etlplus 0.11.11__py3-none-any.whl → 0.12.1__py3-none-any.whl

etlplus/file/feather.py

@@ -1,48 +1,119 @@
 """
 :mod:`etlplus.file.feather` module.
 
-Stub helpers for FEATHER read/write.
+Helpers for reading/writing Feather files.
 """
 
 from __future__ import annotations
 
 from pathlib import Path
+from typing import Any
+from typing import cast
 
 from ..types import JSONData
+from ..types import JSONDict
+from ..types import JSONList
 
 # SECTION: EXPORTS ========================================================== #
 
 
-def read(path: Path) -> JSONData:
+__all__ = [
+    'read',
+    'write',
+]
+
+
+# SECTION: INTERNAL CONSTANTS =============================================== #
+
+
+_PANDAS_CACHE: dict[str, Any] = {}
+
+
+# SECTION: INTERNAL FUNCTIONS =============================================== #
+
+
+def _get_pandas() -> Any:
+    """
+    Return the pandas module, importing it on first use.
+
+    Raises an informative ImportError if the optional dependency is missing.
+    """
+    mod = _PANDAS_CACHE.get('mod')
+    if mod is not None:  # pragma: no cover - tiny branch
+        return mod
+    try:
+        _pd = __import__('pandas')  # type: ignore[assignment]
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'Feather support requires optional dependency "pandas".\n'
+            'Install with: pip install pandas',
+        ) from e
+    _PANDAS_CACHE['mod'] = _pd
+
+    return _pd
+
+
+def _normalize_records(data: JSONData) -> JSONList:
     """
-    Read FEATHER content from ``path``.
+    Normalize JSON payloads into a list of dictionaries.
+
+    Raises TypeError when payloads contain non-dict items.
+    """
+    if isinstance(data, list):
+        if not all(isinstance(item, dict) for item in data):
+            raise TypeError(
+                'Feather payloads must contain only objects (dicts)',
+            )
+        return cast(JSONList, data)
+    return [cast(JSONDict, data)]
+
+
+# SECTION: FUNCTIONS ======================================================== #
+
+
+def read(
+    path: Path,
+) -> JSONList:
+    """
+    Read Feather content from ``path``.
 
     Parameters
     ----------
     path : Path
-        Path to the FEATHER file on disk.
+        Path to the Feather file on disk.
 
     Returns
     -------
-    JSONData
-        Parsed payload.
+    JSONList
+        The list of dictionaries read from the Feather file.
 
     Raises
     ------
-    NotImplementedError
-        FEATHER :func:`read` is not implemented yet.
+    ImportError
+        When optional dependency "pyarrow" is missing.
     """
-    raise NotImplementedError('FEATHER read is not implemented yet')
+    pandas = _get_pandas()
+    try:
+        frame = pandas.read_feather(path)
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'Feather support requires optional dependency "pyarrow".\n'
+            'Install with: pip install pyarrow',
+        ) from e
+    return cast(JSONList, frame.to_dict(orient='records'))
 
 
-def write(path: Path, data: JSONData) -> int:
+def write(
+    path: Path,
+    data: JSONData,
+) -> int:
     """
-    Write ``data`` to FEATHER at ``path``.
+    Write ``data`` to Feather at ``path`` and return record count.
 
     Parameters
     ----------
     path : Path
-        Path to the FEATHER file on disk.
+        Path to the Feather file on disk.
     data : JSONData
         Data to write.
 
@@ -53,7 +124,21 @@ def write(path: Path, data: JSONData) -> int:
 
     Raises
     ------
-    NotImplementedError
-        FEATHER :func:`write` is not implemented yet.
+    ImportError
+        When optional dependency "pyarrow" is missing.
     """
-    raise NotImplementedError('FEATHER write is not implemented yet')
+    records = _normalize_records(data)
+    if not records:
+        return 0
+
+    pandas = _get_pandas()
+    path.parent.mkdir(parents=True, exist_ok=True)
+    frame = pandas.DataFrame.from_records(records)
+    try:
+        frame.to_feather(path)
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'Feather support requires optional dependency "pyarrow".\n'
+            'Install with: pip install pyarrow',
+        ) from e
+    return len(records)

etlplus/file/gz.py

@@ -1,21 +1,37 @@
 """
 :mod:`etlplus.file.gz` module.
 
-Stub helpers for GZ read/write.
+Helpers for reading/writing GZ files.
 """
 
 from __future__ import annotations
 
+import gzip
+import tempfile
 from pathlib import Path
 
 from ..types import JSONData
+from .enums import CompressionFormat
+from .enums import FileFormat
+from .enums import infer_file_format_and_compression
 
 # SECTION: EXPORTS ========================================================== #
 
 
-def read(path: Path) -> JSONData:
+__all__ = [
+    'read',
+    'write',
+]
+
+
+# SECTION: INTERNAL FUNCTIONS =============================================== #
+
+
+def _resolve_format(
+    path: Path,
+) -> FileFormat:
     """
-    Read GZ content from ``path``.
+    Resolve the inner file format from a .gz filename.
 
