datachain 0.28.0__tar.gz → 0.28.1__tar.gz

{datachain-0.28.0 → datachain-0.28.1}/.pre-commit-config.yaml

@@ -24,7 +24,7 @@ repos:
       - id: trailing-whitespace
         exclude: '^LICENSES/'
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.12.4'
+    rev: 'v0.12.5'
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix]

{datachain-0.28.0 → datachain-0.28.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.28.0
+Version: 0.28.1
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.28.0 → datachain-0.28.1}/src/datachain/lib/dc/datachain.py

@@ -2419,9 +2419,11 @@ class DataChain:
             ds.to_storage("gs://mybucket", placement="filename")
             ```
         """
+        chain = self.persist()
+        count = chain.count()
+
         if placement == "filename" and (
-            self._query.distinct(pathfunc.name(C(f"{signal}__path"))).count()
-            != self._query.count()
+            chain._query.distinct(pathfunc.name(C(f"{signal}__path"))).count() != count
         ):
             raise ValueError("Files with the same name found")
 
@@ -2433,7 +2435,7 @@ class DataChain:
             unit=" files",
             unit_scale=True,
             unit_divisor=10,
-            total=self.count(),
+            total=count,
             leave=False,
         )
         file_exporter = FileExporter(
@@ -2444,7 +2446,10 @@ class DataChain:
             max_threads=num_threads or 1,
             client_config=client_config,
         )
-        file_exporter.run(self.to_values(signal), progress_bar)
+        file_exporter.run(
+            (rows[0] for rows in chain.to_iter(signal)),
+            progress_bar,
+        )
 
     def shuffle(self) -> "Self":
         """Shuffle the rows of the chain deterministically."""

{datachain-0.28.0 → datachain-0.28.1}/src/datachain/lib/file.py

@@ -23,7 +23,7 @@ from pydantic import Field, field_validator
 
 from datachain.client.fileslice import FileSlice
 from datachain.lib.data_model import DataModel
-from datachain.lib.utils import DataChainError
+from datachain.lib.utils import DataChainError, rebase_path
 from datachain.nodes_thread_pool import NodesThreadPool
 from datachain.sql.types import JSON, Boolean, DateTime, Int, String
 from datachain.utils import TIME_ZERO
@@ -634,6 +634,40 @@ class File(DataModel):
             location=self.location,
         )
 
+    def rebase(
+        self,
+        old_base: str,
+        new_base: str,
+        suffix: str = "",
+        extension: str = "",
+    ) -> str:
+        """
+        Rebase the file's URI from one base directory to another.
+
+        Args:
+            old_base: Base directory to remove from the file's URI
+            new_base: New base directory to prepend
+            suffix: Optional suffix to add before file extension
+            extension: Optional new file extension (without dot)
+
+        Returns:
+            str: Rebased URI with new base directory
+
+        Raises:
+            ValueError: If old_base is not found in the file's URI
+
+        Examples:
+            >>> file = File(source="s3://bucket", path="data/2025-05-27/file.wav")
+            >>> file.rebase("s3://bucket/data", "s3://output-bucket/processed", \
+                    extension="mp3")
+            's3://output-bucket/processed/2025-05-27/file.mp3'
+
+            >>> file.rebase("data/audio", "/local/output", suffix="_ch1",
+                    extension="npy")
+            '/local/output/file_ch1.npy'
+        """
+        return rebase_path(self.get_uri(), old_base, new_base, suffix, extension)
+
 
 def resolve(file: File) -> File:
     """
@@ -1219,6 +1253,24 @@ class Audio(DataModel):
     codec: str = Field(default="")
     bit_rate: int = Field(default=-1)
 
+    @staticmethod
+    def get_channel_name(num_channels: int, channel_idx: int) -> str:
+        """Map channel index to meaningful name based on common audio formats"""
+        channel_mappings = {
+            1: ["Mono"],
+            2: ["Left", "Right"],
+            4: ["W", "X", "Y", "Z"],  # First-order Ambisonics
+            6: ["FL", "FR", "FC", "LFE", "BL", "BR"],  # 5.1 surround
+            8: ["FL", "FR", "FC", "LFE", "BL", "BR", "SL", "SR"],  # 7.1 surround
+        }
+
+        if num_channels in channel_mappings:
+            channels = channel_mappings[num_channels]
+            if 0 <= channel_idx < len(channels):
+                return channels[channel_idx]
+
+        return f"Ch{channel_idx + 1}"
+
 
 class ArrowRow(DataModel):
     """`DataModel` for reading row from Arrow-supported file."""

