feat: Add case_insensitive param to some Elastic DSL queries

docs/reference/es_compatible_api.md

@@ -695,10 +695,11 @@ When working on text, it is recommended to only use `term` queries on fields con
 
 #### Supported Parameters
 
-| Variable | Type     | Description                                                                  | Default |
-| -------- | -------- | ---------------------------------------------------------------------------- | ------- |
-| `value`  | String   | Term value. This is the string representation of a token after tokenization. | -       |
-| `boost`  | `Number` | Multiplier boost for score computation                                       | 1.0     |
+| Variable           | Type    | Description                                                                  | Default |
+| ------------------ | ------- | ---------------------------------------------------------------------------- | ------- |
+| `value`            | String  | Term value. This is the string representation of a token after tokenization. | -       |
+| `boost`            | Number  | Multiplier boost for score computation                                       | 1.0     |
+| `case_insensitive` | Boolean | Allows ASCII case insensitive matching of the value.                         | false   |
 
 
 
@@ -763,9 +764,10 @@ Returns documents that contain a specific prefix in a provided field.
 
 #### Supported Parameters
 
-| Variable | Type   | Description                                     | Default |
-| -------- | ------ | ----------------------------------------------- | ------- |
-| `value`  | String | Beginning characters of terms you wish to find. | -       |
+| Variable           | Type    | Description                                          | Default |
+| ------------------ | ------- | ---------------------------------------------------- | ------- |
+| `value`            | String  | Beginning characters of terms you wish to find.      | -       |
+| `case_insensitive` | Boolean | Allows ASCII case insensitive matching of the value. | false   |
 
 ### `wildcard`
 
@@ -791,9 +793,39 @@ Returns documents that contain terms matching a wildcard pattern:
 
 #### Supported Parameters
 
-| Variable | Type   | Description                                  | Default |
-| -------- | ------ | -------------------------------------------- | ------- |
-| `value`  | String | Wildcard pattern for terms you wish to find. | -       |
+| Variable           | Type    | Description                                          | Default |
+| ------------------ | ------- | ---------------------------------------------------- | ------- |
+| `value`            | String  | Wildcard pattern for terms you wish to find.         | -       |
+| `boost`            | Number  | Multiplier boost for score computation.              | 1.0     |
+| `case_insensitive` | Boolean | Allows ASCII case insensitive matching of the value. | false   |
+
+
+### `regexp`
+
+[Elasticsearch reference documentation](https://www.elastic.co/guide/en/elasticsearch/reference/8.8/query-dsl-regexp-query.html)
+
+Returns documents that contain terms matching a regular expression.
+
+#### Example
+
+```json
+{
+  "query": {
+    "regexp": {
+      "author.login" {
+        "value": "adm.*n",
+      }
+    }
+  }
+}
+```
+
+#### Supported Parameters
+
+| Variable           | Type    | Description                                          | Default |
+| ------------------ | ------- | ---------------------------------------------------- | ------- |
+| `value`            | String  | Wildcard pattern for terms you wish to find.         | -       |
+| `case_insensitive` | Boolean | Allows ASCII case insensitive matching of the value. | false   |
 
 
 ### About the `lenient` argument

quickwit/quickwit-query/src/elastic_query_dsl/prefix_query.rs

@@ -29,6 +29,8 @@ pub(crate) struct PrefixQuery {
 #[serde(deny_unknown_fields)]
 pub struct PrefixQueryParams {
     value: String,
+    #[serde(default)]
+    case_insensitive: bool,
 }
 
 impl ConvertibleToQueryAst for PrefixQuery {
@@ -45,6 +47,7 @@ impl ConvertibleToQueryAst for PrefixQuery {
             field: self.field,
             value: wildcard,
             lenient: true,
+            case_insensitive: self.params.case_insensitive,
         }
         .into())
     }
@@ -64,7 +67,10 @@ impl From<OneFieldMap<StringOrStructForSerialization<PrefixQueryParams>>> for Pr
 
 impl From<String> for PrefixQueryParams {
     fn from(value: String) -> PrefixQueryParams {
-        PrefixQueryParams { value }
+        PrefixQueryParams {
+            value,
+            case_insensitive: false,
+        }
     }
 }
 

quickwit/quickwit-query/src/elastic_query_dsl/regex_query.rs

@@ -22,16 +22,22 @@ use crate::query_ast::{QueryAst, RegexQuery as AstRegexQuery};
 #[serde(deny_unknown_fields)]
 pub struct RegexQueryParams {
     value: String,
-    // we could probably add case_insensitive
+    #[serde(default)]
+    case_insensitive: bool,
 }
 
 pub type RegexQuery = OneFieldMap<RegexQueryParams>;
 
