Merge pull request #45 from saukymo/master

Bump tantivy version 0.17.0

Author: ChillFish8

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "tantivy"
-version = "0.16.0"
+version = "0.17.0"
 readme = "README.md"
 authors = ["Damir Jelić <[email protected]>"]
 edition = "2018"
@@ -11,14 +11,15 @@ name = "tantivy"
 crate-type = ["cdylib"]
 
 [build-dependencies]
-pyo3-build-config = "0.15.1"
+pyo3-build-config = "0.16.3"
 
 [dependencies]
 chrono = "0.4.19"
-tantivy = "0.16.1"
-itertools = "0.10.0"
-futures = "0.3.5"
+tantivy = "0.17"
+itertools = "0.10.3"
+futures = "0.3.21"
+serde_json = "1.0.64"
 
 [dependencies.pyo3]
-version = "0.15.1"
+version = "0.16.3"
 features = ["extension-module"]

src/document.rs

@@ -14,10 +14,37 @@ use chrono::{offset::TimeZone, Datelike, Timelike, Utc};
 use tantivy as tv;
 
 use crate::{facet::Facet, to_pyerr};
-use pyo3::{PyMappingProtocol, PyObjectProtocol};
-use std::{collections::BTreeMap, fmt};
+use serde_json::Value as JsonValue;
+use std::{
+    collections::{BTreeMap, HashMap},
+    fmt,
+};
 use tantivy::schema::Value;
 
+fn value_to_object(val: &JsonValue, py: Python<'_>) -> PyObject {
+    match val {
+        JsonValue::Null => py.None(),
+        JsonValue::Bool(b) => b.to_object(py),
+        JsonValue::Number(n) => match n {
+            n if n.is_i64() => n.as_i64().to_object(py),
+            n if n.is_u64() => n.as_u64().to_object(py),
+            n if n.is_f64() => n.as_f64().to_object(py),
+            _ => panic!("number too large"),
+        },
+        JsonValue::String(s) => s.to_object(py),
+        JsonValue::Array(v) => {
+            let inner: Vec<_> =
+                v.iter().map(|x| value_to_object(x, py)).collect();
+            inner.to_object(py)
+        }
+        JsonValue::Object(m) => {
+            let inner: HashMap<_, _> =
+                m.iter().map(|(k, v)| (k, value_to_object(v, py))).collect();
+            inner.to_object(py)
+        }
+    }
+}
+
 fn value_to_py(py: Python, value: &Value) -> PyResult<PyObject> {
     Ok(match value {
         Value::Str(text) => text.into_py(py),
@@ -42,6 +69,13 @@ fn value_to_py(py: Python, value: &Value) -> PyResult<PyObject> {
         )?
         .into_py(py),
         Value::Facet(f) => Facet { inner: f.clone() }.into_py(py),
+        Value::JsonObject(json_object) => {
+            let inner: HashMap<_, _> = json_object
+                .iter()
+                .map(|(k, v)| (k, value_to_object(&v, py)))
+                .collect();
+            inner.to_object(py)
+        }
     })
 }
 
@@ -58,6 +92,9 @@ fn value_to_string(value: &Value) -> String {
             // TODO implement me
             unimplemented!();
         }
+        Value::JsonObject(json_object) => {
+            serde_json::to_string(&json_object).unwrap()
+        }
     }
 }
 
@@ -293,6 +330,17 @@ impl Document {
         add_value(self, field_name, bytes);
     }
 
+    /// Add a bytes value to the document.
+    ///
+    /// Args:
+    ///     field_name (str): The field for which we are adding the bytes.
+    ///     value (str): The json object that will be added to the document.
+    fn add_json(&mut self, field_name: String, json: &str) {
+        let json_object: serde_json::Value =
+            serde_json::from_str(json).unwrap();
+        add_value(self, field_name, json_object);
+    }
+
     /// Returns the number of added fields that have been added to the document
     #[getter]
     fn num_fields(&self) -> usize {
@@ -337,6 +385,16 @@ impl Document {
             .map(|value| value_to_py(py, value))
             .collect::<PyResult<Vec<_>>>()
     }
+
+    fn __getitem__(&self, field_name: &str) -> PyResult<Vec<PyObject>> {
+        let gil = Python::acquire_gil();
+        let py = gil.python();
+        self.get_all(py, field_name)
+    }
+
+    fn __repr__(&self) -> PyResult<String> {
+        Ok(format!("{:?}", self))
+    }
 }
 