datachain-0.28.1/src/datachain/lib/utils.py

@@ -0,0 +1,155 @@
+import re
+from abc import ABC, abstractmethod
+from collections.abc import Sequence
+from pathlib import PurePosixPath
+from urllib.parse import urlparse
+
+
+class AbstractUDF(ABC):
+    @abstractmethod
+    def process(self, *args, **kwargs):
+        pass
+
+    @abstractmethod
+    def setup(self):
+        pass
+
+    @abstractmethod
+    def teardown(self):
+        pass
+
+
+class DataChainError(Exception):
+    pass
+
+
+class DataChainParamsError(DataChainError):
+    pass
+
+
+class DataChainColumnError(DataChainParamsError):
+    def __init__(self, col_name: str, msg: str):
+        super().__init__(f"Error for column {col_name}: {msg}")
+
+
+def normalize_col_names(col_names: Sequence[str]) -> dict[str, str]:
+    """Returns normalized_name -> original_name dict."""
+    gen_col_counter = 0
+    new_col_names = {}
+    org_col_names = set(col_names)
+
+    for org_column in col_names:
+        new_column = org_column.lower()
+        new_column = re.sub("[^0-9a-z]+", "_", new_column)
+        new_column = new_column.strip("_")
+
+        generated_column = new_column
+
+        while (
+            not generated_column.isidentifier()
+            or generated_column in new_col_names
+            or (generated_column != org_column and generated_column in org_col_names)
+        ):
+            if new_column:
+                generated_column = f"c{gen_col_counter}_{new_column}"
+            else:
+                generated_column = f"c{gen_col_counter}"
+            gen_col_counter += 1
+
+        new_col_names[generated_column] = org_column
+
+    return new_col_names
+
+
+def rebase_path(
+    src_path: str,
+    old_base: str,
+    new_base: str,
+    suffix: str = "",
+    extension: str = "",
+) -> str:
+    """
+    Rebase a file path from one base directory to another.
+
+    Args:
+        src_path: Source file path (can include URI scheme like s3://)
+        old_base: Base directory to remove from src_path
+        new_base: New base directory to prepend
+        suffix: Optional suffix to add before file extension
+        extension: Optional new file extension (without dot)
+
+    Returns:
+        str: Rebased path with new base directory
+
+    Raises:
+        ValueError: If old_base is not found in src_path
+    """
+    # Parse URIs to handle schemes properly
+    src_parsed = urlparse(src_path)
+    old_base_parsed = urlparse(old_base)
+    new_base_parsed = urlparse(new_base)
+
+    # Get the path component (without scheme)
+    if src_parsed.scheme:
+        src_path_only = src_parsed.netloc + src_parsed.path
+    else:
+        src_path_only = src_path
+
+    if old_base_parsed.scheme:
+        old_base_only = old_base_parsed.netloc + old_base_parsed.path
+    else:
+        old_base_only = old_base
+
+    # Normalize paths
+    src_path_norm = PurePosixPath(src_path_only).as_posix()
+    old_base_norm = PurePosixPath(old_base_only).as_posix()
+
+    # Find where old_base appears in src_path
+    if old_base_norm in src_path_norm:
+        # Find the index where old_base appears
+        idx = src_path_norm.find(old_base_norm)
+        if idx == -1:
+            raise ValueError(f"old_base '{old_base}' not found in src_path")
+
+        # Extract the relative path after old_base
+        relative_start = idx + len(old_base_norm)
+        # Skip leading slash if present
+        if relative_start < len(src_path_norm) and src_path_norm[relative_start] == "/":
+            relative_start += 1
+        relative_path = src_path_norm[relative_start:]
+    else:
+        raise ValueError(f"old_base '{old_base}' not found in src_path")
+
+    # Parse the filename
+    path_obj = PurePosixPath(relative_path)
+    stem = path_obj.stem
+    current_ext = path_obj.suffix
+
+    # Apply suffix and extension changes
+    new_stem = stem + suffix if suffix else stem
+    if extension:
+        new_ext = f".{extension}"
+    elif current_ext:
+        new_ext = current_ext
+    else:
+        new_ext = ""
+
+    # Build new filename
+    new_name = new_stem + new_ext
+
+    # Reconstruct path with new base
+    parent = str(path_obj.parent)
+    if parent == ".":
+        new_relative_path = new_name
+    else:
+        new_relative_path = str(PurePosixPath(parent) / new_name)
+
+    # Handle new_base URI scheme
+    if new_base_parsed.scheme:
+        # Has schema like s3://
+        base_path = new_base_parsed.netloc + new_base_parsed.path
+        base_path = PurePosixPath(base_path).as_posix()
+        full_path = str(PurePosixPath(base_path) / new_relative_path)
+        return f"{new_base_parsed.scheme}://{full_path}"
+    # Regular path
+    return str(PurePosixPath(new_base) / new_relative_path)

