Supported CSV features

Temetra Analysis recognizes and supports the following elements within CSV import files:

  • Column separator. Supports the auto-detection of the following separator characters (in recognition order):

    • \t (tabulator)

    • | (vertical bar/pipe)

    • ; (semicolon)

    • , (comma)

    Note: The tabulator character string (\t) is the default separator if auto-recognition fails. A single file must use the same separator throughout the file.

  • Header. Auto-recognition of column headers is supported. Each header is case sensitive. Extra punctuation characters within header names are discarded, such as white space, periods (.), underscores (_), and hyphens (-). If a header is present and recognized, the following options are supported: 

    • omitting optional columns

    • custom column ordering

    If any header is omitted, the file must use the default column order, all columns must be present within the file, and all records must contain the same number of columns as specified in the default file format.

  • Empty lines or files. Empty or white-space-only lines are automatically discarded. Completely empty or header-only CSV files are supported.

  • Double quotation mark ("). This is the only supported quote/escape method. Single quotation marks (') are not supported (per RFC 4180 specifications).

    Note: Quoting is optional. It is required only if a field contains a separator character.

  • Auto-detect CSV file encoding. Supports localized, culture-specific characters for properly encoded CSV files.

  • Empty/null values. For null values, simply leave the field empty. Do not use special null markers (such as NULL) for empty fields, unless a field is explicitly defined as resettable.

  • Rules. Use the following best practices for file formatting:

    • If possible, use UTF-8 encoded files with byte order mark (BOM) to ensure special character support and to significantly boost encoding detection speed.

    • Use the tabulator (\t) as the separator character:

      • Typically, no quoting is needed as fields rarely contain tabulators as data.
      • The file is human readable.
      • Excel can open tab-separated CSV files with any locale setting (as opposed to semicolon- or comma-separated CSV files).

    • Always use column headers.

    • Always include all defined columns in the data files.