|
<< Click to Display Table of Contents >> Character set encoding |
  
|
|
<< Click to Display Table of Contents >> Character set encoding |
|
When exporting data to text format (.csv, .txt or .tab) character coding for characters in the extended ASCII character set above character code 127 (often mislabelled ANSI characters) may be of either the current standard Windows System Settings (Windows Default, in Western Countries = Windows 1252) or by specific coding as Windows 1252 Western, UTF-8 or UNICODE.
See also the Character codes external link.
Besides optionally exporting non-coded simple text, where import character interpretation will depend on the current system default character set, PdxEditor may also export using the universal Unicode UTF-8 encoding to overcome the usual challenge interpreting high bit characters correctly according to the original data - i.e. reading character codes > ASCII #127 correctly independent of code page settings of the currently local computer.
When exporting as UTF-8 coded text, the file will be Byte Order Marked (BOM) to ensure proper decoding when imported by other applications.
If exporting targeting e.g. a MySQL database, you may deselect the use of BOM when exporting UTF-8 encoded text files, as e.g. MySQL is not import compliant with BOM encoded UTF-8 files.
See also BDE language drivers.
The Byte Order Mark as an ISO standard is specified by the first three bytes of the text file telling that the receiving application should interpret text by the UTF-8 encoding.
With older text handling software imported text may present with the leading characters of  (on PCs with character set 1252) indicating the application is not fully Unicode compliant (or will only decode correctly by manual selection), as these first three characters (Byte Order Mark; binary: EF BB BF) actually should have been read as information about the UTF-8 encoding of the text document and not as literal characters.
Some applications not depending on the BOM will however decode UTF-8 encoded texts based on content analysis, though this method may be prone to be unreliable on short volumes of text.