Merge branch 'master' into expose-regex-query

.github/workflows/codeql.yml

@@ -47,7 +47,7 @@ jobs:
 
       # Initializes the CodeQL tools for scanning.
       - name: Initialize CodeQL
-        uses: github/codeql-action/init@df5a14dc28094dc936e103b37d749c6628682b60 # v2.2.5
+        uses: github/codeql-action/init@8f596b4ae3cb3c588a5c46780b86dd53fef16c52 # v2.2.5
         with:
           languages: ${{ matrix.language }}
           # If you wish to specify custom queries, you can do so here or in a config file.
@@ -60,7 +60,7 @@ jobs:
       # Autobuild attempts to build any compiled languages  (C/C++, C#, or Java).
       # If this step fails, then you should remove it and run the build manually (see below)
       - name: Autobuild
-        uses: github/codeql-action/autobuild@df5a14dc28094dc936e103b37d749c6628682b60 # v2.2.5
+        uses: github/codeql-action/autobuild@8f596b4ae3cb3c588a5c46780b86dd53fef16c52 # v2.2.5
 
       # ℹ️ Command-line programs to run using the OS shell.
       # 📚 See https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idstepsrun
@@ -73,6 +73,6 @@ jobs:
       #   ./location_of_script_within_repo/buildscript.sh
 
       - name: Perform CodeQL Analysis
-        uses: github/codeql-action/analyze@df5a14dc28094dc936e103b37d749c6628682b60 # v2.2.5
+        uses: github/codeql-action/analyze@8f596b4ae3cb3c588a5c46780b86dd53fef16c52 # v2.2.5
         with:
           category: "/language:${{matrix.language}}"

.github/workflows/publish.yaml

@@ -54,7 +54,7 @@ jobs:
           args: --release --sdist -o dist -i 3.8 3.9 3.10 3.11 3.12
 
       - name: Upload wheels
-        uses: actions/upload-artifact@5d5d22a31266ced268874388b861e4b58bb5c2f3  # 4.3.1
+        uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808  # 4.3.3
         with:
           name: wheels
           path: dist
@@ -84,7 +84,7 @@ jobs:
           args: --release -o dist
 
       - name: Upload wheels
-        uses: actions/upload-artifact@5d5d22a31266ced268874388b861e4b58bb5c2f3  # 4.3.1
+        uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808  # 4.3.3
         with:
           name: wheels
           path: dist
@@ -116,7 +116,7 @@ jobs:
           args: --release -o dist
 
       - name: Upload wheels
-        uses: actions/upload-artifact@5d5d22a31266ced268874388b861e4b58bb5c2f3  # 4.3.1
+        uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808  # 4.3.3
         with:
           name: wheels
           path: dist
@@ -151,7 +151,7 @@ jobs:
         with:
           python-version: 3.8
 
-      - uses: actions/download-artifact@c850b930e6ba138125429b7e5c93fc707a7f8427 # v4.1.4
+      - uses: actions/download-artifact@9c19ed7fe5d278cd354c7dfd5d3b88589c7e2395 # v4.1.6
         with:
           name: wheels
           path: wheels
@@ -175,7 +175,7 @@ jobs:
         with:
           egress-policy: audit # TODO: change to 'egress-policy: block' after couple of runs
 
-      - uses: actions/download-artifact@c850b930e6ba138125429b7e5c93fc707a7f8427 # v4.1.4
+      - uses: actions/download-artifact@9c19ed7fe5d278cd354c7dfd5d3b88589c7e2395 # v4.1.6
         with:
           name: wheels
           path: wheels

.github/workflows/scorecards.yml

@@ -56,14 +56,14 @@ jobs:
       # Upload the results as artifacts (optional). Commenting out will disable uploads of run results in SARIF
       # format to the repository Actions tab.
       - name: "Upload artifact"
-        uses: actions/upload-artifact@5d5d22a31266ced268874388b861e4b58bb5c2f3 # v4.3.1
+        uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808 # v4.3.3
         with:
           name: SARIF file
           path: results.sarif
           retention-days: 5
 
