pointcloudset.io.pointcloud.delimited module

pointcloudset.io.pointcloud.delimited.ensure_coordinate_columns(df: DataFrame, file_path: Path, format_name: str) → DataFrame

Ensure lowercase coordinate columns exist, inferring for headerless data.

If x, y, z are missing but at least three columns exist, the first three columns are renamed to x, y, z and remaining columns are named property_1, property_2, and so on.

Parameters:

  • df – Parsed tabular data.

  • file_path – Source file path used in warning/error messages.

  • format_name – Human-readable format name (for example, CSV or XYZ).

Returns:

A dataframe containing lowercase coordinate columns.

Raises:

ValueError – If fewer than three columns are available.

pointcloudset.io.pointcloud.delimited.read_delimited_coordinates(file_path: Path | str, *, format_name: str, default_sep: str | None, fallback_sep: str | None = None, normalize_xyz: bool = False, **kwargs) → DataFrame

Read delimited coordinate files with consistent coordinate handling.

The function first attempts to parse with normal header inference, then falls back to headerless parsing when needed. Coordinate handling is unified via _normalize_xyz_columns.

Parameters:

  • file_path – Path to the input file.

  • format_name – Human-readable format name.

  • default_sep – Default delimiter passed to pandas.read_csv.

  • fallback_sep – Optional fallback delimiter used when default parsing fails.

  • normalize_xyz – Whether to convert uppercase X/Y/Z headers to lowercase x/y/z.

  • **kwargs – Additional keyword arguments forwarded to pandas.read_csv.

Returns:

Parsed dataframe with valid coordinate columns.

Raises:

ValueError – If parsing fails, coordinates are invalid, or uppercase coordinates are provided without enabling normalization.

pointcloudset.io.pointcloud.delimited.write_delimited_coordinates(pointcloud: PointCloud, file_path: Path, *, header: bool, sep: str, columns: list[str] | None = None) → None

Write pointcloud tabular data to a delimited text file.

Parameters:

  • pointcloud – PointCloud instance to serialize.

  • file_path – Destination file path.

  • header – Whether to write a header row.

  • sep – Column separator.

  • columns – Optional column subset to write; writes all columns when omitted.