pixeltable 0.3.1__py3-none-any.whl → 0.3.3__py3-none-any.whl

pixeltable/io/pandas.py

@@ -9,10 +9,13 @@ import pixeltable.type_system as ts
 
 
 def import_pandas(
-    tbl_name: str, df: pd.DataFrame, *, schema_overrides: Optional[dict[str, pxt.ColumnType]] = None,
+    tbl_name: str,
+    df: pd.DataFrame,
+    *,
+    schema_overrides: Optional[dict[str, pxt.ColumnType]] = None,
     primary_key: Optional[Union[str, list[str]]] = None,
     num_retained_versions: int = 10,
-    comment: str = ''
+    comment: str = '',
 ) -> pxt.Table:
     """Creates a new base table from a Pandas
     [`DataFrame`](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), with the
@@ -45,17 +48,21 @@ def import_pandas(
 
     schema, pxt_pk = __df_to_pxt_schema(df, schema_overrides, primary_key)
     tbl_rows = (dict(__df_row_to_pxt_row(row, schema)) for row in df.itertuples())
-    table = pxt.create_table(tbl_name, schema, primary_key=pxt_pk, num_retained_versions=num_retained_versions, comment=comment)
+    table = pxt.create_table(
+        tbl_name, schema, primary_key=pxt_pk, num_retained_versions=num_retained_versions, comment=comment
+    )
     table.insert(tbl_rows)
     return table
 
 
 def import_csv(
-    tbl_name: str, filepath_or_buffer, schema_overrides: Optional[dict[str, ts.ColumnType]] = None,
+    tbl_name: str,
+    filepath_or_buffer,
+    schema_overrides: Optional[dict[str, ts.ColumnType]] = None,
     primary_key: Optional[Union[str, list[str]]] = None,
     num_retained_versions: int = 10,
     comment: str = '',
-    **kwargs
+    **kwargs,
 ) -> pxt.Table:
     """
     Creates a new base table from a csv file. This is a convenience method and is equivalent
@@ -67,15 +74,25 @@ def import_csv(
         A handle to the newly created [`Table`][pixeltable.Table].
     """
     df = pd.read_csv(filepath_or_buffer, **kwargs)
-    return import_pandas(tbl_name, df, schema_overrides=schema_overrides, primary_key=primary_key, num_retained_versions=num_retained_versions, comment=comment)
+    return import_pandas(
+        tbl_name,
+        df,
+        schema_overrides=schema_overrides,
+        primary_key=primary_key,
+        num_retained_versions=num_retained_versions,
+        comment=comment,
+    )
 
 
 def import_excel(
-    tbl_name: str, io, *args, schema_overrides: Optional[dict[str, ts.ColumnType]] = None,
+    tbl_name: str,
+    io,
+    *args,
+    schema_overrides: Optional[dict[str, ts.ColumnType]] = None,
     primary_key: Optional[Union[str, list[str]]] = None,
     num_retained_versions: int = 10,
     comment: str = '',
-    **kwargs
+    **kwargs,
 ) -> pxt.Table:
     """
     Creates a new base table from an Excel (.xlsx) file. This is a convenience method and is
@@ -87,7 +104,14 @@ def import_excel(
         A handle to the newly created [`Table`][pixeltable.Table].
     """
     df = pd.read_excel(io, *args, **kwargs)
-    return import_pandas(tbl_name, df, schema_overrides=schema_overrides, primary_key=primary_key, num_retained_versions=num_retained_versions, comment=comment)
+    return import_pandas(
+        tbl_name,
+        df,
+        schema_overrides=schema_overrides,
+        primary_key=primary_key,
+        num_retained_versions=num_retained_versions,
+        comment=comment,
+    )
 
 
 def __df_to_pxt_schema(

pixeltable/io/parquet.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import datetime
 import io
 import json
 import logging
@@ -11,19 +12,19 @@ from typing import Any, Optional, Union
 
 import numpy as np
 import PIL.Image
-import datetime
 
 import pixeltable as pxt
-from pixeltable.env import Env
 import pixeltable.exceptions as exc
 import pixeltable.type_system as ts
+from pixeltable.env import Env
 from pixeltable.utils.transactional_directory import transactional_directory
 
 if typing.TYPE_CHECKING:
     import pyarrow as pa
+
     import pixeltable as pxt
 