 impl Document {
@@ -350,19 +408,3 @@ impl Document {
             .flat_map(|values| values.iter())
     }
 }
-
-#[pyproto]
-impl PyMappingProtocol for Document {
-    fn __getitem__(&self, field_name: &str) -> PyResult<Vec<PyObject>> {
-        let gil = Python::acquire_gil();
-        let py = gil.python();
-        self.get_all(py, field_name)
-    }
-}
-
-#[pyproto]
-impl PyObjectProtocol for Document {
-    fn __repr__(&self) -> PyResult<String> {
-        Ok(format!("{:?}", self))
-    }
-}

src/facet.rs

@@ -1,4 +1,4 @@
-use pyo3::{basic::PyObjectProtocol, prelude::*, types::PyType};
+use pyo3::{prelude::*, types::PyType};
 use tantivy::schema;
 
 /// A Facet represent a point in a given hierarchy.
@@ -63,10 +63,7 @@ impl Facet {
     fn to_path_str(&self) -> String {
         self.inner.to_string()
     }
-}
 
-#[pyproto]
-impl PyObjectProtocol for Facet {
     fn __repr__(&self) -> PyResult<String> {
         Ok(format!("Facet({})", self.to_path_str()))
     }

src/index.rs

@@ -41,7 +41,7 @@ impl IndexWriter {
     pub fn add_document(&mut self, doc: &Document) -> PyResult<u64> {
         let named_doc = NamedFieldDocument(doc.field_values.clone());
         let doc = self.schema.convert_named_doc(named_doc).map_err(to_pyerr)?;
-        Ok(self.inner_index_writer.add_document(doc))
+        self.inner_index_writer.add_document(doc).map_err(to_pyerr)
     }
 
     /// Helper for the `add_document` method, but passing a json string.
@@ -55,7 +55,7 @@ impl IndexWriter {
     pub fn add_json(&mut self, json: &str) -> PyResult<u64> {
         let doc = self.schema.parse_document(json).map_err(to_pyerr)?;
         let opstamp = self.inner_index_writer.add_document(doc);
-        Ok(opstamp)
+        opstamp.map_err(to_pyerr)
     }
 
     /// Commits all of the pending changes
@@ -134,6 +134,12 @@ impl IndexWriter {
                     field_name
                 )))
             }
+            Value::JsonObject(_) => {
+                return Err(exceptions::PyValueError::new_err(format!(
+                    "Field `{}` is json object type not deletable.",
+                    field_name
+                )))
+            }
         };
         Ok(self.inner_index_writer.delete_term(term))
     }
@@ -281,7 +287,7 @@ impl Index {
     #[staticmethod]
     fn exists(path: &str) -> PyResult<bool> {
         let directory = MmapDirectory::open(path).map_err(to_pyerr)?;
-        Ok(tv::Index::exists(&directory).unwrap())
+        tv::Index::exists(&directory).map_err(to_pyerr)
     }
 
     /// The schema of the current index.
@@ -304,7 +310,7 @@ impl Index {
     ///
     /// Args:
     ///     query: the query, following the tantivy query language.
-    ///     default_fields (List[Field]): A list of fields used to search if no
+    ///     default_fields_names (List[Field]): A list of fields used to search if no
     ///         field is specified in the query.
     ///
     #[args(reload_policy = "RELOAD_POLICY")]

src/query.rs

@@ -1,4 +1,4 @@
-use pyo3::{prelude::*, PyObjectProtocol};
+use pyo3::prelude::*;
 use tantivy as tv;
 
