IBM Support

HD25700: INCORRECT ENCODING OF NON-ASCII CHARACTERS IN P21 FILES CREATED BY LCA

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • DETAILED DESCRIPTION:
    The STEP part 21 files created by Enovia LCA do
    not encode non-ascii
    characters correctly.
    I have only tested using swedish special characters
    ᄄ,ᄄ and ᄄ, but the
    error is likely to affect other characters as well.
    Enovia LCA uses the format '\X2\hh\X0\' where hh
    is a one-octet
    hexadecimal number. This is wrong. The standard
    says '\X2\hhhh\X0\',
    where hhhh is a two-octet hexadecimal number.
    There is also a discrepancy between the P21 writer
    and the P21 reader of
    Enovia LCA. The reader requires the format '\X2
    \hhhh\X0\' which means
    that the IEnovIn.sh (VPMImporCDM) program can
    not read its own temporary
    files.
    If you do not wish to use the ISO-10646 character
    set, use the '\X\hh'
    format to encode ISO-8859-1 as described in
    chapter 6.3.3.3 of
    ISO-10303- 21.
    Chapter 6.3.3.2 "Encoding the character sets of
    ISO 10646 within a
    string" from ISO standard 10303-21:
    The control directive reverse solidus capital letter X
    digit two reverse
    solidus "\X2\"shall be used to indicate that the
    following sequence of
    multiples of four hexadecimal characters shall be
    interpreted as
    encoding the two-octet representation of characters
    from the BMP in ISO
    10646.
    The encoding in a string in the exchange structure
    shall be as follows:
    ? each character in the representation of 10646 to
    be encoded shall be
    converted to two 8-bit bytes as specified in ISO
    10646;
    ? each of the two resulting 8-bit bytes shall be
    encoded as two
    hexadecimal characters in the basic alphabet
    corresponding to the
    graphic representation of the hexadecimal digit.
    Example:
    Original text, in Swedish: 'EMBALLAGE/RACK
    LᄄDA'.
    Encoded according to ch 6.3.3.1 of ISO-10303-
    21: 'EMBALLAGE/RACK
    L\S\EDA'.
    Incorrect encoding by Enovia
    LCA: 'EMBALLAGE/RACK L\X2\c5\X0\DA'.
    Correctly encoded according to ch 6.3.3.2 of ISO-
    10303-21:
    'EMBALLAGE/RACK
    L\X2\00c5\X0\DA'.
    .
    Step by step scenario:
    None - see above full description.
    .
    

Local fix

Problem summary

  • Incorrect encoding of non-ascii characters in P21             f
    iles created by LCA
    DETAILED DESCRIPTION:
    The STEP part 21 files created by Enovia LCA do
    not encode non-ascii
    characters correctly.
    I have only tested using swedish special characters
     ,  and  , but the
    error is likely to affect other characters as well.
    Enovia LCA uses the format '\X2\hh\X0\' where hh
    is a one-octet
    hexadecimal number. This is wrong. The standard
    says '\X2\hhhh\X0\',
    where hhhh is a two-octet hexadecimal number.
    There is also a discrepancy between the P21 writer
    and the P21 reader of
    Enovia LCA. The reader requires the format '\X2
    \hhhh\X0\' which means
    that the IEnovIn.sh (VPMImporCDM) program can
    not read its own temporary
    files.
    If you do not wish to use the ISO-10646 character
    set, use the '\X\hh'
    format to encode ISO-8859-1 as described in
    chapter 6.3.3.3 of
    ISO-10303- 21.
    Chapter 6.3.3.2 "Encoding the character sets of
    ISO 10646 within a
    string" from ISO standard 10303-21:
    The control directive reverse solidus capital letter X
    digit two reverse
    solidus "\X2\"shall be used to indicate that the
    following sequence of
    multiples of four hexadecimal characters shall be
    interpreted as
    encoding the two-octet representation of characters
    from the BMP in ISO
    10646.
    The encoding in a string in the exchange structure
    shall be as follows:
    ? each character in the representation of 10646 to
    be encoded shall be
    converted to two 8-bit bytes as specified in ISO
    10646;
    ? each of the two resulting 8-bit bytes shall be
    encoded as two
    hexadecimal characters in the basic alphabet
    corresponding to the
    graphic representation of the hexadecimal digit.
    Example:
    Original text, in Swedish: 'EMBALLAGE/RACK
    L DA'.
    Encoded according to ch 6.3.3.1 of ISO-10303-
    21: 'EMBALLAGE/RACK
    L\S\EDA'.
    Incorrect encoding by Enovia
    LCA: 'EMBALLAGE/RACK L\X2\c5\X0\DA'.
    Correctly encoded according to ch 6.3.3.2 of ISO-
    10303-21:
    'EMBALLAGE/RACK
    L\X2\00c5\X0\DA'.
    .
    Step by step scenario:
    None - see above full description.
    .
    

Problem conclusion

  • THIS PROBLEM WILL BE FIXED ON ENOVIA
    VERSION 5 RELEASE 15 GA level.
    NOTE THAT THIS PROBLEM WILL ALSO BE FIXED
    ON V5R13SP08 AND V5R14SP03.
    .
    

Temporary fix

Comments

APAR Information

  • APAR number

    HD25700

  • Reported component name

    ENOVIA LCA AIX

  • Reported component ID

    569101700

  • Reported release

    512

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2004-07-27

  • Closed date

    2004-11-04

  • Last modified date

    2009-06-24

  • APAR is sysrouted FROM one or more of the following:

  • APAR is sysrouted TO one or more of the following:

Fix information

  • Fixed component name

    ENOVIA LCA AIX

  • Fixed component ID

    569101700

Applicable component levels

  • R513 PSN SP51308

       UP04/12/29 N 1000

  • R514 PSN SP51403

       UP04/12/02 I 1000

[{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSDJRN","label":"ENOVIA LCA"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"512","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
24 June 2009