classfiltercsv.markdown

layout	title	published
default	classfiltercsv	true

[%CFEngine_function_prototype(filename, has_header, class_column, optional_sort_column)%]

Description: Parses CSV data from an RFC 4180 compliant file filename (CRLF line endings required), and returns a data variable that is filtered by defined classes. If has_header is set to true, the columns in the first line of the CSV file are used as keys for the data. class_column specifies which column contains class names to filter by.

If optional_sort_column is defined, the data containers will be sorted by the given column. Both class_column and optional_sort_column must be integer indices starting from 0, and must be at most the total amount of columns minus 1.

[%CFEngine_function_attributes(filename, has_header, class_column, optional_sort_column)%]

Example:

Prepare CSV:

[%CFEngine_include_snippet(classfiltercsv.cf, #+begin_src prep, .*end_src)%]

Policy:

[%CFEngine_include_snippet(classfiltercsv.cf, #+begin_src cfengine3, .*end_src)%]

Output:

[%CFEngine_include_snippet(classfiltercsv.cf, #+begin_src\s+example_output\s*, .*end_src)%]

Notes:

• If the CSV file is stored in a git repository the .gitattributes file can be used to ensure proper line endings.

  For example:

  # .gitattribtues
  *.csv     text eol=crlf
  RFC-4180-non-compliant-line-endings.csv eol=lf
  *.mustache text
  *.sh       text eol=lf
  

See also: [data_expand()][data_expand], [readcsv()][readcsv], [classmatch()][classmatch]

History:

• Introduced in CFEngine 3.14