-_logger = logging.getLogger(__name__)
+_logger = logging.getLogger('pixeltable')
 
 
 def _write_batch(value_batch: dict[str, deque], schema: pa.Schema, output_path: Path) -> None:
@@ -43,11 +44,11 @@ def _write_batch(value_batch: dict[str, deque], schema: pa.Schema, output_path:
 
 
 def export_parquet(
-            table_or_df: Union[pxt.Table, pxt.DataFrame],
-            parquet_path: Path,
-            partition_size_bytes: int = 100_000_000,
-            inline_images: bool = False
-            ) -> None:
+    table_or_df: Union[pxt.Table, pxt.DataFrame],
+    parquet_path: Path,
+    partition_size_bytes: int = 100_000_000,
+    inline_images: bool = False,
+) -> None:
     """
     Exports a dataframe's data to one or more Parquet files. Requires pyarrow to be installed.
 
@@ -159,11 +160,7 @@ def parquet_schema_to_pixeltable_schema(parquet_path: str) -> dict[str, Optional
 
 
 def import_parquet(
-    table: str,
-    *,
-    parquet_path: str,
-    schema_overrides: Optional[dict[str, ts.ColumnType]] = None,
-    **kwargs: Any,
+    table: str, *, parquet_path: str, schema_overrides: Optional[dict[str, ts.ColumnType]] = None, **kwargs: Any
 ) -> pxt.Table:
     """Creates a new base table from a Parquet file or set of files. Requires pyarrow to be installed.
 

pixeltable/iterators/__init__.py

@@ -1,3 +1,4 @@
+from .audio import AudioSplitter
 from .base import ComponentIterator
 from .document import DocumentSplitter
 from .image import TileIterator

