zut.csv.CsvFormat

Bases: object

Format options for CSV files.

The default format is a balanced decision between compatibility with Excel and PostgreSQL bulk COPY operation: the delimiter is set to the same locale-dependent value as Excel and encoding is set to UTF-8 with BOM, so that Excel can open the file and display accents and columns automatically. All other format options are set as expected by PostgreSQL COPY operation. Thus, the written files can be directly bulk loaded into PostgreSQL by specifying the DELIMITER option of the COPY operation and by skipping the BOM characters in the loading function.

This default format may be overridden by specifing each format option individually, or using a modified set of predefined defaults with the fmt parameter (or CSV_FORMAT environment variable). Accepted values are: - pg (or postgres, postgresql): directly bulk loadable with PostgreSQL COPY operation without specifying the delimiter or skipping the BOM characters (sets encoding=utf-8 delimiter= ). - excel (or xlsx or xls): expand the quality of Excel displaying by modify CSV values themselves (sets decimal_separator=locale, lists=visual, dicts=visual, enums=name, no_microsecond=True, no_scientific_notation=True, tz=local). - visual: sets lists=visual, dicts=visual, enums=name. - json: sets lists=json, dicts=json.

__init__(fmt: CsvFormat | str | None = None, *, encoding: str | None = None, delimiter: str | Literal['locale'] | None = None, decimal_separator: str | Literal['locale'] | None = None, nullval: str | None = None, quotechar: str | None = None, newline: str | None = None, lists: Literal['pg', 'json', 'visual'] | None = None, dicts: Literal['json', 'visual'] | None = None, enums: Literal['value', 'name'] | None = None, no_microsecond: bool | None = None, no_scientific_notation: bool | None = None, tz: tzinfo | str | Literal['local', 'utc'] | None = None, default_formatter: Callable[[Any], Any] | None = None, default_parser: Callable[[str], Any] | None = None)

Methods

`__init__`([fmt, encoding, delimiter, ...])
`builtin_formatter`(value)
`builtin_parser`(value)
`escape_value`(value)
`examine`(-> ~zut.csv.ExamineResult)
`format_row`(row, *[, formatters])
`format_value`(value, *[, formatter])
`parse_row`(row, *[, parsers])
`parse_value`(value, *[, parser])

Attributes

`encoding`	The character encoding used for the CSV file.
`delimiter`	The character used to delimit columns.
`decimal_separator`	The character used as the decimal separator.
`nullval`	The character used to represent null values.
`quotechar`	The character used to quote values when necessary.
`newline`	The character used to separate lines.
`lists`	The format used to write lists.
`dicts`	The format used to write dicts.
`enums`	Whether to write enum values as their names.
`no_microsecond`	If set, omit datetime microseconds.
`no_scientific_notation`	If set, disable scientific notation for numbers.
`tz`	If set, convert aware datetimes to naive datetimes in the given timezone when writing, and back when reading.
`default_parser`	Default parser to use when reading values from CSV.
`default_formatter`	Default formatter to use when writing values to CSV.

decimal_separator: str: The character used as the decimal separator. Defaults to ..

default_formatter: Callable[[Any], Any]: Default formatter to use when writing values to CSV.

default_parser: Callable[[str], Any]: Default parser to use when reading values from CSV.

delimiter: str: The character used to delimit columns. Defaults to get_locale_csv_delimiter().

dicts: Literal['json', 'visual']: The format used to write dicts. Defaults to json.

encoding: str: The character encoding used for the CSV file. Defaults to utf-8-sig (UTF-8 with BOM).

enums: Literal['value', 'name']: Whether to write enum values as their names. Defaults to “value”.

lists: Literal['pg', 'json', 'visual']: The format used to write lists. Defaults to pg (PostgreSQL array format).

newline: str: The character used to separate lines. Defaults to n.

no_microsecond: bool: If set, omit datetime microseconds.

no_scientific_notation: bool: If set, disable scientific notation for numbers.

nullval: str: The character used to represent null values. Defaults to an empty string (distinguished from non-null values which are expressed as two quotechars).

quotechar: str: The character used to quote values when necessary. Defaults to “ (double quote).

tz: tzinfo | Literal['local', 'utc'] | None: If set, convert aware datetimes to naive datetimes in the given timezone when writing, and back when reading.