 impl ConvertibleToQueryAst for RegexQuery {
     fn convert_to_query_ast(self) -> anyhow::Result<QueryAst> {
+        let regex = if self.value.case_insensitive {
+            format!("(?i){}", self.value.value)
+        } else {
+            self.value.value.clone()
+        };
         Ok(AstRegexQuery {
             field: self.field,
-            regex: self.value.value,
+            regex,
         }
         .into())
     }

quickwit/quickwit-query/src/elastic_query_dsl/term_query.rs

@@ -41,6 +41,7 @@ impl From<String> for TermQueryParams {
         TermQueryParams {
             value: query,
             boost: None,
+            case_insensitive: false,
         }
     }
 }
@@ -70,6 +71,8 @@ pub struct TermQueryParams {
     pub value: String,
     #[serde(default)]
     pub boost: Option<NotNaNf32>,
+    #[serde(default)]
+    case_insensitive: bool,
 }
 
 pub fn term_query_from_field_value(field: impl ToString, value: impl ToString) -> TermQuery {
@@ -78,6 +81,7 @@ pub fn term_query_from_field_value(field: impl ToString, value: impl ToString) -
         value: TermQueryParams {
             value: value.to_string(),
             boost: None,
+            case_insensitive: false,
         },
     }
 }
@@ -90,7 +94,20 @@ impl From<TermQuery> for ElasticQueryDslInner {
 
 impl ConvertibleToQueryAst for TermQuery {
     fn convert_to_query_ast(self) -> anyhow::Result<QueryAst> {
-        let TermQueryParams { value, boost } = self.value;
+        let TermQueryParams {
+            value,
+            boost,
+            case_insensitive,
+        } = self.value;
+        if case_insensitive {
+            let ci_value = format!("(?i){}", regex::escape(&value));
+            let term_ast: QueryAst = query_ast::RegexQuery {
+                field: self.field,
+                regex: ci_value,
+            }
+            .into();
+            return Ok(term_ast.boost(boost));
+        }
         let term_ast: QueryAst = query_ast::TermQuery {
             field: self.field,
             value,

quickwit/quickwit-query/src/elastic_query_dsl/wildcard_query.rs

@@ -32,6 +32,8 @@ pub struct WildcardQueryParams {
     value: String,
     #[serde(default)]
     pub boost: Option<NotNaNf32>,
+    #[serde(default)]
+    case_insensitive: bool,
 }
 
 impl ConvertibleToQueryAst for WildcardQuery {
@@ -40,6 +42,7 @@ impl ConvertibleToQueryAst for WildcardQuery {
             field: self.field,
             value: self.params.value,
             lenient: true,
+            case_insensitive: self.params.case_insensitive,
         }
         .into();
         Ok(wildcard_ast.boost(self.params.boost))
@@ -60,7 +63,11 @@ impl From<OneFieldMap<StringOrStructForSerialization<WildcardQueryParams>>> for
 
 impl From<String> for WildcardQueryParams {
     fn from(value: String) -> WildcardQueryParams {
-        WildcardQueryParams { value, boost: None }
+        WildcardQueryParams {
+            value,
+            boost: None,
+            case_insensitive: false,
+        }
     }
 }
 

quickwit/quickwit-query/src/query_ast/user_input_query.rs

@@ -293,6 +293,7 @@ fn convert_user_input_literal(
                     field: field_name,
                     value: phrase.clone(),
                     lenient,
+                    case_insensitive: false,
                 }
                 .into()
             } else {

quickwit/quickwit-query/src/query_ast/wildcard_query.rs

@@ -32,6 +32,7 @@ pub struct WildcardQuery {
     pub value: String,
     /// Support missing fields
     pub lenient: bool,
+    pub case_insensitive: bool,
 }
 
 impl From<WildcardQuery> for QueryAst {
@@ -133,6 +134,11 @@ impl WildcardQuery {
                 let tokenizer_name = text_field_indexing.tokenizer();
                 let regex =
                     sub_query_parts_to_regex(sub_query_parts, tokenizer_name, tokenizer_manager)?;
+                let regex = if self.case_insensitive {
+                    format!("(?i){}", regex)
+                } else {
+                    regex
+                };
 
