quickner - pypi Package Compare versions

Comparing version 0.0.1a15 to 0.0.1a16

Cargo.lock

@@ -28,5 +28,5 @@ # This file is automatically @generated by Cargo.
 name = "block-buffer"
-version = "0.10.3"
+version = "0.10.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "69cce20737498f97b993470a6e536b8523f0af7892a4f928cceb1ac5e52ebe7e"
+checksum = "3078c7629b62d3f0439517fa394996acacc5cbc91c5a20d8c658e77abd503a71"
 dependencies = [
@@ -162,5 +162,5 @@ "generic-array",
 name = "csv"
-version = "1.2.0"
+version = "1.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "af91f40b7355f82b0a891f50e70399475945bb0b0da4f1700ce60761c9d3e359"
+checksum = "0b015497079b9a9d69c02ad25de6c0a6edef051ea6360a327d0bd05802ef64ad"
 dependencies = [
@@ -311,5 +311,5 @@ "csv-core",
 name = "io-lifetimes"
-version = "1.0.5"
+version = "1.0.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1abeb7a0dd0f8181267ff8adc397075586500b81b28a73e8a0208b00fc170fb3"
+checksum = "cfa919a82ea574332e2de6e74b4c36e74d41982b335080fa59d4ef31be20fdf3"
 dependencies = [
@@ -346,5 +346,5 @@ "libc",
 name = "libc"
-version = "0.2.139"
+version = "0.2.140"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "201de327520df007757c1f0adce6e827fe8562fbc28bfd9c15571c66ca1f5f79"
+checksum = "99227334921fae1a979cf0bfdfcc6b3e5ce376ef57e16fb6fb3ea2ed6095f80c"
 
@@ -377,2 +377,11 @@ [[package]]
 [[package]]
+name = "matrixmultiply"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "add85d4dd35074e6fedc608f8c8f513a3548619a9024b751949ef0e8e45a4d84"
+dependencies = [
+ "rawpointer",
+]
+
+[[package]]
 name = "memchr"
@@ -393,2 +402,43 @@ version = "2.5.0"
 [[package]]
+name = "ndarray"
+version = "0.15.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "adb12d4e967ec485a5f71c6311fe28158e9d6f4bc4a447b474184d0f91a8fa32"
+dependencies = [
+ "matrixmultiply",
+ "num-complex",
+ "num-integer",
+ "num-traits",
+ "rawpointer",
+]
+
+[[package]]
+name = "num-complex"
+version = "0.4.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "02e0d21255c828d6f128a1e41534206671e8c3ea0c62f32291e808dc82cff17d"
+dependencies = [
+ "num-traits",
+]
+
+[[package]]
+name = "num-integer"
+version = "0.1.45"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "225d3389fb3509a24c93f5c29eb6bde2586b98d9f016636dff58d7c6f7569cd9"
+dependencies = [
+ "autocfg",
+ "num-traits",
+]
+
+[[package]]
+name = "num-traits"
+version = "0.2.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "578ede34cf02f8924ab9447f50c28075b4d3e5b269972345e7e0372b38c6cdcd"
+dependencies = [
+ "autocfg",
+]
+
+[[package]]
 name = "num_cpus"
@@ -410,2 +460,17 @@ version = "1.15.0"
 [[package]]
+name = "numpy"
+version = "0.18.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "96b0fee4571867d318651c24f4a570c3f18408cf95f16ccb576b3ce85496a46e"
+dependencies = [
+ "libc",
+ "ndarray",
+ "num-complex",
+ "num-integer",
+ "num-traits",
+ "pyo3",
+ "rustc-hash",
+]
+
+[[package]]
 name = "once_cell"
@@ -546,4 +611,5 @@ version = "1.17.1"
 name = "quickner"
-version = "0.0.1-alpha.15"
+version = "0.0.1-alpha.16"
 dependencies = [
+ "numpy",
  "pyo3",
@@ -556,3 +622,3 @@ "quickner-core",
 name = "quickner-core"
-version = "0.0.1-alpha.15"
+version = "0.0.1-alpha.16"
 dependencies = [
@@ -582,2 +648,8 @@ "aho-corasick",
 [[package]]
+name = "rawpointer"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "60a357793950651c4ed0f3f52338f53b2f809f32d83a07f72909fa13e4c6c1e3"
+
+[[package]]
 name = "rayon"
@@ -631,2 +703,8 @@ version = "1.7.0"
 [[package]]
+name = "rustc-hash"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
+
+[[package]]
 name = "rustix"
@@ -659,5 +737,5 @@ version = "0.36.9"
 name = "serde"
-version = "1.0.152"
+version = "1.0.154"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bb7d1f0d3021d347a83e556fc4683dea2ea09d87bccdf88ff5c12545d89d5efb"
+checksum = "8cdd151213925e7f1ab45a9bbfb129316bd00799784b174b7cc7bcd16961c49e"
 dependencies = [
@@ -669,5 +747,5 @@ "serde_derive",
 name = "serde_derive"
-version = "1.0.152"
+version = "1.0.154"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "af487d118eecd09402d70a5d72551860e788df87b464af30e5ea6a38c75c541e"
+checksum = "4fc80d722935453bcafdc2c9a73cd6fac4dc1938f0346035d84bf99fa9e33217"
 dependencies = [
@@ -681,5 +759,5 @@ "proc-macro2",
 name = "serde_json"
-version = "1.0.93"
+version = "1.0.94"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cad406b69c91885b5107daf2c29572f6c8cdb3c66826821e286c533490c0bc76"
+checksum = "1c533a59c9d8a93a09c6ab31f0fd5e5f4dd1b8fc9434804029839884765d04ea"
 dependencies = [
@@ -772,5 +850,5 @@ "itoa",
 name = "toml_edit"
-version = "0.19.4"
+version = "0.19.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9a1eb0622d28f4b9c90adc4ea4b2b46b47663fde9ac5fafcb14a1369d5508825"
+checksum = "7082a95d48029677a28f181e5f6422d0c8339ad8396a39d3f33d62a90c1f6c30"
 dependencies = [
@@ -792,5 +870,5 @@ "indexmap",
 name = "unicode-ident"
-version = "1.0.7"
+version = "1.0.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "775c11906edafc97bc378816b94585fbd9a054eabaf86fdd0ced94af449efab7"
+checksum = "e5464a87b239f13a63a501f2701565754bae92d243d4bb7eb12f6d57d2269bf4"
 
