. NOTE: the bleve Index is not updated until the batch is executed.

func (*Batch) IndexAdvanced added in v0.6.0

func (b *Batch) IndexAdvanced(doc *document.Document) (err error)

IndexAdvanced adds the specified index operation to the batch which skips the mapping. NOTE: the bleve Index is not updated until the batch is executed.

func (*Batch) LastDocSize added in v0.8.0

func (b *Batch) LastDocSize() uint64

func (*Batch) Merge added in v0.8.0

func (b *Batch) Merge(o *Batch)

func (*Batch) PersistedCallback added in v0.8.0

func (b *Batch) PersistedCallback() index.BatchCallback

func (*Batch) Reset

func (b *Batch) Reset()

Reset returns a Batch to the empty state so that it can be re-used in the future.

func (*Batch) SetInternal

func (b *Batch) SetInternal(key, val []byte)

SetInternal adds the specified set internal operation to the batch. NOTE: the bleve Index is not updated until the batch is executed.

func (*Batch) SetPersistedCallback added in v0.8.0

func (b *Batch) SetPersistedCallback(f index.BatchCallback)

func (*Batch) Size

func (b *Batch) Size() int

Size returns the total number of operations inside the batch including normal index operations and internal operations.

func (*Batch) String

func (b *Batch) String() string

String prints a user friendly string representation of what is inside this batch.

func (*Batch) TotalDocsSize added in v0.8.0

func (b *Batch) TotalDocsSize() uint64

type Error

type Error int

Error represents a more strongly typed bleve error for detecting and handling specific types of errors.

const (
	ErrorIndexPathExists Error = iota
	ErrorIndexPathDoesNotExist
	ErrorIndexMetaMissing
	ErrorIndexMetaCorrupt
	ErrorUnknownStorageType
	ErrorIndexClosed
	ErrorAliasMulti
	ErrorAliasEmpty
	ErrorUnknownIndexType
	ErrorEmptyID
	ErrorIndexReadInconsistency
)

Constant Error values which can be compared to determine the type of error

func (Error) Error

func (e Error) Error() string

type FacetRequest

type FacetRequest struct {
	Size           int              `json:"size"`
	Field          string           `json:"field"`
	NumericRanges  []*numericRange  `json:"numeric_ranges,omitempty"`
	DateTimeRanges []*dateTimeRange `json:"date_ranges,omitempty"`
}

A FacetRequest describes a facet or aggregation of the result document set you would like to be built.

func NewFacetRequest

func NewFacetRequest(field string, size int) *FacetRequest

NewFacetRequest creates a facet on the specified field that limits the number of entries to the specified size.

Example
facet := NewFacetRequest("Name", 1)
query := NewMatchAllQuery()
searchRequest := NewSearchRequest(query)
searchRequest.AddFacet("facet name", facet)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

// total number of terms
fmt.Println(searchResults.Facets["facet name"].Total)
// numer of docs with no value for this field
fmt.Println(searchResults.Facets["facet name"].Missing)
// term with highest occurrences in field name
fmt.Println(searchResults.Facets["facet name"].Terms[0].Term)
Output:

5
2
one

func (*FacetRequest) AddDateTimeRange

func (fr *FacetRequest) AddDateTimeRange(name string, start, end time.Time)

AddDateTimeRange adds a bucket to a field containing date values. Documents with a date value falling into this range are tabulated as part of this bucket/range.

Example
facet := NewFacetRequest("Created", 1)
facet.AddDateTimeRange("range name", time.Unix(0, 0), time.Now())
query := NewMatchAllQuery()
searchRequest := NewSearchRequest(query)
searchRequest.AddFacet("facet name", facet)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

// dates in field Created since starting of unix time till now
fmt.Println(searchResults.Facets["facet name"].DateRanges[0].Count)
Output:

2

func (*FacetRequest) AddDateTimeRangeString added in v0.6.0

func (fr *FacetRequest) AddDateTimeRangeString(name string, start, end *string)

AddDateTimeRangeString adds a bucket to a field containing date values.

func (*FacetRequest) AddNumericRange

func (fr *FacetRequest) AddNumericRange(name string, min, max *float64)

AddNumericRange adds a bucket to a field containing numeric values. Documents with a numeric value falling into this range are tabulated as part of this bucket/range.