                 Ok((field, None, regex))
             }
@@ -147,6 +153,11 @@ impl WildcardQuery {
                 let tokenizer_name = text_field_indexing.tokenizer();
                 let regex =
                     sub_query_parts_to_regex(sub_query_parts, tokenizer_name, tokenizer_manager)?;
+                let regex = if self.case_insensitive {
+                    format!("(?i){}", regex)
+                } else {
+                    regex
+                };
 
                 let mut term_for_path = Term::from_field_json_path(
                     field,
@@ -219,6 +230,7 @@ mod tests {
             field: "text_field".to_string(),
             value: "MyString Wh1ch?a.nOrMal Tokenizer would*cut".to_string(),
             lenient: false,
+            case_insensitive: false,
         };
 
         let tokenizer_manager = create_default_quickwit_tokenizer_manager();
@@ -261,6 +273,7 @@ mod tests {
             field: "text_field".to_string(),
             value: "MyString Wh1ch\\?a.nOrMal Tokenizer would\\*cut".to_string(),
             lenient: false,
+            case_insensitive: false,
         };
 
         let tokenizer_manager = create_default_quickwit_tokenizer_manager();
@@ -305,6 +318,7 @@ mod tests {
             field: "json_field.Inner.Fie*ld".to_string(),
             value: "MyString Wh1ch?a.nOrMal Tokenizer would*cut".to_string(),
             lenient: false,
+            case_insensitive: false,
         };
 
         let tokenizer_manager = create_default_quickwit_tokenizer_manager();
@@ -347,6 +361,7 @@ mod tests {
             field: "my_missing_field".to_string(),
             value: "My query value*".to_string(),
             lenient: false,
+            case_insensitive: false,
         };
         let tokenizer_manager = create_default_quickwit_tokenizer_manager();
         let schema = single_text_field_schema("my_field", "whitespace");
@@ -359,4 +374,47 @@ mod tests {
         };
         assert_eq!(missing_field_full_path, "my_missing_field");
     }
+
+    #[test]
+    fn test_wildcard_query_to_regex_on_text_case_insensitive() {
+        let query = WildcardQuery {
+            field: "text_field".to_string(),
+            value: "MyString Wh1ch?a.nOrMal Tokenizer would*cut".to_string(),
+            lenient: false,
+            case_insensitive: true,
+        };
+
+        let tokenizer_manager = create_default_quickwit_tokenizer_manager();
+        for tokenizer in ["raw", "whitespace"] {
+            let mut schema_builder = TantivySchema::builder();
+            let text_options = TextOptions::default()
+                .set_indexing_options(TextFieldIndexing::default().set_tokenizer(tokenizer));
+            schema_builder.add_text_field("text_field", text_options);
+            let schema = schema_builder.build();
+
+            let (_field, path, regex) = query.to_regex(&schema, &tokenizer_manager).unwrap();
+            assert_eq!(regex, "(?i)MyString Wh1ch.a\\.nOrMal Tokenizer would.*cut");
+            assert!(path.is_none());
+        }
+
+        for tokenizer in [
+            "raw_lowercase",
+            "lowercase",
+            "default",
+            "en_stem",
+            "chinese_compatible",
+            "source_code_default",
+            "source_code_with_hex",
+        ] {
+            let mut schema_builder = TantivySchema::builder();
+            let text_options = TextOptions::default()
+                .set_indexing_options(TextFieldIndexing::default().set_tokenizer(tokenizer));
+            schema_builder.add_text_field("text_field", text_options);
+            let schema = schema_builder.build();
+
+            let (_field, path, regex) = query.to_regex(&schema, &tokenizer_manager).unwrap();
+            assert_eq!(regex, "(?i)mystring wh1ch.a\\.normal tokenizer would.*cut");
+            assert!(path.is_none());
+        }
+    }
 }

quickwit/rest-api-tests/scenarii/es_compatibility/0006-term_query.yaml

@@ -1,5 +1,3 @@
-# case_insensitive not supported.
-engines: ["elasticsearch"]
 params:
   # this overrides the query sent in body apparently
   size: 3

quickwit/rest-api-tests/scenarii/es_compatibility/0029-wildcard.yaml

@@ -26,3 +26,25 @@ expected:
   hits:
     total:
       value: 2
+---
+json:
+  query:
+    wildcard:
+      repo.name:
+        value: RUS*
+        case_insensitive: true
+expected:
+  hits:
+    total:
+      value: 1
+---
+json:
+  query:
+    wildcard:
+      repo.name:
+        value: RUS*
+        case_insensitive: false
+expected:
+  hits:
+    total:
+      value: 0

quickwit/rest-api-tests/scenarii/es_compatibility/0030-prefix.yaml

@@ -26,3 +26,25 @@ expected:
   hits:
     total:
       value: 2
+---
+json:
+  query:
+    prefix:
+      repo.name:
+        value: RUST
+        case_insensitive: true
+expected:
+  hits:
+    total:
+      value: 1
+---
+json:
+  query:
+    prefix:
+      repo.name:
+        value: RUST
+        case_insensitive: false
+expected:
+  hits:
+    total:
+      value: 0