{datachain-0.28.0 → datachain-0.28.1}/src/datachain.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.28.0
+Version: 0.28.1
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.28.0 → datachain-0.28.1}/tests/func/test_datachain.py

@@ -9,7 +9,7 @@ import uuid
 from collections.abc import Iterator
 from datetime import datetime, timedelta, timezone
 from pathlib import Path, PurePosixPath
-from unittest.mock import patch
+from unittest.mock import Mock, patch
 
 import numpy as np
 import pandas as pd
@@ -358,15 +358,24 @@ def test_to_storage(
     file_type,
     num_threads,
 ):
+    mapper = Mock(side_effect=lambda file_path: len(file_path))
+
     ctc = cloud_test_catalog
     df = dc.read_storage(ctc.src_uri, type=file_type, session=test_session)
     if use_map:
-        df.settings(cache=use_cache).map(
-            res=lambda file: file.export(tmp_dir / "output", placement=placement)
-        ).exec()
+        (
+            df.settings(cache=use_cache)
+            .map(mapper, params=["file.path"], output={"path_len": int})
+            .map(res=lambda file: file.export(tmp_dir / "output", placement=placement))
+            .exec()
+        )
     else:
-        df.settings(cache=use_cache).to_storage(
-            tmp_dir / "output", placement=placement, num_threads=num_threads
+        (
+            df.settings(cache=use_cache)
+            .map(mapper, params=["file.path"], output={"path_len": int})
+            .to_storage(
+                tmp_dir / "output", placement=placement, num_threads=num_threads
+            )
         )
 
     expected = {
@@ -387,6 +396,8 @@ def test_to_storage(
         with open(tmp_dir / "output" / file_path) as f:
             assert f.read() == expected[file.name]
 
+    assert mapper.call_count == len(expected)
+
 
 @pytest.mark.parametrize("use_cache", [True, False])
 def test_export_images_files(test_session, tmp_dir, tmp_path, use_cache):

{datachain-0.28.0 → datachain-0.28.1}/tests/unit/lib/test_file.py

@@ -7,7 +7,7 @@ from fsspec.implementations.local import LocalFileSystem
 from PIL import Image
 
 from datachain.catalog import Catalog
-from datachain.lib.file import File, FileError, ImageFile, TextFile, resolve
+from datachain.lib.file import Audio, File, FileError, ImageFile, TextFile, resolve
 
 
 def create_file(source: str):
@@ -409,3 +409,49 @@ def test_path_normalized(path, expected, raises):
             file.get_path_normalized()
     else:
         assert file.get_path_normalized() == expected
+
+
+def test_file_rebase_method():
+    """Test File.rebase() method"""
+    file = File(source="s3://bucket", path="data/audio/file.wav")
+
+    # Basic rebase
+    result = file.rebase("s3://bucket/data/audio", "s3://output-bucket/waveforms")
+    assert result == "s3://output-bucket/waveforms/file.wav"
+
+    # With suffix and extension
+    result = file.rebase(
+        "s3://bucket/data/audio",
+        "s3://output-bucket/processed",
+        suffix="_ch1",
+        extension="npy",
+    )
+    assert result == "s3://output-bucket/processed/file_ch1.npy"
+
+
+def test_file_rebase_local_path():
+    """Test File.rebase() with local file paths"""
+    file = File(source="file://", path="/data/audio/folder/file.mp3")
+
+    result = file.rebase("file:///data/audio", "/output/processed")
+    assert result == "/output/processed/folder/file.mp3"
+
+
+def test_audio_get_channel_name():
+    # Test known channel configurations
+    assert Audio.get_channel_name(1, 0) == "Mono"
+    assert Audio.get_channel_name(2, 0) == "Left"
+    assert Audio.get_channel_name(2, 1) == "Right"
+    assert Audio.get_channel_name(4, 2) == "Y"  # Ambisonics
+    assert Audio.get_channel_name(6, 3) == "LFE"  # 5.1 surround
+    assert Audio.get_channel_name(8, 7) == "SR"  # 7.1 surround
+
+    # Test fallback for unknown configurations
+    assert Audio.get_channel_name(-1, 0) == "Ch1"
+    assert Audio.get_channel_name(3, 0) == "Ch1"
+    assert Audio.get_channel_name(5, 4) == "Ch5"
+    assert Audio.get_channel_name(10, 9) == "Ch10"
+
+    # Test out of range indices
+    assert Audio.get_channel_name(2, 5) == "Ch6"
+    assert Audio.get_channel_name(1, 1) == "Ch2"