Example
value1 := float64(11)

facet := NewFacetRequest("Priority", 1)
facet.AddNumericRange("range name", &value1, nil)
query := NewMatchAllQuery()
searchRequest := NewSearchRequest(query)
searchRequest.AddFacet("facet name", facet)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

// number documents with field Priority in the given range
fmt.Println(searchResults.Facets["facet name"].NumericRanges[0].Count)
Output:

1

func (*FacetRequest) Validate added in v0.3.0

func (fr *FacetRequest) Validate() error

type FacetsRequest

type FacetsRequest map[string]*FacetRequest

FacetsRequest groups together all the FacetRequest objects for a single query.

func (FacetsRequest) Validate added in v0.3.0

func (fr FacetsRequest) Validate() error

type HighlightRequest

type HighlightRequest struct {
	Style  *string  `json:"style"`
	Fields []string `json:"fields"`
}

HighlightRequest describes how field matches should be highlighted.

func NewHighlight

func NewHighlight() *HighlightRequest

NewHighlight creates a default HighlightRequest.

Example
query := NewMatchQuery("nameless")
searchRequest := NewSearchRequest(query)
searchRequest.Highlight = NewHighlight()
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].Fragments["Name"][0])
Output:

great <mark>nameless</mark> one

func NewHighlightWithStyle

func NewHighlightWithStyle(style string) *HighlightRequest

NewHighlightWithStyle creates a HighlightRequest with an alternate style.

Example
query := NewMatchQuery("nameless")
searchRequest := NewSearchRequest(query)
searchRequest.Highlight = NewHighlightWithStyle(ansi.Name)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].Fragments["Name"][0])
Output:

great �[43mnameless�[0m one

func (*HighlightRequest) AddField

func (h *HighlightRequest) AddField(field string)

type Index

type Index interface {
	// Index analyzes, indexes or stores mapped data fields. Supplied
	// identifier is bound to analyzed data and will be retrieved by search
	// requests. See Index interface documentation for details about mapping
	// rules.
	Index(id string, data interface{}) error
	Delete(id string) error

	NewBatch() *Batch
	Batch(b *Batch) error

	// Document returns specified document or nil if the document is not
	// indexed or stored.
	Document(id string) (*document.Document, error)
	// DocCount returns the number of documents in the index.
	DocCount() (uint64, error)

	Search(req *SearchRequest) (*SearchResult, error)
	SearchInContext(ctx context.Context, req *SearchRequest) (*SearchResult, error)

	Fields() ([]string, error)

	FieldDict(field string) (index.FieldDict, error)
	FieldDictRange(field string, startTerm []byte, endTerm []byte) (index.FieldDict, error)
	FieldDictPrefix(field string, termPrefix []byte) (index.FieldDict, error)

	Close() error

	Mapping() mapping.IndexMapping

	Stats() *IndexStat
	StatsMap() map[string]interface{}

	GetInternal(key []byte) ([]byte, error)
	SetInternal(key, val []byte) error
	DeleteInternal(key []byte) error

	// Name returns the name of the index (by default this is the path)
	Name() string
	// SetName lets you assign your own logical name to this index
	SetName(string)

	// Advanced returns the indexer and data store, exposing lower level
	// methods to enumerate records and access data.
	Advanced() (index.Index, store.KVStore, error)
}

An Index implements all the indexing and searching capabilities of bleve. An Index can be created using the New() and Open() methods.

Index() takes an input value, deduces a DocumentMapping for its type, assigns string paths to its fields or values then applies field mappings on them.

The DocumentMapping used to index a value is deduced by the following rules:

  1. If value implements mapping.bleveClassifier interface, resolve the mapping from BleveType().
  2. If value implements mapping.Classifier interface, resolve the mapping from Type().
  3. If value has a string field or value at IndexMapping.TypeField.

(defaulting to "_type"), use it to resolve the mapping. Fields addressing is described below. 4) If IndexMapping.DefaultType is registered, return it. 5) Return IndexMapping.DefaultMapping.

Each field or nested field of the value is identified by a string path, then mapped to one or several FieldMappings which extract the result for analysis.

Struct values fields are identified by their "json:" tag, or by their name. Nested fields are identified by prefixing with their parent identifier, separated by a dot.

Map values entries are identified by their string key. Entries not indexed by strings are ignored. Entry values are identified recursively like struct fields.

Slice and array values are identified by their field name. Their elements are processed sequentially with the same FieldMapping.

String, float64 and time.Time values are identified by their field name. Other types are ignored.

Each value identifier is decomposed in its parts and recursively address SubDocumentMappings in the tree starting at the root DocumentMapping. If a mapping is found, all its FieldMappings are applied to the value. If no mapping is found and the root DocumentMapping is dynamic, default mappings are used based on value type and IndexMapping default configurations.