pixeltable/iterators/audio.py

@@ -0,0 +1,205 @@
+import logging
+import math
+import uuid
+from fractions import Fraction
+from pathlib import Path
+from typing import Any, Optional
+
+import av  # type: ignore[import-untyped]
+
+import pixeltable.env as env
+import pixeltable.exceptions as excs
+import pixeltable.type_system as ts
+
+from .base import ComponentIterator
+
+_logger = logging.getLogger('pixeltable')
+
+
+class AudioSplitter(ComponentIterator):
+    """
+    Iterator over chunks of an audio file. The audio file is split into smaller chunks, where the duration of each chunk is determined by chunk_duration_sec.
+    The iterator yields audio chunks as pxt.Audio, along with the start and end time of each chunk.
+    If the input contains no audio, no chunks are yielded.
+
+    Args:
+        chunk_duration_sec: Audio chunk duration in seconds
+        overlap_sec: Overlap between consecutive chunks in seconds.
+        min_chunk_duration_sec: Drop the last chunk if it is smaller than min_chunk_duration_sec
+    """
+
+    # Input parameters
+    audio_path: Path
+    chunk_duration_sec: float
+    overlap_sec: float
+
+    # audio stream details
+    container: av.container.input.InputContainer
+    audio_time_base: Fraction  # seconds per presentation time
+
+    # List of chunks to extract
+    # Each chunk is defined by start and end presentation timestamps in audio file (int)
+    chunks_to_extract_in_pts: Optional[list[tuple[int, int]]] = []
+    # next chunk to extract
+    next_pos: int
+
+    __codec_map = {
+        'mp3': 'mp3',  # MP3 decoder -> mp3/libmp3lame encoder
+        'mp3float': 'mp3',  # MP3float decoder -> mp3 encoder
+        'aac': 'aac',  # AAC decoder -> AAC encoder
+        'vorbis': 'libvorbis',  # Vorbis decoder -> libvorbis encoder
+        'opus': 'libopus',  # Opus decoder -> libopus encoder
+        'flac': 'flac',  # FLAC decoder -> FLAC encoder
+        'wavpack': 'wavpack',  # WavPack decoder -> WavPack encoder
+        'alac': 'alac',  # ALAC decoder -> ALAC encoder
+    }
+
+    def __init__(
+        self, audio: str, chunk_duration_sec: float, *, overlap_sec: float = 0.0, min_chunk_duration_sec: float = 0.0
+    ):
+        if chunk_duration_sec <= 0.0:
+            raise excs.Error('chunk_duration_sec must be a positive number')
+        if chunk_duration_sec < min_chunk_duration_sec:
+            raise excs.Error('chunk_duration_sec must be at least min_chunk_duration_sec')
+        if overlap_sec >= chunk_duration_sec:
+            raise excs.Error('overlap_sec must be less than chunk_duration_sec')
+        audio_path = Path(audio)
+        assert audio_path.exists() and audio_path.is_file()
+        self.audio_path = audio_path
+        self.next_pos = 0
+        self.container = av.open(str(audio_path))
+        if len(self.container.streams.audio) == 0:
+            # No audio stream
+            return
+        self.chunk_duration_sec = chunk_duration_sec
+        self.overlap_sec = overlap_sec
+        self.min_chunk_duration_sec = min_chunk_duration_sec
+        self.audio_time_base = self.container.streams.audio[0].time_base
+
+        audio_start_time_pts = self.container.streams.audio[0].start_time or 0
+        audio_start_time_sec = float(audio_start_time_pts * self.audio_time_base)
+        total_audio_duration_pts = self.container.streams.audio[0].duration or 0
+        total_audio_duration_sec = float(total_audio_duration_pts * self.audio_time_base)
+
+        self.chunks_to_extract_in_pts = [
+            (round(start / self.audio_time_base), round(end / self.audio_time_base))
+            for (start, end) in self.build_chunks(
+                audio_start_time_sec, total_audio_duration_sec, chunk_duration_sec, overlap_sec, min_chunk_duration_sec
+            )
+        ]
+        _logger.debug(
+            f'AudioIterator: path={self.audio_path} total_audio_duration_pts={total_audio_duration_pts} chunks_to_extract_in_pts={self.chunks_to_extract_in_pts}'
+        )
+
+    @classmethod
+    def build_chunks(
+        cls,
+        start_time_sec: float,
+        total_duration_sec: float,
+        chunk_duration_sec: float,
+        overlap_sec: float,
+        min_chunk_duration_sec: float,
+    ) -> list[tuple[float, float]]:
+        chunks_to_extract_in_sec: list[tuple[float, float]] = []
+        current_pos = start_time_sec
+        end_time = start_time_sec + total_duration_sec
+        while current_pos < end_time:
+            chunk_start = current_pos
+            chunk_end = min(chunk_start + chunk_duration_sec, end_time)
+            chunks_to_extract_in_sec.append((chunk_start, chunk_end))
+            if chunk_end >= end_time:
+                break
+            current_pos = chunk_end - overlap_sec
+        # If the last chunk is smaller than min_chunk_duration_sec then drop the last chunk from the list
+        if (
+            len(chunks_to_extract_in_sec) > 0
+            and (chunks_to_extract_in_sec[-1][1] - chunks_to_extract_in_sec[-1][0]) < min_chunk_duration_sec
+        ):
+            return chunks_to_extract_in_sec[:-1]  # return all but the last chunk
+        return chunks_to_extract_in_sec
+
+    @classmethod
+    def input_schema(cls) -> dict[str, ts.ColumnType]:
+        return {
+            'audio': ts.AudioType(nullable=False),
+            'chunk_duration_sec': ts.FloatType(nullable=True),
+            'overlap_sec': ts.FloatType(nullable=True),
+            'min_chunk_duration_sec': ts.FloatType(nullable=True),
+        }
+
+    @classmethod
+    def output_schema(cls, *args: Any, **kwargs: Any) -> tuple[dict[str, ts.ColumnType], list[str]]:
+        return {
+            'start_time_sec': ts.FloatType(),
+            'end_time_sec': ts.FloatType(),
+            'audio_chunk': ts.AudioType(nullable=True),
+        }, []
+
+    def __next__(self) -> dict[str, Any]:
+        if self.next_pos >= len(self.chunks_to_extract_in_pts):
+            raise StopIteration
+        target_chunk_start, target_chunk_end = self.chunks_to_extract_in_pts[self.next_pos]
+        chunk_start_pts = 0
+        chunk_end_pts = 0
+        chunk_file = str(env.Env.get().tmp_dir / f'{uuid.uuid4()}{self.audio_path.suffix}')
+        output_container = av.open(chunk_file, mode='w')
+        input_stream = self.container.streams.audio[0]
+        codec_name = AudioSplitter.__codec_map.get(input_stream.codec_context.name, input_stream.codec_context.name)
+        output_stream = output_container.add_stream(codec_name, rate=input_stream.codec_context.sample_rate)
+        frame_count = 0
+        # Since frames don't align with chunk boundaries, we may have read an extra frame in previous iteration
+        # Seek to the nearest frame in stream at current chunk start time
+        self.container.seek(target_chunk_start, backward=True, stream=self.container.streams.audio[0])
+        while True:
+            try:
+                frame = next(self.container.decode(audio=0))
+            except EOFError as e:
+                raise excs.Error(f'Failed to read audio file `{self.audio_path}`, error `{e}`')
+            except StopIteration:
+                # no more frames to scan
+                break
+            if frame.pts < target_chunk_start:
+                # Current frame is behind chunk's start time, always get frame next to chunk's start time
+                continue
+            if frame.pts >= target_chunk_end:
+                # Frame has crossed the chunk boundary, it should be picked up by next chunk, throw away the current frame
+                break
+            frame_end = frame.pts + frame.samples
+            if frame_count == 0:
+                # Record start of the first frame
+                chunk_start_pts = frame.pts
+            # Write frame to output container
+            frame_count += 1
+            # If encode returns packets, write them to output container. Some encoders will buffer the frames.
+            output_container.mux(output_stream.encode(frame))
+            # record this frame's end as chunks end
+            chunk_end_pts = frame_end
+            # Check if frame's end has crossed the chunk boundary
+            if frame_end >= target_chunk_end:
+                break
+
+        # record result
+        if frame_count > 0:
+            # flush encoder
+            output_container.mux(output_stream.encode(None))
+            output_container.close()
+            result = {
+                'start_time_sec': round(float(chunk_start_pts * self.audio_time_base), 4),
+                'end_time_sec': round(float(chunk_end_pts * self.audio_time_base), 4),
+                'audio_chunk': chunk_file if frame_count > 0 else None,
+            }
+            _logger.debug('audio chunk result: %s', result)
+            self.next_pos += 1
+            return result
+        else:
+            # It's possible that there are no frames in the range of the last chunk, stop the iterator in this case.
+            # Note that start_time points at the first frame so case applies only for the last chunk
+            assert self.next_pos == len(self.chunks_to_extract_in_pts) - 1
+            self.next_pos += 1
+            raise StopIteration
+
+    def close(self) -> None:
+        self.container.close()
+
+    def set_pos(self, pos: int) -> None:
+        pass

