#[macro_use] extern crate pest_derive; mod criterion; mod error; mod external_documents_ids; mod fields_ids_map; mod search; pub mod facet; pub mod heed_codec; pub mod index; pub mod proximity; pub mod tree_level; pub mod update; use std::borrow::Cow; use std::collections::HashMap; use std::hash::BuildHasherDefault; use std::result::Result as StdResult; use fxhash::{FxHasher32, FxHasher64}; use serde_json::{Map, Value}; pub use self::criterion::{Criterion, default_criteria}; pub use self::error::Error; pub use self::external_documents_ids::ExternalDocumentsIds; pub use self::fields_ids_map::FieldsIdsMap; pub use self::heed_codec::{BEU32StrCodec, StrStrU8Codec, StrLevelPositionCodec, ObkvCodec, FieldIdWordCountCodec}; pub use self::heed_codec::{RoaringBitmapCodec, BoRoaringBitmapCodec, CboRoaringBitmapCodec}; pub use self::heed_codec::{RoaringBitmapLenCodec, BoRoaringBitmapLenCodec, CboRoaringBitmapLenCodec}; pub use self::index::Index; pub use self::search::{Search, FacetDistribution, FilterCondition, SearchResult, MatchingWords}; pub use self::tree_level::TreeLevel; pub type Result = std::result::Result; pub type FastMap4 = HashMap>; pub type FastMap8 = HashMap>; pub type SmallString32 = smallstr::SmallString<[u8; 32]>; pub type SmallVec16 = smallvec::SmallVec<[T; 16]>; pub type SmallVec32 = smallvec::SmallVec<[T; 32]>; pub type SmallVec8 = smallvec::SmallVec<[T; 8]>; pub type BEU32 = heed::zerocopy::U32; pub type BEU64 = heed::zerocopy::U64; pub type Attribute = u32; pub type DocumentId = u32; pub type FieldId = u8; pub type Position = u32; pub type FieldsDistribution = HashMap; type MergeFn = for<'a> fn(&[u8], &[Cow<'a, [u8]>]) -> StdResult, E>; /// Transform a raw obkv store into a JSON Object. pub fn obkv_to_json( displayed_fields: &[FieldId], fields_ids_map: &FieldsIdsMap, obkv: obkv::KvReader, ) -> Result> { displayed_fields.iter() .copied() .flat_map(|id| obkv.get(id).map(|value| (id, value))) .map(|(id, value)| { let name = fields_ids_map.name(id).ok_or(error::FieldIdMapMissingEntry::FieldId { field_id: id, from_db_name: "documents", })?; let value = serde_json::from_slice(value).map_err(error::InternalError::SerdeJson)?; Ok((name.to_owned(), value)) }) .collect() } /// Transform a JSON value into a string that can be indexed. pub fn json_to_string(value: &Value) -> Option { fn inner(value: &Value, output: &mut String) -> bool { use std::fmt::Write; match value { Value::Null => false, Value::Bool(boolean) => write!(output, "{}", boolean).is_ok(), Value::Number(number) => write!(output, "{}", number).is_ok(), Value::String(string) => write!(output, "{}", string).is_ok(), Value::Array(array) => { let mut count = 0; for value in array { if inner(value, output) { output.push_str(". "); count += 1; } } // check that at least one value was written count != 0 }, Value::Object(object) => { let mut buffer = String::new(); let mut count = 0; for (key, value) in object { buffer.clear(); let _ = write!(&mut buffer, "{}: ", key); if inner(value, &mut buffer) { buffer.push_str(". "); // We write the "key: value. " pair only when // we are sure that the value can be written. output.push_str(&buffer); count += 1; } } // check that at least one value was written count != 0 }, } } let mut string = String::new(); if inner(value, &mut string) { Some(string) } else { None } } #[cfg(test)] mod tests { use super::*; use serde_json::json; #[test] fn json_to_string_object() { let value = json!({ "name": "John Doe", "age": 43, "not_there": null, }); let string = json_to_string(&value).unwrap(); assert_eq!(string, "name: John Doe. age: 43. "); } #[test] fn json_to_string_array() { let value = json!([ { "name": "John Doe" }, 43, "hello", [ "I", "am", "fine" ], null, ]); let string = json_to_string(&value).unwrap(); // We don't care about having two point (.) after the other as // the distance of hard separators is clamped to 8 anyway. assert_eq!(string, "name: John Doe. . 43. hello. I. am. fine. . "); } }