csvpy 命令

常用示例

说明

csvpy is part of csvkit that loads a CSV file into an interactive Python shell for exploration and analysis. If IPython is installed, it is used as the shell for a richer experience. In default mode, the CSV data is loaded into a variable named reader, a csv.reader object. With --dict, it becomes a DictReader where each row is a dictionary. With --agate, it becomes a table variable with full analysis capabilities including sorting, filtering, and aggregation. This tool is useful for quick data exploration, testing data transformations, and prototyping analysis code. The interactive environment allows immediate feedback while working with the data.

参数

-d _CHAR_, --delimiter _CHAR_

Field delimiter character (default: comma).

-t, --tabs

Use tab characters as the field delimiter.

-q _CHAR_, --quotechar _CHAR_

Character used for quoting fields.

-e _ENCODING_, --encoding _ENCODING_

Specify the input file encoding (e.g., utf-8, latin1).

-H, --no-header-row

Indicate that the file has no header row.

-K _N_, --skip-lines _N_

Skip the first N lines of the input before processing.

--agate

Load the file as an agate Table instead of a csv.reader object. The data is available as a variable named table.

--dict

Load as a csv.DictReader where each row is a dictionary keyed by column headers.

-I, --no-inference

Disable type inference when used with --agate, keeping all values as strings.

-S

Disable CSV dialect sniffing.

--blanks

Do not convert empty strings to None.

--null-value _VALUES_

Convert specified values to None (can specify multiple).

-y _N_, --sniff-limit _N_

Limit the number of bytes used for CSV dialect sniffing.

-l, --linenumbers

Insert a column of line numbers at the beginning of the output.

-z _N_, --field-size-limit _N_

Maximum length of a single field in the input CSV.

-V, --version

Display version information.