pixeltable/iterators/document.py

@@ -35,6 +35,7 @@ class Separator(enum.Enum):
 @dataclasses.dataclass
 class DocumentSectionMetadata:
     """Metadata for a subsection of a document (ie, a structural element like a heading or paragraph)"""
+
     # html and markdown metadata
     sourceline: Optional[int] = None
     # the stack of headings up to the most recently observed one;
@@ -50,6 +51,7 @@ class DocumentSectionMetadata:
 @dataclasses.dataclass
 class DocumentSection:
     """A single document chunk, according to some of the splitting criteria"""
+
     text: Optional[str]
     metadata: Optional[DocumentSectionMetadata]
 
@@ -93,6 +95,7 @@ class DocumentSplitter(ComponentIterator):
 
     Chunked text will be cleaned with `ftfy.fix_text` to fix up common problems with unicode sequences.
     """
+
     METADATA_COLUMN_TYPES = {
         ChunkMetadata.TITLE: StringType(nullable=True),
         ChunkMetadata.HEADING: JsonType(nullable=True),
@@ -102,10 +105,16 @@ class DocumentSplitter(ComponentIterator):
     }
 
     def __init__(
-            self, document: str, *, separators: str, limit: Optional[int] = None, overlap: Optional[int] = None,
-            metadata: str = '',
-            html_skip_tags: Optional[list[str]] = None, tiktoken_encoding: Optional[str] = 'cl100k_base',
-            tiktoken_target_model: Optional[str] = None
+        self,
+        document: str,
+        *,
+        separators: str,
+        limit: Optional[int] = None,
+        overlap: Optional[int] = None,
+        metadata: str = '',
+        html_skip_tags: Optional[list[str]] = None,
+        tiktoken_encoding: Optional[str] = 'cl100k_base',
+        tiktoken_target_model: Optional[str] = None,
     ):
         """Init method for `DocumentSplitter` class.
 