{datachain-0.28.0 → datachain-0.28.1}/tests/unit/lib/test_utils.py

@@ -5,7 +5,7 @@ import pytest
 from pydantic import BaseModel
 
 from datachain.lib.convert.python_to_sql import python_to_sql
-from datachain.lib.utils import normalize_col_names
+from datachain.lib.utils import normalize_col_names, rebase_path
 from datachain.sql.types import Array, String
 
 
@@ -110,3 +110,72 @@ def test_normalize_column_names_repeat_generated_after_normalize():
     res = normalize_col_names(["c0_CoLuMn", "_column", "column"])
 
     assert res == {"c0_column": "c0_CoLuMn", "c1_column": "_column", "column": "column"}
+
+
+def test_rebase_path_basic():
+    result = rebase_path(
+        "/data/audio/folder1/file.wav", "/data/audio", "/output/waveforms"
+    )
+    assert result == "/output/waveforms/folder1/file.wav"
+
+
+def test_rebase_path_with_s3_uri():
+    result = rebase_path(
+        "s3://bucket/data/audio/folder/file.wav",
+        "data/audio",
+        "s3://output-bucket/waveforms",
+    )
+    assert result == "s3://output-bucket/waveforms/folder/file.wav"
+
+
+def test_rebase_path_mixed_uri_schemes():
+    result = rebase_path(
+        "/local/data/audio/file.mp3", "/local/data/audio", "s3://bucket/output"
+    )
+    assert result == "s3://bucket/output/file.mp3"
+
+
+def test_rebase_path_with_suffix():
+    result = rebase_path(
+        "/data/audio/file.wav", "/data/audio", "/output", suffix="_processed"
+    )
+    assert result == "/output/file_processed.wav"
+
+
+def test_rebase_path_with_extension_change():
+    result = rebase_path("/data/audio/file.wav", "audio", "/output", extension="npy")
+    assert result == "/output/file.npy"
+
+
+def test_rebase_path_base_dir_not_in_path():
+    with pytest.raises(
+        ValueError, match="old_base '/data/audio' not found in src_path"
+    ):
+        rebase_path("/different/path/file.wav", "/data/audio", "/output")
+
+
+def test_rebase_path_partial_match_base_dir():
+    result = rebase_path("/home/user/data/audio/file.wav", "data/audio", "/output")
+    assert result == "/output/file.wav"
+
+
+def test_rebase_path_complex_s3_paths():
+    result = rebase_path(
+        "s3://bucket/balanced_train_segments/audio/folder/file.flac",
+        "s3://bucket/balanced_train_segments",
+        "s3://output-bucket/waveforms",
+        suffix="_ch1",
+        extension="npy",
+    )
+    assert result == "s3://output-bucket/waveforms/audio/folder/file_ch1.npy"
+
+
+def test_rebase_path_file_without_extension():
+    result = rebase_path("/data/audio/file_no_ext", "/data/audio", "/output")
+    assert result == "/output/file_no_ext"
+
+    # With new extension
+    result = rebase_path(
+        "/data/audio/file_no_ext", "/data/audio", "/output", extension="txt"
+    )
+    assert result == "/output/file_no_ext.txt"

datachain-0.28.0/src/datachain/lib/utils.py

@@ -1,59 +0,0 @@
-import re
-from abc import ABC, abstractmethod
-from collections.abc import Sequence
-
-
-class AbstractUDF(ABC):
-    @abstractmethod
-    def process(self, *args, **kwargs):
-        pass
-
-    @abstractmethod
-    def setup(self):
-        pass
-
-    @abstractmethod
-    def teardown(self):
-        pass
-
-
-class DataChainError(Exception):
-    pass
-
-
-class DataChainParamsError(DataChainError):
-    pass
-
-
-class DataChainColumnError(DataChainParamsError):
-    def __init__(self, col_name: str, msg: str):
-        super().__init__(f"Error for column {col_name}: {msg}")
-
-
-def normalize_col_names(col_names: Sequence[str]) -> dict[str, str]:
-    """Returns normalized_name -> original_name dict."""
-    gen_col_counter = 0
-    new_col_names = {}
-    org_col_names = set(col_names)
-
-    for org_column in col_names:
-        new_column = org_column.lower()
-        new_column = re.sub("[^0-9a-z]+", "_", new_column)
-        new_column = new_column.strip("_")
-
-        generated_column = new_column
-
-        while (
-            not generated_column.isidentifier()
-            or generated_column in new_col_names
-            or (generated_column != org_column and generated_column in org_col_names)
-        ):
-            if new_column:
-                generated_column = f"c{gen_col_counter}_{new_column}"
-            else:
-                generated_column = f"c{gen_col_counter}"
-            gen_col_counter += 1
-
-        new_col_names[generated_column] = org_column
-
-    return new_col_names