@@ -929,7 +1007,7 @@ [[package]]
 name = "winnow"
-version = "0.3.4"
+version = "0.3.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c95fb4ff192527911dd18eb138ac30908e7165b8944e528b6af93aa4c842d345"
+checksum = "ee7b2c67f962bf5042bfd8b6a916178df33a26eec343ae064cb8e069f638fa6f"
 dependencies = [
  "memchr",
 ]

Cargo.toml

 [package]
 name = "quickner"
-version = "0.0.1-alpha.15"
+version = "0.0.1-alpha.16"
 edition = "2021"
@@ -24,1 +24,2 @@ authors = ["Omar MHAIMDAT"]
 serde = { version = "1.0.130", features = ["derive"] }
+numpy = "0.18"

local_dependencies/quickner-core/Cargo.toml

 [package]
 name = "quickner-core"
-version = "0.0.1-alpha.15"
+version = "0.0.1-alpha.16"
 edition = "2021"
@@ -5,0 +5,0 @@ authors = ["Omar MHAIMDAT"]

PKG-INFO

 Metadata-Version: 2.1
 Name: quickner
-Version: 0.0.1a15
+Version: 0.0.1a16
 Classifier: Programming Language :: Rust
@@ -5,0 +5,0 @@ Classifier: Programming Language :: Python :: Implementation :: CPython

pyproject.toml

@@ -14,3 +14,2 @@ [build-system]
 
-
 [project.scripts]
@@ -17,0 +16,0 @@ Quickner = "quickner:Quickner"

quickner.pyi

@@ -312,14 +312,2 @@ from ast import Dict
 
-    def add_string(self, string: str) -> None:
-        """
-        Add a string to the list of documents.
-
-        Parameters:
-            string (str): String to add.
-
-        Returns:
-            None
-        """
-        ...
-
     def add_entity(self, entity: Entity) -> None:
@@ -362,1 +350,10 @@ """
         ...
+
+    def numpy(self) -> NDArray:
+        """
+        Convert the list of documents to a Numpy array.
+
+        Returns:
+            NDArray: Numpy array of documents.
+        """
+        ...