@@ -234,13 +243,14 @@ class DocumentSplitter(ComponentIterator):
     def _html_sections(self) -> Iterator[DocumentSection]:
         """Create DocumentSections reflecting the html-specific separators"""
         import bs4
+
         emit_on_paragraph = Separator.PARAGRAPH in self._separators or Separator.SENTENCE in self._separators
         emit_on_heading = Separator.HEADING in self._separators or emit_on_paragraph
         # current state
         accumulated_text: list[str] = []  # currently accumulated text
         # accumulate pieces then join before emit to avoid quadratic complexity of string concatenation
 
-        headings: dict[str, str] = {}   # current state of observed headings (level -> text)
+        headings: dict[str, str] = {}  # current state of observed headings (level -> text)
         sourceline = 0  # most recently seen sourceline
 
         def update_metadata(el: bs4.Tag) -> None:
@@ -300,7 +310,7 @@ class DocumentSplitter(ComponentIterator):
         # current state
         accumulated_text: list[str] = []  # currently accumulated text
         # accumulate pieces then join before emit to avoid quadratic complexity of string concatenation
-        headings: dict[str, str] = {}   # current state of observed headings (level -> text)
+        headings: dict[str, str] = {}  # current state of observed headings (level -> text)
 
         def update_headings(heading: dict) -> None:
             # update current state
@@ -353,6 +363,7 @@ class DocumentSplitter(ComponentIterator):
     def _pdf_sections(self) -> Iterator[DocumentSection]:
         """Create DocumentSections reflecting the pdf-specific separators"""
         import fitz  # type: ignore[import-untyped]
+
         doc: fitz.Document = self._doc_handle.pdf_doc
         assert doc is not None
 
@@ -385,8 +396,7 @@ class DocumentSplitter(ComponentIterator):
                     yield DocumentSection(text=_emit_text(), metadata=metadata)
 
             if accumulated_text and emit_on_page and not emit_on_paragraph:
-                yield DocumentSection(text=_emit_text(),
-                                      metadata=DocumentSectionMetadata(page=page_number))
+                yield DocumentSection(text=_emit_text(), metadata=DocumentSectionMetadata(page=page_number))
                 accumulated_text = []
 
         if accumulated_text and not emit_on_page:
@@ -411,6 +421,7 @@ class DocumentSplitter(ComponentIterator):
 
     def _token_chunks(self, input: Iterable[DocumentSection]) -> Iterator[DocumentSection]:
         import tiktoken
+
         if self._tiktoken_target_model is not None:
             encoding = tiktoken.encoding_for_model(self._tiktoken_target_model)
         else:

pixeltable/iterators/image.py

@@ -30,15 +30,9 @@ class TileIterator(ComponentIterator):
     __i: int
     __j: int
 
-    def __init__(
-        self,
-        image: PIL.Image.Image,
-        *,
-        tile_size: tuple[int, int],
-        overlap: tuple[int, int] = (0, 0),
-    ):
+    def __init__(self, image: PIL.Image.Image, *, tile_size: tuple[int, int], overlap: tuple[int, int] = (0, 0)):
         if overlap[0] >= tile_size[0] or overlap[1] >= tile_size[1]:
-            raise excs.Error(f"overlap dimensions {overlap} are not strictly smaller than tile size {tile_size}")
+            raise excs.Error(f'overlap dimensions {overlap} are not strictly smaller than tile size {tile_size}')
 
         self.__image = image
         self.__image.load()
@@ -64,11 +58,7 @@ class TileIterator(ComponentIterator):
         x2 = x1 + self.__tile_size[0]
         y2 = y1 + self.__tile_size[1]
         tile = self.__image.crop((x1, y1, x2, y2))
-        result = {
-            'tile': tile,
-            'tile_coord': [self.__i, self.__j],
-            'tile_box': [x1, y1, x2, y2]
-        }
+        result = {'tile': tile, 'tile_coord': [self.__i, self.__j], 'tile_box': [x1, y1, x2, y2]}
 
         self.__i += 1
         if self.__i >= self.__xlen:
@@ -85,16 +75,8 @@ class TileIterator(ComponentIterator):
 