     Parameters
     ----------
@@ -24,20 +40,61 @@ def read(path: Path) -> JSONData:
 
     Returns
     -------
-    JSONData
-        Parsed payload.
+    FileFormat
+        The inferred inner file format.
 
     Raises
     ------
-    NotImplementedError
-        GZ :func:`read` is not implemented yet.
+    ValueError
+        If the file format cannot be inferred from the filename.
+    """
+    fmt, compression = infer_file_format_and_compression(path)
+    if compression is not CompressionFormat.GZ:
+        raise ValueError(f'Not a gzip file: {path}')
+    if fmt is None:
+        raise ValueError(
+            f'Cannot infer file format from compressed file {path!r}',
+        )
+    return fmt
+
+
+# SECTION: FUNCTIONS ======================================================== #
+
+
+def read(
+    path: Path,
+) -> JSONData:
+    """
+    Read GZ content from ``path`` and parse the inner payload.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the GZ file on disk.
+
+    Returns
+    -------
+    JSONData
+        Parsed payload.
     """
-    raise NotImplementedError('GZ read is not implemented yet')
+    fmt = _resolve_format(path)
+    with gzip.open(path, 'rb') as handle:
+        payload = handle.read()
+
+    with tempfile.TemporaryDirectory() as tmpdir:
+        tmp_path = Path(tmpdir) / f'payload.{fmt.value}'
+        tmp_path.write_bytes(payload)
+        from .core import File
 
+        return File(tmp_path, fmt).read()
 
-def write(path: Path, data: JSONData) -> int:
+
+def write(
+    path: Path,
+    data: JSONData,
+) -> int:
     """
-    Write ``data`` to GZ at ``path``.
+    Write ``data`` to GZ at ``path`` and return record count.
 
     Parameters
     ----------
@@ -50,10 +107,17 @@ def write(path: Path, data: JSONData) -> int:
     -------
     int
         Number of records written.
-
-    Raises
-    ------
-    NotImplementedError
-        GZ :func:`write` is not implemented yet.
     """
-    raise NotImplementedError('GZ write is not implemented yet')
+    fmt = _resolve_format(path)
+    with tempfile.TemporaryDirectory() as tmpdir:
+        tmp_path = Path(tmpdir) / f'payload.{fmt.value}'
+        from .core import File
+
+        count = File(tmp_path, fmt).write(data)
+        payload = tmp_path.read_bytes()
+
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with gzip.open(path, 'wb') as handle:
+        handle.write(payload)
+
+    return count

etlplus/file/json.py

@@ -1,7 +1,7 @@
 """
 :mod:`etlplus.file.json` module.
 
-JSON read/write helpers.
+Helpers for reading/writing JSON files.
 """
 
 from __future__ import annotations
@@ -15,6 +15,15 @@ from ..types import JSONDict
 from ..types import JSONList
 from ..utils import count_records
 
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = [
+    'read',
+    'write',
+]
+
+
 # SECTION: FUNCTIONS ======================================================== #
 
 
@@ -22,7 +31,9 @@ def read(
     path: Path,
 ) -> JSONData:
     """
-    Load and validate JSON payloads from ``path``.
+    Read JSON content from ``path``.
+
+    Validates that the JSON root is a dict or a list of dicts.
 
     Parameters
     ----------

etlplus/file/ndjson.py

@@ -1,19 +1,35 @@
 """
 :mod:`etlplus.file.ndjson` module.
 
-Stub helpers for NDJSON read/write.
+Helpers for reading/writing NDJSON files.
 """
 
 from __future__ import annotations
 
+import json
 from pathlib import Path
+from typing import cast
 
 from ..types import JSONData
+from ..types import JSONDict
+from ..types import JSONList
+from ..utils import count_records
 
 # SECTION: EXPORTS ========================================================== #
 
 
-def read(path: Path) -> JSONData:
+__all__ = [
+    'read',
+    'write',
+]
+
+
+# SECTION: FUNCTIONS ======================================================== #
+
+
+def read(
+    path: Path,
+) -> JSONList:
     """
     Read NDJSON content from ``path``.
 
@@ -24,18 +40,33 @@ def read(path: Path) -> JSONData:
 
     Returns
     -------
-    JSONData
-        Parsed payload.
+    JSONList
+        The list of dictionaries read from the NDJSON file.
 
     Raises
     ------
-    NotImplementedError
-        NDJSON :func:`read` is not implemented yet.
+    TypeError
+        If any line in the NDJSON file is not a JSON object (dict).
     """
-    raise NotImplementedError('NDJSON read is not implemented yet')
+    rows: JSONList = []
+    with path.open('r', encoding='utf-8') as handle:
+        for idx, line in enumerate(handle, start=1):
+            text = line.strip()
+            if not text:
+                continue
+            payload = json.loads(text)
+            if not isinstance(payload, dict):
+                raise TypeError(
+                    f'NDJSON lines must be objects (dicts) (line {idx})',
+                )
+            rows.append(cast(JSONDict, payload))
+    return rows
 
 
-def write(path: Path, data: JSONData) -> int:
+def write(
+    path: Path,
+    data: JSONData,
+) -> int:
     """
     Write ``data`` to NDJSON at ``path``.
 