Finally, mapped values are analyzed, indexed or stored. See FieldMapping.Analyzer to know how an analyzer is resolved for a given field.

Examples:

type Date struct {
  Day string `json:"day"`
  Month string
  Year string
}

type Person struct {
  FirstName string `json:"first_name"`
  LastName string
  BirthDate Date `json:"birth_date"`
}

A Person value FirstName is mapped by the SubDocumentMapping at "first_name". Its LastName is mapped by the one at "LastName". The day of BirthDate is mapped to the SubDocumentMapping "day" of the root SubDocumentMapping "birth_date". It will appear as the "birth_date.day" field in the index. The month is mapped to "birth_date.Month".

Example (Indexing)
data := struct {
	Name    string
	Created time.Time
	Age     int
}{Name: "named one", Created: time.Now(), Age: 50}
data2 := struct {
	Name    string
	Created time.Time
	Age     int
}{Name: "great nameless one", Created: time.Now(), Age: 25}

// index some data
err = exampleIndex.Index("document id 1", data)
if err != nil {
	panic(err)
}
err = exampleIndex.Index("document id 2", data2)
if err != nil {
	panic(err)
}

// 2 documents have been indexed
count, err := exampleIndex.DocCount()
if err != nil {
	panic(err)
}

fmt.Println(count)
Output:

2

func New

func New(path string, mapping mapping.IndexMapping) (Index, error)

New index at the specified path, must not exist. The provided mapping will be used for all Index/Search operations.

Example
indexMapping = NewIndexMapping()
exampleIndex, err = New("path_to_index", indexMapping)
if err != nil {
	panic(err)
}
count, err := exampleIndex.DocCount()
if err != nil {
	panic(err)
}

fmt.Println(count)
Output:

0

func NewMemOnly added in v0.5.0

func NewMemOnly(mapping mapping.IndexMapping) (Index, error)

NewMemOnly creates a memory-only index. The contents of the index is NOT persisted, and will be lost once closed. The provided mapping will be used for all Index/Search operations.

func NewUsing

func NewUsing(path string, mapping mapping.IndexMapping, indexType string, kvstore string, kvconfig map[string]interface{}) (Index, error)

NewUsing creates index at the specified path, which must not already exist. The provided mapping will be used for all Index/Search operations. The specified index type will be used. The specified kvstore implementation will be used and the provided kvconfig will be passed to its constructor. Note that currently the values of kvconfig must be able to be marshaled and unmarshaled using the encoding/json library (used when reading/writing the index metadata file).

func Open

func Open(path string) (Index, error)

Open index at the specified path, must exist. The mapping used when it was created will be used for all Index/Search operations.

func OpenUsing

func OpenUsing(path string, runtimeConfig map[string]interface{}) (Index, error)

OpenUsing opens index at the specified path, must exist. The mapping used when it was created will be used for all Index/Search operations. The provided runtimeConfig can override settings persisted when the kvstore was created.

type IndexAlias

type IndexAlias interface {
	Index

	Add(i ...Index)
	Remove(i ...Index)
	Swap(in, out []Index)
}

An IndexAlias is a wrapper around one or more Index objects. It has two distinct modes of operation. 1. When it points to a single index, ALL index operations are valid and will be passed through to the underlying index. 2. When it points to more than one index, the only valid operation is Search. In this case the search will be performed across all the underlying indexes and the results merged. Calls to Add/Remove/Swap the underlying indexes are atomic, so you can safely change the underlying Index objects while other components are performing operations.

type IndexErrMap

type IndexErrMap map[string]error

IndexErrMap tracks errors with the name of the index where it occurred

func (IndexErrMap) MarshalJSON

func (iem IndexErrMap) MarshalJSON() ([]byte, error)