       # Upload the results to GitHub's code scanning dashboard.
       - name: "Upload to code-scanning"
-        uses: github/codeql-action/upload-sarif@df5a14dc28094dc936e103b37d749c6628682b60 # v3.25.0
+        uses: github/codeql-action/upload-sarif@8f596b4ae3cb3c588a5c46780b86dd53fef16c52 # v3.25.2
         with:
           sarif_file: results.sarif

src/lib.rs

@@ -15,7 +15,7 @@ mod snippet;
 use document::Document;
 use facet::Facet;
 use index::Index;
-use query::Query;
+use query::{Occur, Query};
 use schema::Schema;
 use schemabuilder::SchemaBuilder;
 use searcher::{DocAddress, Order, SearchResult, Searcher};
@@ -87,6 +87,7 @@ fn tantivy(_py: Python, m: &PyModule) -> PyResult<()> {
     m.add_class::<Query>()?;
     m.add_class::<Snippet>()?;
     m.add_class::<SnippetGenerator>()?;
+    m.add_class::<Occur>()?;
 
     m.add_wrapped(wrap_pymodule!(query_parser_error))?;
 

src/query.rs

@@ -1,13 +1,56 @@
 use crate::{get_field, make_term, to_pyerr, Schema};
-use pyo3::{exceptions, prelude::*, types::PyAny, types::PyString};
+use pyo3::{
+    exceptions, prelude::*, types::PyAny, types::PyString, types::PyTuple,
+};
 use tantivy as tv;
 
+/// Custom Tuple struct to represent a pair of Occur and Query
+/// for the BooleanQuery
+struct OccurQueryPair(Occur, Query);
+
+impl<'source> FromPyObject<'source> for OccurQueryPair {
+    fn extract(ob: &'source PyAny) -> PyResult<Self> {
+        let tuple = ob.downcast::<PyTuple>()?;
+        let occur = tuple.get_item(0)?.extract()?;
+        let query = tuple.get_item(1)?.extract()?;
+
+        Ok(OccurQueryPair(occur, query))
+    }
+}
+
+/// Tantivy's Occur
+#[pyclass(frozen, module = "tantivy.tantivy")]
+#[derive(Clone)]
+pub enum Occur {
+    Must,
+    Should,
+    MustNot,
+}
+
+impl From<Occur> for tv::query::Occur {
+    fn from(occur: Occur) -> tv::query::Occur {
+        match occur {
+            Occur::Must => tv::query::Occur::Must,
+            Occur::Should => tv::query::Occur::Should,
+            Occur::MustNot => tv::query::Occur::MustNot,
+        }
+    }
+}
+
 /// Tantivy's Query
 #[pyclass(frozen, module = "tantivy.tantivy")]
 pub(crate) struct Query {
     pub(crate) inner: Box<dyn tv::query::Query>,
 }
 
+impl Clone for Query {
+    fn clone(&self) -> Self {
+        Query {
+            inner: self.inner.box_clone(),
+        }
+    }
+}
+
 impl Query {
     pub(crate) fn get(&self) -> &dyn tv::query::Query {
         &self.inner
@@ -110,4 +153,21 @@ impl Query {
             Err(e) => Err(to_pyerr(e)),
         }
     }
+
+    #[staticmethod]
+    #[pyo3(signature = (subqueries))]
+    pub(crate) fn boolean_query(
+        subqueries: Vec<(Occur, Query)>,
+    ) -> PyResult<Query> {
+        let dyn_subqueries = subqueries
+            .into_iter()
+            .map(|(occur, query)| (occur.into(), query.inner.box_clone()))
+            .collect::<Vec<_>>();
+
+        let inner = tv::query::BooleanQuery::from(dyn_subqueries);
+
+        Ok(Query {
+            inner: Box::new(inner),
+        })
+    }
 }