@@ -53,7 +84,26 @@ def write(path: Path, data: JSONData) -> int:
 
     Raises
     ------
-    NotImplementedError
-        NDJSON :func:`write` is not implemented yet.
+    TypeError
+        If ``data`` is a list containing non-dict items.
     """
-    raise NotImplementedError('NDJSON write is not implemented yet')
+    rows: JSONList
+    if isinstance(data, list):
+        if not all(isinstance(item, dict) for item in data):
+            raise TypeError(
+                'NDJSON payloads must contain only objects (dicts)',
+            )
+        rows = cast(JSONList, data)
+    else:
+        rows = [cast(JSONDict, data)]
+
+    if not rows:
+        return 0
+
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with path.open('w', encoding='utf-8') as handle:
+        for row in rows:
+            handle.write(json.dumps(row, ensure_ascii=False))
+            handle.write('\n')
+
+    return count_records(rows)

etlplus/file/orc.py

@@ -1,19 +1,77 @@
 """
 :mod:`etlplus.file.orc` module.
 
-Stub helpers for ORC read/write.
+Helpers for reading/writing ORC files.
 """
 
 from __future__ import annotations
 
 from pathlib import Path
+from typing import Any
+from typing import cast
 
 from ..types import JSONData
+from ..types import JSONDict
+from ..types import JSONList
 
 # SECTION: EXPORTS ========================================================== #
 
 
-def read(path: Path) -> JSONData:
+__all__ = [
+    'read',
+    'write',
+]
+
+
+# SECTION: INTERNAL CONSTANTS =============================================== #
+
+
+_PANDAS_CACHE: dict[str, Any] = {}
+
+
+# SECTION: INTERNAL FUNCTIONS =============================================== #
+
+
+def _get_pandas() -> Any:
+    """
+    Return the pandas module, importing it on first use.
+
+    Raises an informative ImportError if the optional dependency is missing.
+    """
+    mod = _PANDAS_CACHE.get('mod')
+    if mod is not None:  # pragma: no cover - tiny branch
+        return mod
+    try:
+        _pd = __import__('pandas')  # type: ignore[assignment]
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'ORC support requires optional dependency "pandas".\n'
+            'Install with: pip install pandas',
+        ) from e
+    _PANDAS_CACHE['mod'] = _pd
+
+    return _pd
+
+
+def _normalize_records(data: JSONData) -> JSONList:
+    """
+    Normalize JSON payloads into a list of dictionaries.
+
+    Raises TypeError when payloads contain non-dict items.
+    """
+    if isinstance(data, list):
+        if not all(isinstance(item, dict) for item in data):
+            raise TypeError('ORC payloads must contain only objects (dicts)')
+        return cast(JSONList, data)
+    return [cast(JSONDict, data)]
+
+
+# SECTION: FUNCTIONS ======================================================== #
+
+
+def read(
+    path: Path,
+) -> JSONList:
     """
     Read ORC content from ``path``.
 
@@ -24,20 +82,31 @@ def read(path: Path) -> JSONData:
 
     Returns
     -------
-    JSONData
-        Parsed payload.
+    JSONList
+        The list of dictionaries read from the ORC file.
 
     Raises
     ------
-    NotImplementedError
-        ORC :func:`read` is not implemented yet.
+    ImportError
+        When optional dependency "pyarrow" is missing.
     """
-    raise NotImplementedError('ORC read is not implemented yet')
+    pandas = _get_pandas()
+    try:
+        frame = pandas.read_orc(path)
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'ORC support requires optional dependency "pyarrow".\n'
+            'Install with: pip install pyarrow',
+        ) from e
+    return cast(JSONList, frame.to_dict(orient='records'))
 
 
-def write(path: Path, data: JSONData) -> int:
+def write(
+    path: Path,
+    data: JSONData,
+) -> int:
     """
-    Write ``data`` to ORC at ``path``.
+    Write ``data`` to ORC at ``path`` and return record count.
 
     Parameters
     ----------
@@ -53,7 +122,21 @@ def write(path: Path, data: JSONData) -> int:
 
     Raises
     ------
-    NotImplementedError
-        ORC :func:`write` is not implemented yet.
+    ImportError
+        When optional dependency "pyarrow" is missing.
     """
-    raise NotImplementedError('ORC write is not implemented yet')
+    records = _normalize_records(data)
+    if not records:
+        return 0
+
+    pandas = _get_pandas()
+    path.parent.mkdir(parents=True, exist_ok=True)
+    frame = pandas.DataFrame.from_records(records)
+    try:
+        frame.to_orc(path, index=False)
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'ORC support requires optional dependency "pyarrow".\n'
+            'Install with: pip install pyarrow',
+        ) from e
+    return len(records)