Converting 8-bit characters

In an UDO source file you can use higher characters without having to know how a character has to look like in a destination format like LaTeX or Windows Help. So you can enter a German ß or ä without any fear, UDO converts it for you automatically.

UDO expects files containing chars of the system charset of your operating system. If you run UDO on a MS-DOS computer UDO expects text files that are written with the IBM PC character set by default. If UDO runs on an Atari computer UDO will expect the TOS character set by default.

But UDO can manage file that are written with another character set, too. You have simply to tell UDO which character set your source file uses with !code_source [<charset>].

Below is an overview of the character sets UDO knows about:

UDO supports various codepages for various systems. Below you see a list of all currently supported systems and codepages, some of which with multiple descriptors for the same codepage. It doesn't matter if you use these descriptors upper- or lowercase. (The descriptors base on the former UDO descriptors and on those supported by the Unix command iconv.)

Important: If you have used latin1 in your old UDO documents, you should switch it to e.g. cp1252 because UDO used to assign Windows codepage 1252 to it before version 7 which correctly assigns ISO-8859-1 to it!

When you use so-called 1-byte codepages (all codepages supported by UDO, except Unicode) and use one codepage for your UDO documents, but a different one for your output documents, you might want to keep in mind that all codepages have different settings. A codepage is a collection of 256 characters from the whole range of all characters which have been defined in the Unicode standard already.

Imagine you have created an UDO document using the DOS encoding and use DOS graphic signs, but your target format is e.g. Apple MacRoman. Then you will not be able to see your DOS graphic signs. When you have used the Hebrew letters from the Atari TOS encoding, you will not be lucky to see them in most other codepages.

In these cases we recommend to use UTF-8, if it is available for the target format. Internally, UDO keeps all codepages in Unicode format so you will be able to use e.g. the Hebrew Alef from the TOS character set and see it properly even in UTF-8 and Windows codepage 1255.

Convert multiple files to Unicode

If you want to convert older project files from a 1-byte codepage to UTF-8 but don't have fun doing this conversion yourself for each single file, chances are that the Unix command iconv helps much. It can usually be found on Unix machines and on Mac OS X.
Here is a simple example how to convert any number of files with the suffix *.cs recursively (i.d. in any number of subfolders) at once from Czech (e.g. coded in ISO-8859-2) to UTF-8, using the bash shell in the Terminal application on Mac OS X. Regard the special apostrophes which enclose the find command!

for x in `find . -name '*.cs'`; do iconv -f ISO-8859-2 -t UTF-8 $x > "$x.utf8"; rm $x; mv "$x.utf8" $x; done

The encoding conversion cannot be done directly in the same file because it would be empty afterwards; thus we need the temporary *.utf8 files which are renamed with the original files names after the original files have been deleted.