Polish and test coverage

.github/workflows/check.yml

@@ -0,0 +1,36 @@
+name: Check
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  check:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install Poetry
+        run: pipx install poetry
+
+      - name: Install dependencies
+        run: poetry install --with dev --extras all
+
+      - name: Lint
+        run: poetry run flake8 .
+
+      - name: Typecheck
+        run: poetry run mypy .
+
+      - name: Format
+        run: poetry run black .
+
+      - name: Test
+        run: poetry run pytest -n auto

avrokit/__init__.py

@@ -77,8 +77,9 @@
     ...             writer.roll()  # Create a new partition file
 """
 
-from .url import URL, parse_url, create_url_mapping
+from .url import URL, FileURL, parse_url, create_url_mapping, flatten_urls
 from .io import (
+    Appendable,
     PartitionedAvroReader,
     PartitionedAvroWriter,
     TimePartitionedAvroWriter,
@@ -88,16 +89,46 @@
     avro_writer,
     avro_records,
     compact_avro_data,
+    read_avro_schema,
+    read_avro_schema_from_first_nonempty_file,
     validate_avro_schema_evolution,
 )
 from .asyncio import DeferredAvroWriter, BlockingQueueAvroReader
+from .tools import (
+    CatTool,
+    ConcatTool,
+    FileSortTool,
+    FromParquetTool,
+    GetMetaTool,
+    GetSchemaTool,
+    HttpServerTool,
+    PartitionTool,
+    RepairTool,
+    StatsTool,
+    ToJsonTool,
+    ToParquetTool,
+)
 
 __all__ = [
+    "Appendable",
     "BlockingQueueAvroReader",
+    "CatTool",
+    "ConcatTool",
     "DeferredAvroWriter",
+    "FileSortTool",
+    "FileURL",
+    "FromParquetTool",
+    "GetMetaTool",
+    "GetSchemaTool",
+    "HttpServerTool",
+    "PartitionTool",
     "PartitionedAvroReader",
     "PartitionedAvroWriter",
+    "RepairTool",
+    "StatsTool",
     "TimePartitionedAvroWriter",
+    "ToJsonTool",
+    "ToParquetTool",
     "URL",
     "add_avro_schema_fields",
     "avro_reader",
@@ -106,6 +137,9 @@
     "avro_records",
     "compact_avro_data",
     "create_url_mapping",
+    "flatten_urls",
     "parse_url",
+    "read_avro_schema",
+    "read_avro_schema_from_first_nonempty_file",
     "validate_avro_schema_evolution",
 ]

pyproject.toml

@@ -4,7 +4,7 @@
 
 [tool.poetry]
 name = "avrokit"
-version = "0.0.1"
+version = "0.0.2"
 description = "Python utilities for working with Avro data files"
 authors = ["Greg Brandt <brandt.greg@gmail.com>"]
 license = "Apache-2.0"

tests/test_additional_coverage.py

@@ -20,7 +20,7 @@ def test_file_url_nonexistent_read(self):
         """Test reading from non-existent file."""
         url = parse_url("/tmp/nonexistent_file_12345.avro")
         with pytest.raises(FileNotFoundError):
-            with url.with_mode("rb") as f:
+            with url.with_mode("rb") as _:
                 pass
 
     def test_file_url_exists_check(self):