 /// Tantivy's Query
@@ -13,8 +13,8 @@ impl Query {
     }
 }
 
-#[pyproto]
-impl PyObjectProtocol for Query {
+#[pymethods]
+impl Query {
     fn __repr__(&self) -> PyResult<String> {
         Ok(format!("Query({:?})", self.get()))
     }

src/schemabuilder.rs

@@ -73,26 +73,11 @@ impl SchemaBuilder {
         index_option: &str,
     ) -> PyResult<Self> {
         let builder = &mut self.builder;
-        let index_option = match index_option {
-            "position" => schema::IndexRecordOption::WithFreqsAndPositions,
-            "freq" => schema::IndexRecordOption::WithFreqs,
-            "basic" => schema::IndexRecordOption::Basic,
-            _ => return Err(exceptions::PyValueError::new_err(
-                "Invalid index option, valid choices are: 'basic', 'freq' and 'position'"
-            ))
-        };
-
-        let indexing = schema::TextFieldIndexing::default()
-            .set_tokenizer(tokenizer_name)
-            .set_index_option(index_option);
-
-        let options =
-            schema::TextOptions::default().set_indexing_options(indexing);
-        let options = if stored {
-            options.set_stored()
-        } else {
-            options
-        };
+        let options = SchemaBuilder::build_text_option(
+            stored,
+            tokenizer_name,
+            index_option,
+        )?;
 
         if let Some(builder) = builder.write().unwrap().as_mut() {
             builder.add_text_field(name, options);
@@ -230,6 +215,55 @@ impl SchemaBuilder {
         Ok(self.clone())
     }
 
+    /// Add a new json field to the schema.
+    ///
+    /// Args:
+    ///     name (str): the name of the field.
+    ///     stored (bool, optional): If true sets the field as stored, the
+    ///         content of the field can be later restored from a Searcher.
+    ///         Defaults to False.
+    ///     tokenizer_name (str, optional): The name of the tokenizer that
+    ///         should be used to process the field. Defaults to 'default'
+    ///     index_option (str, optional): Sets which information should be
+    ///         indexed with the tokens. Can be one of 'position', 'freq' or
+    ///         'basic'. Defaults to 'position'. The 'basic' index_option
+    ///         records only the document ID, the 'freq' option records the
+    ///         document id and the term frequency, while the 'position' option
+    ///         records the document id, term frequency and the positions of
+    ///         the term occurrences in the document.
+    ///
+    /// Returns the associated field handle.
+    /// Raises a ValueError if there was an error with the field creation.
+    #[args(
+        stored = false,
+        tokenizer_name = "TOKENIZER",
+        index_option = "RECORD"
+    )]
+    fn add_json_field(
+        &mut self,
+        name: &str,
+        stored: bool,
+        tokenizer_name: &str,
+        index_option: &str,
+    ) -> PyResult<Self> {
+        let builder = &mut self.builder;
+        let options = SchemaBuilder::build_text_option(
+            stored,
+            tokenizer_name,
+            index_option,
+        )?;
+
+        if let Some(builder) = builder.write().unwrap().as_mut() {
+            builder.add_json_field(name, options);
+        } else {
+            return Err(exceptions::PyValueError::new_err(
+                "Schema builder object isn't valid anymore.",
+            ));
+        }
+
+        Ok(self.clone())
+    }
+
     /// Add a Facet field to the schema.
     /// Args:
     ///     name (str): The name of the field.
@@ -289,8 +323,8 @@ impl SchemaBuilder {
         stored: bool,
         indexed: bool,
         fast: Option<&str>,
-    ) -> PyResult<schema::IntOptions> {
-        let opts = schema::IntOptions::default();
+    ) -> PyResult<schema::NumericOptions> {
+        let opts = schema::NumericOptions::default();
 
         let opts = if stored { opts.set_stored() } else { opts };
         let opts = if indexed { opts.set_indexed() } else { opts };
@@ -317,4 +351,33 @@ impl SchemaBuilder {
 
         Ok(opts)
     }
+
+    fn build_text_option(
+        stored: bool,
+        tokenizer_name: &str,
+        index_option: &str,
+    ) -> PyResult<schema::TextOptions> {
+        let index_option = match index_option {
+            "position" => schema::IndexRecordOption::WithFreqsAndPositions,
+            "freq" => schema::IndexRecordOption::WithFreqs,
+            "basic" => schema::IndexRecordOption::Basic,
+            _ => return Err(exceptions::PyValueError::new_err(
+                "Invalid index option, valid choices are: 'basic', 'freq' and 'position'"
+            ))
+        };
+
+        let indexing = schema::TextFieldIndexing::default()
+            .set_tokenizer(tokenizer_name)
+            .set_index_option(index_option);
+
+        let options =
+            schema::TextOptions::default().set_indexing_options(indexing);
+        let options = if stored {
+            options.set_stored()
+        } else {
+            options
+        };
+
+        Ok(options)
+    }
 }