Merge #2259

2259: disable typos on words r=MarinPostma a=MarinPostma

Introduce the disable typo setting as per https://github.com/meilisearch/specifications/pull/117.

waiting for https://github.com/meilisearch/milli/pull/474.


Co-authored-by: ad hoc <postma.marin@protonmail.com>

meilisearch-error/src/lib.rs

@@ -120,6 +120,7 @@ pub enum Code {
     IndexAlreadyExists,
     IndexNotFound,
     InvalidIndexUid,
+    InvalidMinWordLengthForTypo,
 
     // invalid state error
     InvalidState,
@@ -271,6 +272,9 @@ impl Code {
             InvalidApiKeyDescription => {
                 ErrCode::invalid("invalid_api_key_description", StatusCode::BAD_REQUEST)
             }
+            InvalidMinWordLengthForTypo => {
+                ErrCode::invalid("invalid_min_word_length_for_typo", StatusCode::BAD_REQUEST)
+            }
         }
     }
 

meilisearch-lib/src/error.rs

@@ -41,7 +41,9 @@ impl ErrorCode for MilliError<'_> {
                     UserError::CriterionError(_) => Code::InvalidRankingRule,
                     UserError::InvalidGeoField { .. } => Code::InvalidGeoField,
                     UserError::SortError(_) => Code::Sort,
-                    UserError::InvalidMinTypoWordLenSetting(_, _) => unreachable!(),
+                    UserError::InvalidMinTypoWordLenSetting(_, _) => {
+                        Code::InvalidMinWordLengthForTypo
+                    }
                 }
             }
         }

meilisearch-lib/src/index/index.rs

@@ -5,6 +5,7 @@ use std::ops::Deref;
 use std::path::Path;
 use std::sync::Arc;
 
+use fst::IntoStreamer;
 use milli::heed::{EnvOpenOptions, RoTxn};
 use milli::update::{IndexerConfig, Setting};
 use milli::{obkv_to_json, FieldDistribution, FieldId};
@@ -17,7 +18,7 @@ use crate::EnvSizer;
 
 use super::error::IndexError;
 use super::error::Result;
-use super::updates::TypoSettings;
+use super::updates::{MinWordLengthTypoSetting, TypoSettings};
 use super::{Checked, Settings};
 
 pub type Document = Map<String, Value>;
@@ -169,8 +170,22 @@ impl Index {
             })
             .collect();
 
+        let min_typo_word_len = MinWordLengthTypoSetting {
+            one_typo: Setting::Set(self.min_word_len_one_typo(txn)?),
+            two_typos: Setting::Set(self.min_word_len_two_typos(txn)?),
+        };
+
+        let disabled_words = self
+            .exact_words(txn)?
+            .into_stream()
+            .into_strs()?
+            .into_iter()
+            .collect();
+
         let typo_tolerance = TypoSettings {
             enabled: Setting::Set(self.authorize_typos(txn)?),
+            min_word_length_for_typo: Setting::Set(min_typo_word_len),
+            disable_on_words: Setting::Set(disabled_words),
         };
 
         Ok(Settings {

meilisearch-lib/src/index/updates.rs

@@ -37,14 +37,33 @@ pub struct Checked;
 #[derive(Clone, Default, Debug, Serialize, Deserialize, PartialEq)]
 pub struct Unchecked;
 
+#[cfg_attr(test, derive(proptest_derive::Arbitrary))]
 #[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq)]
 #[serde(deny_unknown_fields)]
 #[serde(rename_all = "camelCase")]
+pub struct MinWordLengthTypoSetting {
+    #[cfg_attr(test, proptest(strategy = "test::setting_strategy()"))]
+    #[serde(default, skip_serializing_if = "Setting::is_not_set")]
+    pub one_typo: Setting<u8>,
+    #[cfg_attr(test, proptest(strategy = "test::setting_strategy()"))]
+    #[serde(default, skip_serializing_if = "Setting::is_not_set")]
+    pub two_typos: Setting<u8>,
+}
+
 #[cfg_attr(test, derive(proptest_derive::Arbitrary))]
+#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq)]
+#[serde(deny_unknown_fields)]
+#[serde(rename_all = "camelCase")]
 pub struct TypoSettings {
     #[cfg_attr(test, proptest(strategy = "test::setting_strategy()"))]
     #[serde(default, skip_serializing_if = "Setting::is_not_set")]
     pub enabled: Setting<bool>,
+    #[cfg_attr(test, proptest(strategy = "test::setting_strategy()"))]
+    #[serde(default, skip_serializing_if = "Setting::is_not_set")]
+    pub min_word_length_for_typo: Setting<MinWordLengthTypoSetting>,
+    #[cfg_attr(test, proptest(strategy = "test::setting_strategy()"))]
+    #[serde(default, skip_serializing_if = "Setting::is_not_set")]
+    pub disable_on_words: Setting<BTreeSet<String>>,
 }
 /// Holds all the settings for an index. `T` can either be `Checked` if they represents settings
 /// whose validity is guaranteed, or `Unchecked` if they need to be validated. In the later case, a
@@ -352,14 +371,44 @@ pub fn apply_settings_to_builder(
     }
 
     match settings.typo {
-        Setting::Set(ref value) => match value.enabled {
+        Setting::Set(ref value) => {
-            Setting::Set(val) => builder.set_autorize_typos(val),
+            match value.enabled {
-            Setting::Reset => builder.reset_authorize_typos(),
+                Setting::Set(val) => builder.set_autorize_typos(val),
-            Setting::NotSet => (),
+                Setting::Reset => builder.reset_authorize_typos(),
-        },
+                Setting::NotSet => (),
+            }
+            match value.min_word_length_for_typo {
+                Setting::Set(ref setting) => {
+                    match setting.one_typo {
+                        Setting::Set(val) => builder.set_min_word_len_one_typo(val),
+                        Setting::Reset => builder.reset_min_word_len_one_typo(),
+                        Setting::NotSet => (),
+                    }
+                    match setting.two_typos {
+                        Setting::Set(val) => builder.set_min_word_len_two_typos(val),
+                        Setting::Reset => builder.reset_min_word_len_two_typos(),
+                        Setting::NotSet => (),
+                    }
+                }
+                Setting::Reset => {
+                    builder.reset_min_word_len_one_typo();
+                    builder.reset_min_word_len_two_typos();
+                }
+                Setting::NotSet => (),
+            }
+            match value.disable_on_words {
+                Setting::Set(ref words) => {
+                    builder.set_exact_words(words.clone());
+                }
+                Setting::Reset => builder.reset_exact_words(),
+                Setting::NotSet => (),
+            }
+        }
         Setting::Reset => {
             // all typo settings need to be reset here.
             builder.reset_authorize_typos();
+            builder.reset_min_word_len_one_typo();
+            builder.reset_min_word_len_two_typos();
         }
         Setting::NotSet => (),
     }