Expand description
Schema definition for tantivy’s indices.
Setting your schema in Tantivy
Tantivy has a very strict schema. The schema defines information about the fields your index contains, that is, for each field:
- the field name (may only contain letters
[a-zA-Z]
, number[0-9]
, and_
) - the type of the field (currently only
text
andu64
are supported) - how the field should be indexed / stored.
This very last point is critical as it will enable / disable some of the functionality for your index.
Tantivy’s schema is stored within the meta.json
file at the root of your
directory.
Building a schema “programmatically”
Setting a text field
Example
use tantivy::schema::*;
let mut schema_builder = Schema::builder();
let title_options = TextOptions::default()
.set_stored()
.set_indexing_options(TextFieldIndexing::default()
.set_tokenizer("default")
.set_index_option(IndexRecordOption::WithFreqsAndPositions));
schema_builder.add_text_field("title", title_options);
let schema = schema_builder.build();
We can split the problem of generating a search result page into two phases:
- identifying the list of 10 or so documents to be displayed (Conceptually
query -> doc_ids[]
) - for each of these documents, retrieving the information required to generate the search
results page. (
doc_ids[] -> Document[]
)
In the first phase, the ability to search for documents by the given field is determined by the
IndexRecordOption
of our TextOptions
.
The effect of each possible setting is described more in detail in TextOptions
.
On the other hand setting the field as stored or not determines whether the field should be
returned when Searcher::doc()
is called.
Setting a u64, a i64 or a f64 field
Example
use tantivy::schema::*;
let mut schema_builder = Schema::builder();
let num_stars_options = NumericOptions::default()
.set_stored()
.set_indexed();
schema_builder.add_u64_field("num_stars", num_stars_options);
let schema = schema_builder.build();
Just like for Text fields (see above),
setting the field as stored defines whether the field will be
returned when Searcher::doc()
is called,
and setting the field as indexed means that we will be able perform queries such as
num_stars:10
. Note that unlike text fields, numeric fields can only be indexed in one way for
the moment.
Shortcuts
For convenience, it is possible to define your field indexing options by combining different
flags using the |
operator.
For instance, a schema containing the two fields defined in the example above could be rewritten:
use tantivy::schema::*;
let mut schema_builder = Schema::builder();
schema_builder.add_u64_field("num_stars", INDEXED | STORED);
schema_builder.add_text_field("title", TEXT | STORED);
let schema = schema_builder.build();
Fast fields
This functionality is somewhat similar to Lucene’s DocValues
.
Fields that are indexed as FAST
will be stored in a special data structure that will
make it possible to access the value given the doc id rapidly. This is useful if the value
of the field is required during scoring or collection for instance.
use tantivy::schema::*;
let mut schema_builder = Schema::builder();
schema_builder.add_u64_field("population", STORED | FAST);
schema_builder.add_text_field("zip_code", STRING | FAST);
let schema = schema_builder.build();
Structs
Field
is represented by an unsigned 32-bit integer type.
The schema holds the mapping between field names and Field
objects.FieldEntry
represents a field and its configuration.
Schema
are a collection of FieldEntry
FieldValue
holds together a Field
and its Value
.JsonObjectOptions
make it possible to
configure how a json object field should be indexed and stored.u64
, i64
, or f64
field should be handled by tantivy.Enums
FieldType
describes the type (text, u64) of a field as well as
how it should be handled by tantivy.IndexRecordOption
describes an amount information associated
with a given indexed field.Constants
Traits
Functions
field_name
.
Returns true if the name can be use for a field name.Type Definitions
NumericOptions
instead.