DatasetReader str paths and Writer.add_reads type checking

CHANGELOG.md

@@ -7,6 +7,16 @@ All notable changes, updates, and fixes to pod5 will be documented here
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.3.11]
+
+## Added
+
+- Typechecking on `Writer.add_reads` to inform users incorrectly passing `ReadRecords`
+
+## Fixed
+
+- `DatasetReader` correctly handles string paths
+
 ## [0.3.10]
 
 ## Added
@@ -100,7 +110,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Removed python3.7 support
 
-
 ## [0.2.9] 2023-11-02
 
 ### Fixed
@@ -119,7 +128,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Remove exposed artifactory URL env var from gitlab ci config.
 - `convert to_fast5` writes byte encoded read_ids to match Minkow (was `str`)
 
-
 ## [0.2.7] 2023-09-11
 
 ### Added

docs/requirements.txt

@@ -2,4 +2,4 @@ sphinx-rtd-theme
 sphinx==v5.3.0
 myst-parser
 # Paths are relative to project root for ReadTheDocs and docs/Makefile
-pod5==0.3.10
+pod5==0.3.11

python/pod5/pyproject.toml

@@ -22,7 +22,7 @@ classifiers=[
 ]
 
 dependencies = [
-    "lib_pod5 == 0.3.10",
+    "lib_pod5 == 0.3.11",
     "iso8601",
     'importlib-metadata; python_version<"3.8"',
     "more_itertools",

python/pod5/src/pod5/dataset.py

@@ -290,13 +290,18 @@ def _collect_dataset(
         threads: int,
     ) -> Set[Path]:
         if isinstance(paths, (str, Path, os.PathLike)):
-            paths = [Path(paths)]
+            paths = [paths]
 
-        _paths = list(paths)
+        if not isinstance(paths, Collection):
+            raise TypeError(
+                f"paths must be a Collection[PathOrStr] but found {type(paths)=}"
+            )
+
+        paths = [Path(p) for p in paths]
         collected: Set[Path] = set()
         with ThreadPoolExecutor(max_workers=threads) as executor:
             search = partial(search_path, recursive=recursive, patterns=[pattern])
-            for coll in executor.map(search, _paths):
+            for coll in executor.map(search, paths):
                 collected.update(coll)
         return collected
 

python/pod5/src/pod5/writer.py

@@ -1,6 +1,7 @@
 """
 Tools for writing POD5 data
 """
+
 import datetime
 import itertools
 from pathlib import Path
@@ -19,6 +20,7 @@
 
 import lib_pod5 as p5b
 import numpy as np
+from pod5.reader import ReadRecord
 import pytz
 
 from pod5.api_utils import Pod5ApiException, safe_close
@@ -296,6 +298,11 @@ def add_reads(self, reads: Sequence[Union[Read, CompressedRead]]) -> None:
                 np.concatenate(signal_chunk_lengths).astype(np.uint32),  # type: ignore [no-untyped-call]
                 signal_chunk_counts,
             )
+        elif isinstance(reads[0], ReadRecord):
+            raise TypeError(
+                "Writer.add_reads(reads) does not take ReadRecords - see ReadRecord.to_read()"
+            )
+        raise TypeError(f"Writer.add_reads(reads) - unexpected type: {type(reads[0])=}")
 
     def _prepare_add_reads_args(self, reads: Sequence[BaseRead]) -> List[Any]:
         """

python/pod5/src/tests/test_dataset.py

@@ -393,3 +393,15 @@ def test_get_path(self, nested_dataset: Path) -> None:
         for path in dataset.paths:
             read_id = dataset.get_reader(path).read_ids[0]
             assert dataset.get_path(read_id) == path
+
+    def test_collect_paths(self, nested_dataset: Path) -> None:
+        """Pass various inputs to DatasetReader._collect_dataset"""
+        collect = p5.DatasetReader._collect_dataset
+
+        expected = {nested_dataset / "root_10.pod5"}
+        kwargs = dict(recursive=False, pattern="*.pod5", threads=1)
+        assert expected == collect(nested_dataset, **kwargs)  # type: ignore
+        assert expected == collect(str(nested_dataset), **kwargs)  # type: ignore
+
+        with pytest.raises(TypeError, match="paths must be a Collection"):
+            collect(1, **kwargs)  # type: ignore

python/pod5/src/tests/test_writer.py

@@ -1,6 +1,7 @@
 """
 Testing Pod5Writer
 """
+
 import math
 import lib_pod5 as p5b
 import numpy as np
@@ -113,3 +114,17 @@ def test_read_copy(self, reader: p5.Reader, writer: p5.Writer) -> None:
             assert before.has_cached_signal == after.has_cached_signal
             assert np.array_equal(before.signal, after.signal)
             assert np.array_equal(before.signal_pa, after.signal_pa)
+
+    def test_read_record_type_check(self, reader: p5.Reader, writer: p5.Writer) -> None:
+        """Check type errors raised when passing ReadRecords to writer"""
+        with pytest.raises(TypeError, match="ReadRecord.to_read"):
+            for record in reader:
+                writer.add_read(record)  # type: ignore
+
+        with pytest.raises(TypeError, match="ReadRecord.to_read"):
+            writer.add_reads([r for r in reader])  # type: ignore
+
+        with pytest.raises(TypeError, match="unexpected type"):
+            writer.add_read([1])  # type: ignore
+
+        writer.close()