HP 3000 Manuals

MPE XL Native Language Programmer's Guide

ROMAN8 to EBCDIC Mapping 

In mapping from ROMAN8 to EBCDIC, characters look the same, or as close
as possible, before and after conversion.  The majority of the symbols
appearing in ROMAN8 also exist in the EBCDIC country extended code pages.
In ROMAN8, there are nine characters that have no similar EBCDIC
character and six undefined characters.  Since there are no undefined
characters in the EBCDIC country extended code pages, 15 characters in
EBCDIC have no look-alikes in ROMAN8.  For these characters a one-to-one
mapping has been defined as shown in Table C-1.

          Table C-1.  ROMAN8 to EBCDIC Mapping



The mapping to and from EBCDIC for the KANA8 character set is defined by
Japanese Industrial Standards (JIS) and IBM.

In all languages, the character mappings defined and implemented are any
character mapped from any Hewlett-Packard 8-bit character set to EBCDIC
and then back again, or vice versa, will result in the original character
value.  A complete listing of the Hewlett-Packard 8-bit character set to
EBCDIC mappings, and vice versa, can be obtained by running the NLUTIL
utility.  (Refer to the MPE XL System Utilities Reference Manual (32650-
90081).)

The mappings can be made available to a program by using NLINFO item=13
or 14.  The mappings are used by the NLTRANSLATE intrinsic, which
performs the Hewlett-Packard 8-bit to EBCDIC translation or the reverse.
The CTRANSLATE intrinsic maps USASCII or JISCII to EBCDIC (and vice
versa).  For the languages NATIVE-3000 and KATAKANA, there is no
difference between the mappings produced by NLTRANSLATE and CTRANSLATE.

MPE/iX 5.0 Documentation