     @classmethod
     def input_schema(cls, *args: Any, **kwargs: Any) -> dict[str, ts.ColumnType]:
-        return {
-            'image': ts.ImageType(),
-            'tile_size': ts.JsonType(),
-            'overlap': ts.JsonType(),
-        }
+        return {'image': ts.ImageType(), 'tile_size': ts.JsonType(), 'overlap': ts.JsonType()}
 
     @classmethod
-    def output_schema(cls,  *args: Any, **kwargs: Any) -> tuple[dict[str, ts.ColumnType], list[str]]:
-        return {
-            'tile': ts.ImageType(),
-            'tile_coord': ts.JsonType(),
-            'tile_box': ts.JsonType(),
-        }, ['tile']
+    def output_schema(cls, *args: Any, **kwargs: Any) -> tuple[dict[str, ts.ColumnType], list[str]]:
+        return {'tile': ts.ImageType(), 'tile_coord': ts.JsonType(), 'tile_box': ts.JsonType()}, ['tile']

pixeltable/iterators/string.py

@@ -1,4 +1,4 @@
-from typing import Iterator, Any
+from typing import Any, Iterator
 
 import pixeltable.exceptions as excs
 import pixeltable.type_system as ts
@@ -30,11 +30,8 @@ class StringSplitter(ComponentIterator):
 
     @classmethod
     def input_schema(cls, *args: Any, **kwargs: Any) -> dict[str, ts.ColumnType]:
-        return {
-            'text': ts.StringType(),
-            'separators': ts.StringType(),
-        }
+        return {'text': ts.StringType(), 'separators': ts.StringType()}
 
     @classmethod
-    def output_schema(cls,  *args: Any, **kwargs: Any) -> tuple[dict[str, ts.ColumnType], list[str]]:
+    def output_schema(cls, *args: Any, **kwargs: Any) -> tuple[dict[str, ts.ColumnType], list[str]]:
         return {'text': ts.StringType()}, []

pixeltable/iterators/video.py

@@ -24,7 +24,6 @@ class FrameIterator(ComponentIterator):
     frame of the video will always be extracted, and the remaining frames will be spaced as evenly as possible.
 
     Args:
-        video: URL or path of the video to use for frame extraction.
         fps: Number of frames to extract per second of video. This may be a fractional value, such as 0.5.
             If omitted or set to 0.0, then the native framerate of the video will be used (all frames will be
             extracted). If `fps` is greater than the frame rate of the video, an error will be raised.
@@ -167,12 +166,7 @@ class FrameIterator(ComponentIterator):
             img = frame.to_image()
             assert isinstance(img, PIL.Image.Image)
             pos_msec = float(pts * self.video_time_base * 1000)
-            result = {
-                'frame_idx': self.next_pos,
-                'pos_msec': pos_msec,
-                'pos_frame': video_idx,
-                'frame': img,
-            }
+            result = {'frame_idx': self.next_pos, 'pos_msec': pos_msec, 'pos_frame': video_idx, 'frame': img}
             self.next_pos += 1
             return result
 

pixeltable/metadata/__init__.py

@@ -10,7 +10,7 @@ import sqlalchemy.orm as orm
 from .schema import SystemInfo, SystemInfoMd
 
 # current version of the metadata; this is incremented whenever the metadata schema changes
-VERSION = 26
+VERSION = 27
 
 
 def create_system_info(engine: sql.engine.Engine) -> None:
@@ -22,20 +22,25 @@ def create_system_info(engine: sql.engine.Engine) -> None:
         session.flush()
         session.commit()
 
+
 # conversion functions for upgrading the metadata schema from one version to the following
 # key: old schema version
 converter_cbs: dict[int, Callable[[sql.engine.Engine], None]] = {}
 
+
 def register_converter(version: int) -> Callable[[Callable[[sql.engine.Engine], None]], None]:
     def decorator(fn: Callable[[sql.engine.Engine], None]) -> None:
         global converter_cbs
         converter_cbs[version] = fn
+
     return decorator
 