src/lib.rs

 use pyo3::prelude::*;
+use pyquickner::QuicknerError;
 mod pyconfig;
@@ -35,3 +36,4 @@ mod pydocument;
     m.add_class::<pyentity::PyEntity>()?;
+    m.add("QuicknerError", _py.get_type::<QuicknerError>())?;
     Ok(())
 }

src/pydocument.rs

@@ -102,3 +102,6 @@ use crate::{
     pub fn __repr__(&self) -> PyResult<String> {
-        let mut repr = format!("Document(id=\"{}\", text={}, label=[", self.id, self.text);
+        let mut repr = format!(
+            "Document(id=\"{}\", text=\"{}\", label=[",
+            self.id, self.text
+        );
         for (start, end, label) in &self.label {
@@ -105,0 +108,0 @@ repr.push_str(&format!("({start}, {end}, {label}), "));

src/pyquickner.rs

@@ -12,8 +12,13 @@ use std::collections::HashMap;
 };
+use numpy::PyArray2;
+use pyo3::create_exception;
 use pyo3::{
     exceptions::{self, PyGeneratorExit},
     prelude::*,
+    types::{PyDict, PyTuple},
 };
 use quickner::{Document, Entity, Quickner, SpacyEntity};
 
+create_exception!(quickner, QuicknerError, exceptions::PyException);
+
 #[pyclass(name = "Quickner")]
@@ -135,8 +140,2 @@ pub struct PyQuickner {
 
-    pub fn add_string(&mut self, text: &str) {
-        let document = Document::from_string(text.to_string());
-        self.quickner.add_document(document.clone());
-        self.add_document(PyDocument::from(document));
-    }
-
     pub fn add_entity(&mut self, entity: PyEntity) {
@@ -365,2 +364,70 @@ // Check if the entity is already in the list
     }
+
+    /// Convert Vec<Document> to numpy array of (string, array of (int, int, string))
+    /// where the first int is the start index and the second int is the end index
+    /// of the entity in the string.
+    /// The string is the text of the document.
+    /// The array of (int, int, string) is the list of entities in the document.
+    /// Return a numpy array like so: array(['rust is made by Mozilla', list([(0, 4, 'PL'), (16, 23, 'ORG')])], dtype=object)
+    /// And type is numpy.ndarray
+    // pub fn numpy(&self) -> Py<PyArray1<PyObject>> {
+    //     Python::with_gil(|py| {
+    //         let numpy = PyModule::import(py, "numpy").unwrap();
+    //         let array = numpy.getattr("array").unwrap();
+    //         let array = array.call1((self.documents.clone(),)).unwrap();
+    //         array.extract().unwrap()
+    //     })
+    // }
+
+    pub fn numpy(&self) -> PyResult<Py<PyArray2<PyObject>>> {
+        Python::with_gil(|py| {
+            let numpy = PyModule::import(py, "numpy").unwrap();
+            let array = numpy.getattr("array").unwrap();
+            let object: Vec<&PyTuple> = self
+                .documents
+                .iter()
+                .map(|document| {
+                    let entities: Vec<&PyTuple> = document
+                        .label
+                        .iter()
+                        .map(|entity| {
+                            PyTuple::new(
+                                py,
+                                &[
+                                    entity.0.to_object(py),
+                                    entity.1.to_object(py),
+                                    entity.2.clone().to_object(py),
+                                ],
+                            )
+                        })
+                        .collect();
+                    PyTuple::new(
+                        py,
+                        &[
+                            document.id.clone().to_object(py),
+                            document.text.clone().to_object(py),
+                            entities.to_object(py),
+                        ],
+                    )
+                })
+                .collect::<Vec<&PyTuple>>();
+            let args = PyDict::new(py);
+            args.set_item("dtype", "object").unwrap();
+            let array = array.call((object,), Some(args));
+            if let Ok(array) = array {
+                let array = array.extract::<Py<PyArray2<PyObject>>>();
+                match array {
+                    Ok(array) => Ok(array),
+                    Err(array) => {
+                        // Raise an exception QuicknerError
+                        Err(PyErr::new::<QuicknerError, _>(array.to_string()))
+                    }
+                }
+            } else {
+                let array = array.unwrap_err();
+                // Raise an exception QuicknerError
+                Err(PyErr::new::<QuicknerError, _>(array.to_string()))
+            }
+        })
+    }
 }
@@ -367,0 +434,0 @@

Alert delta unavailable

Currently unable to show alert delta for PyPI packages.

Improved metrics

Total package byte prevSize

39177056

0.01%

Worsened metrics

Lines of code

286

-0.69%