Comma-separated values

In computing, a comma-separated values (CSV) file stores information you might find in a spreadsheet, such as a table of text, numbers, and dates, in a plain text file.

Format

Typically, the first line in a plain text file would contain header information, which are labels for table information organized vertically. Starting with the second line, each line would represent a row. Each cell in a row is separated by a comma, hence its name.

The following is a simple example of this type of file. The example has three columns (labeled 'name', 'age', and 'favorite color'), and four rows (including the header line at the top).<syntaxhighlight lang="text"> name,age,favorite color Alice,12,green Bob,23,purple Jane,34,blue </syntaxhighlight>

Escaping

Text in a cell may be surrounded by quotation character ("), to note the start and end of a cell (a particular column and row location) in a table. Quotation characters in a cell value in turn can be denoted by doubling the quotation character (""). This is primarily used to signify that a particular cell contains data that may be confusing to a computer that's trying to read the file. A common example of this is when a cell contains a comma as part of it's data. Without the quotation marks, it's unclear if the comma denotes a separation between cells:<syntaxhighlight lang="text"> name,age Smith, Joe,45 Choi, Ceri,56 Plummer, Olive,23 </syntaxhighlight>or if it should be part of the cell's data:<syntaxhighlight lang="text"> "name","age" "Smith, Joe","45" "Choi, Ceri","56" "Plummer, Olive","23" </syntaxhighlight>

Compatibility

CSV is a way to describe table data. However, various applications may require different types of tables. Having a file in CSV format may not be enough for a file to be compatible with a particular application.

Furthermore, CSV is stored in plain text. An application may have problems reading a CSV file because of incompatibility with the plain text format, either because of the character encoding or line break format.