PH32621: CURRENTLY, WHEN DSNHDECP SCCSID=290(KATAKANA), SYSTEM SPS GENERAL XML OUTPUT CONTAIN THE X'1A' CHARACTER.

Obtain the fix for this APAR.

APAR status

• Closed as unreproducible in next release.

Error description

• Currently, when DSNHDECP SCCSID=290 (katakana),
  GET_CONFIG,GET_SYSTEM_INFO, and GET_MESSAGE generate XML output
  documents which contain the x'1A' character when the requested
  information contain lower case characters.
  

Local fix

• N/A
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * All Db2 12 for z/OS users of the following                   *
  * Db2-supplied stored procedures:                              *
  * - SYSPROC.GET_CONFIG                                         *
  * - SYSPROC.GET_MESSAGE                                        *
  * - SYSPROC.GET_SYSTEM_INFO                                    *
  * - SYSPROC.SET_PLAN_HINT                                      *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * When DSNHDECP SCCSID = 290                                   *
  * (Katakana), GET_CONFIG,                                      *
  * GET_SYSTEM_INFO, GET_MESSAGE, and                            *
  * SET_PLAN_HINT cannot complete                                *
  * successfully and the xml_message                             *
  * document they returned contains the                          *
  * substitution character x'1A'.                                *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  With EBCDIC single byte CCSIDs, the character code points for
  the a-z, A-z and 0-9 characters are the same across all the
  EBCDIC CCSIDs, except for CCSID 290 (Katakana). With
  Katakana CCSID 290, the lower case characters (a,b,c,d,...,z)
  have different character code point values than all of the
  other EBCDIC CCSIDs. The hex code point definitions
  for lower case characters are different from the "Latin-1"
  based EBCDIC codepages. For example:
    c = x'83' for all EBCDIC Latin-1 based CCSIDs
    c = x'64' for Katakana EBCDIC CCSID (CCSID 290)
  When these stored procedures convert a source string from
  EBCDIC to Unicode, and set the conversion source CCSID to
  the DSNHDECP SCCSID 290, although the source string contains
  lower case characters that are in CCSID 37/500/1047, the
  lowercase character code points(e.g. x'A5', x'94', x'85', ...)
  do not translate correctly.  In addition, there are actually
  some valid Katakana character code points within the
  lowercase EBCDIC 37/1047/500 range, which can't be converted
  to single byte Unicode (CCSID 367), causing the conversion
  routine to substitute each lower case character with x'1A'
  in the target string.
  

Problem conclusion

Temporary fix

Comments

• This APAR provides limited support for Katakana CCSID 290
  lower case characters for the following stored procedures:
  SYSPROC.GET_CONFIG, SYSPROC.GET_SYSTEM_INFO,
  SYSPROC.GET_MESSAGE, and SYSPROC.SET_PLAN_HINT.
  These stored procedures will continue to support SBCS CCSID
  only, but not MIXED CCSID. In addition, any user-specific
  information specified in the xml_input document must contain
  only valid EBCDIC CCSID 37 characters, and these EBCDIC
  characters must exist in Unicode CCSID 367. For example,
  this requirement applies (but not limited) to content in the
  string inputs for the following items:
   - GET_CONFIG:
      - DB2 Data Sharing Group Members
   - GET_SYSTEM_INFO:
      - SMPCSI Data Set
      - SYSMOD
   - GET_MESSAGE:
      - SQLCODE
      - Message Tokens
   - SET_PLAN_HINT:
      - PLAN_SCHEMA
  Likewise, any requested information, which will be converted
  to Unicode CCSID 367, and written to the xml_output document,
  must also contain only valid EBCDIC CCSID 37 characters and
  these EBCDIC characters must exist in Unicode CCSID 367.
  

APAR Information

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

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

  UI73360

Modules/Macros

• DSNADMGV DSNADMGU DSN@DMGV DSN@DMGU
  

Fix information

• Fixed component name

  DB2 OS/390 & Z/

• Fixed component ID

  5740XYR00

Applicable component levels

• RC10 PSY UI73360

     UP21/01/15 P F101 ¢

Fix is available

• Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.