+
 # load all converter modules
 for _, modname, _ in pkgutil.iter_modules([os.path.dirname(__file__) + '/converters']):
     importlib.import_module('pixeltable.metadata.converters.' + modname)
 
+
 def upgrade_md(engine: sql.engine.Engine) -> None:
     """Upgrade the metadata schema to the current version"""
     with orm.Session(engine) as session:
@@ -47,7 +52,9 @@ def upgrade_md(engine: sql.engine.Engine) -> None:
         while md_version < VERSION:
             if md_version not in converter_cbs:
                 raise RuntimeError(f'No metadata converter for version {md_version}')
-            print(f'Converting metadata from version {md_version} to {md_version + 1}')
+            from pixeltable.env import Env
+
+            Env.get().console_logger.info(f'Converting metadata from version {md_version} to {md_version + 1}')
             converter_cbs[md_version](engine)
             md_version += 1
         # update system info

pixeltable/metadata/converters/convert_10.py

@@ -1,12 +1,12 @@
 import sqlalchemy as sql
 
-from pixeltable.metadata.schema import Table, TableSchemaVersion
 from pixeltable.metadata import register_converter
+from pixeltable.metadata.schema import Table, TableSchemaVersion
 
 
 @register_converter(version=10)
 def _(engine: sql.engine.Engine) -> None:
-    default_table_attrs = {"comment": None, "num_retained_versions": 10}
+    default_table_attrs = {'comment': None, 'num_retained_versions': 10}
     with engine.begin() as conn:
         # Because `parameters` wasn't actually used for anything,
         # we can simply delete it without any data loss.

pixeltable/metadata/converters/convert_15.py

@@ -1,4 +1,3 @@
-
 import inspect
 import logging
 from typing import Any
@@ -37,8 +36,5 @@ def __update_md(orig_d: dict, binary_obj: bytes) -> Any:
         params.append(func.Parameter(name=name, col_type=col_type, kind=kind, default=default, is_batched=is_batched))
     is_batched = 'batch_size' in orig_d
     sig = func.Signature(return_type, params, is_batched=is_batched)
-    d = {
-        'signature': sig.as_dict(),
-        'batch_size': orig_d['batch_size'] if is_batched else None,
-    }
+    d = {'signature': sig.as_dict(), 'batch_size': orig_d['batch_size'] if is_batched else None}
     return d

pixeltable/metadata/converters/convert_16.py

@@ -1,4 +1,5 @@
 from uuid import UUID
+
 import sqlalchemy as sql
 
 from pixeltable.metadata import register_converter
@@ -7,10 +8,7 @@ from pixeltable.metadata.converters.util import convert_table_md
 
 @register_converter(version=16)
 def _(engine: sql.engine.Engine) -> None:
-    convert_table_md(
-        engine,
-        table_md_updater=__update_table_md
-    )
+    convert_table_md(engine, table_md_updater=__update_table_md)
 
 
 def __update_table_md(table_md: dict, table_id: UUID) -> None:

pixeltable/metadata/converters/convert_17.py

@@ -1,4 +1,5 @@
 from uuid import UUID
+
 import sqlalchemy as sql
 
 from pixeltable.metadata import register_converter
@@ -7,10 +8,7 @@ from pixeltable.metadata.converters.util import convert_table_md
 
 @register_converter(version=17)
 def _(engine: sql.engine.Engine) -> None:
-    convert_table_md(
-        engine,
-        table_md_updater=__update_table_md
-    )
+    convert_table_md(engine, table_md_updater=__update_table_md)
 
 
 def __update_table_md(table_md: dict, table_id: UUID) -> None:

pixeltable/metadata/converters/convert_18.py

@@ -1,4 +1,5 @@
 from typing import Any, Optional
+
 import sqlalchemy as sql
 
 from pixeltable.metadata import register_converter
@@ -7,10 +8,7 @@ from pixeltable.metadata.converters.util import convert_table_md
 
 @register_converter(version=18)
 def _(engine: sql.engine.Engine) -> None:
-    convert_table_md(
-        engine,
-        substitution_fn=__substitute_md
-    )
+    convert_table_md(engine, substitution_fn=__substitute_md)
 
 
 def __substitute_md(k: Optional[str], v: Any) -> Optional[tuple[Optional[str], Any]]: