Merge pull request #20 from cauliyang/dev

feat: Update Python function signatures and add normalization function

Cargo.toml

@@ -17,7 +17,7 @@ pyo3 = { version = "0.21.0", features = [
     "extension-module",
     "anyhow",
 ] }
-pyo3-stub-gen = "0.5"
+pyo3-stub-gen = "0.5.1"
 thiserror = "1.0"
 anyhow = "1.0"
 walkdir = { version = "2.4" }

crates/deepbiop-fq/src/python.rs

@@ -31,6 +31,7 @@ impl encode::TensorEncoder {
     }
 }
 
+#[gen_stub_pymethods]
 #[pymethods]
 impl encode::JsonEncoder {
     #[new]
@@ -39,6 +40,7 @@ impl encode::JsonEncoder {
     }
 }
 
+#[gen_stub_pymethods]
 #[pymethods]
 impl encode::ParquetEncoder {
     #[new]
@@ -112,6 +114,7 @@ impl PyRecordData {
     }
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn write_fq(records_data: Vec<PyRecordData>, file_path: Option<PathBuf>) -> Result<()> {
     let records: Vec<encode::RecordData> = records_data
@@ -121,6 +124,7 @@ fn write_fq(records_data: Vec<PyRecordData>, file_path: Option<PathBuf>) -> Resu
     io::write_fq(&records, file_path)
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn write_fq_parallel(
     records_data: Vec<PyRecordData>,
@@ -168,6 +172,28 @@ fn generate_kmers(base: String, k: usize) -> Vec<String> {
         .collect()
 }
 
+/// Normalize a DNA sequence by converting any non-standard nucleotides to standard ones.
+///
+/// This function takes a DNA sequence as a `String` and a boolean flag `iupac` indicating whether to normalize using IUPAC ambiguity codes.
+/// It returns a normalized DNA sequence as a `String`.
+///
+/// # Arguments
+///
+/// * `seq` - A DNA sequence as a `String`.
+/// * `iupac` - A boolean flag indicating whether to normalize using IUPAC ambiguity codes.
+///
+/// # Returns
+///
+/// A normalized DNA sequence as a `String`.
+///
+/// # Examples
+///
+/// ```
+/// use deepbiop_fq as fq;
+/// let seq = "acGTN".to_string();
+/// let normalized_seq = fq::normalize_seq(seq, false);
+/// assert_eq!(normalized_seq, "ACGTN");
+/// ```
 #[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn normalize_seq(seq: String, iupac: bool) -> String {
@@ -267,6 +293,7 @@ fn encode_fq_path_to_tensor(
     ))
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn encode_fq_path_to_json(
     fq_path: PathBuf,
@@ -302,6 +329,7 @@ fn encode_fq_path_to_json(
     Ok(())
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn encode_fq_path_to_parquet_chunk(
     fq_path: PathBuf,
@@ -325,6 +353,7 @@ fn encode_fq_path_to_parquet_chunk(
     Ok(())
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn encode_fq_path_to_parquet(
     fq_path: PathBuf,
@@ -358,6 +387,7 @@ fn encode_fq_path_to_parquet(
     Ok(())
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn encode_fq_paths_to_parquet(
     fq_path: Vec<PathBuf>,
@@ -387,6 +417,7 @@ fn get_label_region(labels: Vec<i8>) -> Vec<(usize, usize)> {
         .collect()
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 fn convert_multiple_fqs_to_one_fq(
     paths: Vec<PathBuf>,
@@ -452,6 +483,7 @@ pub fn load_predicts_from_batch_pts(
     predicts::load_predicts_from_batch_pts(pt_path, ignore_label, &id_table, max_predicts)
 }
 
+#[gen_stub_pyfunction(module = "deepbiop.fq")]
 #[pyfunction]
 pub fn fastq_to_fasta(fastq_path: PathBuf, fasta_path: PathBuf) -> Result<()> {
     let fa_records = io::fastq_to_fasta(&fastq_path)?;

py-deepbiop/deepbiop/fq.pyi

@@ -1,6 +1,8 @@
 # This file is automatically generated by pyo3_stub_gen
 # ruff: noqa: E501, F401
 
+import os
+import pathlib
 import typing
 
 class FqEncoderOption:
@@ -18,8 +20,11 @@ class FqEncoderOption:
         threads: int | None,
     ): ...
 
-class JsonEncoder: ...
-class ParquetEncoder: ...
+class JsonEncoder:
+    def __new__(cls, option: FqEncoderOption): ...
+
+class ParquetEncoder:
+    def __new__(cls, option: FqEncoderOption): ...
 
 class Predict:
     r"""A struct to store the prediction result."""
@@ -78,12 +83,84 @@ class RecordData:
     def set_seq(self, seq: str) -> None: ...
     def set_qual(self, qual: str) -> None: ...
 
+def convert_multiple_fqs_to_one_fq(
+    paths: typing.Sequence[str | os.PathLike | pathlib.Path],
+    result_path: str | os.PathLike | pathlib.Path,
+    parallel: bool,
+) -> None: ...
+def encode_fq_path_to_json(
+    fq_path: str | os.PathLike | pathlib.Path,
+    k: int,
+    bases: str,
+    qual_offset: int,
+    vectorized_target: bool,
+    result_path: str | os.PathLike | pathlib.Path | None,
+) -> None: ...
+def encode_fq_path_to_parquet(
+    fq_path: str | os.PathLike | pathlib.Path,
+    bases: str,
+    qual_offset: int,
+    vectorized_target: bool,
+    result_path: str | os.PathLike | pathlib.Path | None,
+) -> None: ...
+def encode_fq_path_to_parquet_chunk(
+    fq_path: str | os.PathLike | pathlib.Path,
+    chunk_size: int,
+    parallel: bool,
+    bases: str,
+    qual_offset: int,
+    vectorized_target: bool,
+) -> None: ...
+def encode_fq_paths_to_parquet(
+    fq_path: typing.Sequence[str | os.PathLike | pathlib.Path],
+    bases: str,
+    qual_offset: int,
+    vectorized_target: bool,
+) -> None: ...
 def encode_qual(qual: str, qual_offset: int) -> list[int]:
     r"""Convert ASCII quality to Phred score for Phred+33 encoding."""
 
+def fastq_to_fasta(
+    fastq_path: str | os.PathLike | pathlib.Path,
+    fasta_path: str | os.PathLike | pathlib.Path,
+) -> None: ...
 def generate_kmers(base: str, k: int) -> list[str]: ...
 def get_label_region(labels: typing.Sequence[int]) -> list[tuple[int, int]]: ...
 def kmers_to_seq(kmers: typing.Sequence[str]) -> str: ...
-def normalize_seq(seq: str, iupac: bool) -> str: ...
+def normalize_seq(seq: str, iupac: bool) -> str:
+    r"""
+    Normalize a DNA sequence by converting any non-standard nucleotides to standard ones.
+
+    This function takes a DNA sequence as a `String` and a boolean flag `iupac` indicating whether to normalize using IUPAC ambiguity codes.
+    It returns a normalized DNA sequence as a `String`.
+
+    # Arguments
+
+    * `seq` - A DNA sequence as a `String`.
+    * `iupac` - A boolean flag indicating whether to normalize using IUPAC ambiguity codes.
+
+    # Returns
+
+    A normalized DNA sequence as a `String`.
+
+    # Examples
+
+    ```
+    use deepbiop_fq as fq;
+    let seq = "acGTN".to_string();
+    let normalized_seq = fq::normalize_seq(seq, false);
+    assert_eq!(normalized_seq, "ACGTN");
+    ```
+    """
+
 def seq_to_kmers(seq: str, k: int, overlap: bool) -> list[str]: ...
 def test_predicts(predicts: typing.Sequence[Predict]) -> None: ...
+def write_fq(
+    records_data: typing.Sequence[RecordData],
+    file_path: str | os.PathLike | pathlib.Path | None,
+) -> None: ...
+def write_fq_parallel(
+    records_data: typing.Sequence[RecordData],
+    file_path: str | os.PathLike | pathlib.Path,
+    threads: int,
+) -> None: ...

py-deepbiop/docs/conf.py

@@ -1,10 +1,6 @@
 """Sphinx configuration."""
 
-import sys
 from datetime import datetime
-from pathlib import Path
-
-# sys.path.insert(0, (Path().resolve() / "../deepbiop").as_posix())
 
 project = "deepbiop"
 author = "Yangyang Li"