csvclean 命令

常用示例

说明

csvclean is part of csvkit that validates and cleans CSV files. It detects common problems like inconsistent column counts, empty columns, and encoding issues. Since csvkit 2.0, csvclean no longer reports or fixes errors by default. You must explicitly enable checks (such as --length-mismatch or --empty-columns) or fixes (such as --join-short-rows or --fill-short-rows). Output is written to standard output and errors to standard error. The tool handles various CSV dialects and can work with files using different delimiters, quote characters, and encodings. It is essential for preprocessing messy data before analysis.

参数

--length-mismatch

Report rows that are shorter or longer than the header row.

--empty-columns

Report empty columns as errors.

-a, --enable-all-checks

Enable all error reporting checks.

--join-short-rows

Merge consecutive short rows into a single row.

--separator _SEPARATOR_

String used to join short rows (default: newline).

--fill-short-rows

Fill short rows with missing values.

--fillvalue _VALUE_

Value used to fill short rows (default: empty string).

--omit-error-rows

Exclude rows containing errors from standard output.

--label _LABEL_

Add a label column to error output for automated workflows.

--header-normalize-space

Strip leading/trailing whitespace and normalize whitespace in headers.

-d _CHAR_, --delimiter _CHAR_

Field delimiter (default: comma).

-t, --tabs

Use tabs as delimiter.

-q _CHAR_, --quotechar _CHAR_

Quote character (default: double quote).

-p _CHAR_, --escapechar _CHAR_

Escape character for the delimiter or quote character.

-e _ENCODING_, --encoding _ENCODING_

Input file encoding.

-S, --no-header-row

File has no header row.

-H

Omit the header row from output.

-K _N_, --skip-lines _N_

Skip the first N lines of the input file.

-v

Verbose error output.