MarshalJSON seralizes the error into a string for JSON consumption

func (IndexErrMap) UnmarshalJSON added in v0.3.0

func (iem IndexErrMap) UnmarshalJSON(data []byte) error

type IndexStat

type IndexStat struct {
	// contains filtered or unexported fields
}

func (*IndexStat) MarshalJSON

func (is *IndexStat) MarshalJSON() ([]byte, error)

type IndexStats

type IndexStats struct {
	// contains filtered or unexported fields
}

func NewIndexStats

func NewIndexStats() *IndexStats

func (*IndexStats) Register

func (i *IndexStats) Register(index Index)

func (*IndexStats) String

func (i *IndexStats) String() string

func (*IndexStats) UnRegister

func (i *IndexStats) UnRegister(index Index)

type SearchQueryEndCallbackFn added in v0.8.0

type SearchQueryEndCallbackFn func(size uint64) error

type SearchQueryStartCallbackFn added in v0.8.0

type SearchQueryStartCallbackFn func(size uint64) error

type SearchRequest

type SearchRequest struct {
	Query            query.Query       `json:"query"`
	Size             int               `json:"size"`
	From             int               `json:"from"`
	Highlight        *HighlightRequest `json:"highlight"`
	Fields           []string          `json:"fields"`
	Facets           FacetsRequest     `json:"facets"`
	Explain          bool              `json:"explain"`
	Sort             search.SortOrder  `json:"sort"`
	IncludeLocations bool              `json:"includeLocations"`
	Score            string            `json:"score,omitempty"`
	SearchAfter      []string          `json:"search_after"`
	SearchBefore     []string          `json:"search_before"`
}

A SearchRequest describes all the parameters needed to search the index. Query is required. Size/From describe how much and which part of the result set to return. Highlight describes optional search result highlighting. Fields describes a list of field values which should be retrieved for result documents, provided they were stored while indexing. Facets describe the set of facets to be computed. Explain triggers inclusion of additional search result score explanations. Sort describes the desired order for the results to be returned. Score controls the kind of scoring performed SearchAfter supports deep paging by providing a minimum sort key SearchBefore supports deep paging by providing a maximum sort key

A special field named "*" can be used to return all fields.

func NewSearchRequest

func NewSearchRequest(q query.Query) *SearchRequest

NewSearchRequest creates a new SearchRequest for the Query, using default values for all other search parameters.

Example
// finds documents with fields fully matching the given query text
query := NewMatchQuery("named one")
searchRequest := NewSearchRequest(query)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
Output:

document id 1

func NewSearchRequestOptions

func NewSearchRequestOptions(q query.Query, size, from int, explain bool) *SearchRequest

NewSearchRequestOptions creates a new SearchRequest for the Query, with the requested size, from and explanation search parameters. By default results are ordered by score, descending.

func (*SearchRequest) AddFacet

func (r *SearchRequest) AddFacet(facetName string, f *FacetRequest)

AddFacet adds a FacetRequest to this SearchRequest

Example
facet := NewFacetRequest("Name", 1)
query := NewMatchAllQuery()
searchRequest := NewSearchRequest(query)
searchRequest.AddFacet("facet name", facet)
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

// total number of terms
fmt.Println(searchResults.Facets["facet name"].Total)
// numer of docs with no value for this field
fmt.Println(searchResults.Facets["facet name"].Missing)
// term with highest occurrences in field name
fmt.Println(searchResults.Facets["facet name"].Terms[0].Term)
Output:

5
2
one

func (*SearchRequest) SetSearchAfter added in v0.8.0

func (r *SearchRequest) SetSearchAfter(after []string)

SetSearchAfter sets the request to skip over hits with a sort value less than the provided sort after key

func (*SearchRequest) SetSearchBefore added in v0.8.0

func (r *SearchRequest) SetSearchBefore(before []string)

SetSearchBefore sets the request to skip over hits with a sort value greater than the provided sort before key

func (*SearchRequest) SortBy added in v0.4.0

func (r *SearchRequest) SortBy(order []string)

SortBy changes the request to use the requested sort order this form uses the simplified syntax with an array of strings each string can either be a field name or the magic value _id and _score which refer to the doc id and search score any of these values can optionally be prefixed with - to reverse the order

Example
// find docs containing "one", order by Age instead of score
query := NewMatchQuery("one")
searchRequest := NewSearchRequest(query)
searchRequest.SortBy([]string{"Age"})
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
fmt.Println(searchResults.Hits[1].ID)
Output:

document id 2
document id 1

func (*SearchRequest) SortByCustom added in v0.4.0

func (r *SearchRequest) SortByCustom(order search.SortOrder)

SortByCustom changes the request to use the requested sort order

Example
// find all docs, order by Age, with docs missing Age field first
query := NewMatchAllQuery()
searchRequest := NewSearchRequest(query)
searchRequest.SortByCustom(search.SortOrder{
	&search.SortField{
		Field:   "Age",
		Missing: search.SortFieldMissingFirst,
	},
})
searchResults, err := exampleIndex.Search(searchRequest)
if err != nil {
	panic(err)
}

fmt.Println(searchResults.Hits[0].ID)
fmt.Println(searchResults.Hits[1].ID)
fmt.Println(searchResults.Hits[2].ID)
fmt.Println(searchResults.Hits[3].ID)
Output:

document id 3
document id 4
document id 2
document id 1

func (*SearchRequest) UnmarshalJSON

func (r *SearchRequest) UnmarshalJSON(input []byte) error

UnmarshalJSON deserializes a JSON representation of a SearchRequest

func (*SearchRequest) Validate added in v0.3.0

func (r *SearchRequest) Validate() error

type SearchResult

type SearchResult struct {
	Status   *SearchStatus                  `json:"status"`
	Request  *SearchRequest                 `json:"request"`
	Hits     search.DocumentMatchCollection `json:"hits"`
	Total    uint64                         `json:"total_hits"`
	MaxScore float64                        `json:"max_score"`
	Took     time.Duration                  `json:"took"`
	Facets   search.FacetResults            `json:"facets"`
}

A SearchResult describes the results of executing a SearchRequest.

func MultiSearch

func MultiSearch(ctx context.Context, req *SearchRequest, indexes ...Index) (*SearchResult, error)

MultiSearch executes a SearchRequest across multiple Index objects, then merges the results. The indexes must honor any ctx deadline.

func (*SearchResult) Merge

func (sr *SearchResult) Merge(other *SearchResult)

Merge will merge together multiple SearchResults during a MultiSearch

func (*SearchResult) Size added in v0.8.0

func (sr *SearchResult) Size() int

func (*SearchResult) String

func (sr *SearchResult) String() string

type SearchStatus

type SearchStatus struct {
	Total      int         `json:"total"`
	Failed     int         `json:"failed"`
	Successful int         `json:"successful"`
	Errors     IndexErrMap `json:"errors,omitempty"`
}

SearchStatus is a secion in the SearchResult reporting how many underlying indexes were queried, how many were successful/failed and a map of any errors that were encountered

func (*SearchStatus) Merge

func (ss *SearchStatus) Merge(other *SearchStatus)

Merge will merge together multiple SearchStatuses during a MultiSearch

Directories

Path Synopsis
lang/en
Package en implements an analyzer with reasonable defaults for processing English text.
Package en implements an analyzer with reasonable defaults for processing English text.
token/lowercase
Package lowercase implements a TokenFilter which converts tokens to lower case according to unicode rules.
Package lowercase implements a TokenFilter which converts tokens to lower case according to unicode rules.
token/stop
Package stop implements a TokenFilter removing tokens found in a TokenMap.
Package stop implements a TokenFilter removing tokens found in a TokenMap.
tokenizer/exception
package exception implements a Tokenizer which extracts pieces matched by a regular expression from the input data, delegates the rest to another tokenizer, then insert back extracted parts in the token stream.
package exception implements a Tokenizer which extracts pieces matched by a regular expression from the input data, delegates the rest to another tokenizer, then insert back extracted parts in the token stream.
tokenmap
package token_map implements a generic TokenMap, often used in conjunction with filters to remove or process specific tokens.
package token_map implements a generic TokenMap, often used in conjunction with filters to remove or process specific tokens.
cmd
scorch/mergeplan
Package mergeplan provides a segment merge planning approach that's inspired by Lucene's TieredMergePolicy.java and descriptions like http://blog.mikemccandless.com/2011/02/visualizing-lucenes-segment-merges.html
Package mergeplan provides a segment merge planning approach that's inspired by Lucene's TieredMergePolicy.java and descriptions like http://blog.mikemccandless.com/2011/02/visualizing-lucenes-segment-merges.html
store/boltdb
Package boltdb implements a store.KVStore on top of BoltDB.
Package boltdb implements a store.KVStore on top of BoltDB.
store/gtreap
Package gtreap provides an in-memory implementation of the KVStore interfaces using the gtreap balanced-binary treap, copy-on-write data structure.
Package gtreap provides an in-memory implementation of the KVStore interfaces using the gtreap balanced-binary treap, copy-on-write data structure.
store/metrics
Package metrics provides a bleve.store.KVStore implementation that wraps another, real KVStore implementation, and uses go-metrics to track runtime performance metrics.
Package metrics provides a bleve.store.KVStore implementation that wraps another, real KVStore implementation, and uses go-metrics to track runtime performance metrics.
upsidedown
Package upsidedown is a generated protocol buffer package.
Package upsidedown is a generated protocol buffer package.

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL