x

The code page generation list file, install_hlq.ERRNLS.DATA(CPCODEPG), is a list of code pages to generate with the TSGU. You must change this file manually, using an editor, to contain the additional or alternate set of code page numbers you need. You copy the code page information from the list in the known code page file, install_hlq.ERRNLS.DATA(CPXCPTBL).

1. Review your code page configuration requirements.
   • If you modified your code page configuration in a FOCUS release prior to FOCUS 7.7, you can copy the CPCODEPG file, any CPnnnnn files that were modified, and the NLSCFG file from the FOCCTL.DATA library in the prior FOCUS release to the FOCUS 7.7 alias library named install_hlq.ERRNLS.DATA (physical library install_hlq.HOME.ERR).

     Once you have copied the files, continue with step 4.

   • If you did not previously modify your code page configuration but need to do so in FOCUS 7.7, continue to step 2.
2. Copy additional or alternate code pages into CPCODEPG from the known code page file, CPXCPTBL.

   Add the information for each additional or alternate code page on a separate line. Note that only the characters CP and the code page number (for example, CP00280) are required to generate the new transcoding tables. The maximum number of code page entries in the file is 16.

   1. Open the known code page file, member CPXCPTBL in the alias library install_hlq.ERRNLS.DATA.

      The following is a portion of the CPXCPTBL file that contains the code page information for the Italian code page 280 (CP00280):

      CP00280   E SBCS  IBM MF Italy   
      CP00281   E SBCS  IBM MF Japanese English    
      CP00284   E SBCS  IBM MF Spain/Latin America  
      CP00285   E SBCS  IBM MF United Kingdom                                 

      For a description of all of the fields in CPXCPTBL, see Structure of the Known Code Page File (CPXCPTBL).

   2. Copy the lines for the additional or alternate code pages you need from member CPXCPTBL to member CPCODEPG.

      For example, to add the Italian code page 280 to your configuration, copy the following line from CPXCPTBL to CPCODEPG:

      CP00280   E SBCS  IBM MF Italy                                          

      If you started with the default CPCODEPG FILE, the new version will contain the following information:

      CP00037   E SBCS  US IBM MF EBCDIC cod   
      CP00437   A SBCS  US PC ASCII code                 
      CP00137   A SBCS  ANSI Character Set for MS-Windows  
      CP01047   E SBCS  IBM MF Open Systems (Latin 1) 
      CP65001   A UTF8  Unicode (UTF-8)   
      CP00280   E SBCS  IBM MF Italy                                          

      Note: If the desired code page is not listed in the known code page file (CPXCPTBL), refer to the appropriate IBM Character Data Representation Architecture (CDRA) document and create your own, or contact your local Information Builders representative for information about additional code pages.

3. If you require customized monocasing or sorting for one or more of your configured code pages, edit the code page definition file for those code pages as described in Configuring Customized FOCUS NLS Monocasing and Configuring Customized FOCUS NLS Sort Sequences.
4. Add the allocation for the NLS DDNAME to the FOCUS 7.7 CLIST that was created during the ISETUP installation process, and start FOCUS.
   1. Add the allocation for DDNAME NLS to the FOCUSC CLIST.

      ISETUP creates a CLIST named FOCUSC in the install_hlq.CONF.DATA library. This CLIST allocates all of the required libraries for executing FOCUS using the high-level qualifier you chose during the installation process. In order to execute the TSGU, you must add an allocation for DDNAME NLS to the CLIST:

      ALLOC F(NLS)      DA('install_hlq.ERRNLS.DATA') SHR REU

   2. Execute the FOCUSC CLIST to start FOCUS.
5. At the FOCUS command prompt, issue the TSGU command to generate the necessary tables.

   The tables are created as members in the install_hlq.ERRNLS.DATA alias library.

   For example, to generate the transcoding table (TRANTBL) that adds the new or alternate code pages from the CPCODEPG file, issue the following command:

   TSGU

   If you modified the monocasing information in any of the code page definition files, issue the following command to generate the monocasing table (CASETBL):

   TSGU CASE

   If you modified the sorting sequence in any of the code page definition files, issue the following command to generate the sorting table (SORTTBL):

   TSGU SORT

   To create all three types of tables (transcoding, monocasing, and sort tables), issue the following command:

   TSGU GEN

   For complete information about TSGU command syntax, see Generating New Transcoding Tables.

6. Configure the FOCUS default code page and language.

   If you generated a new FOCUS code page in step 5 and want FOCUS to use this new code page, you must modify the NLS configuration file, install_hlq.ERRNLS.DATA(NLSCFG), to identify the new FOCUS code page and language used by FOCUS for error messages. This step is only necessary if you generated a new FOCUS code page in step 5.

   You change the FOCUS code page and language for FOCUS by editing the LANG setting in the NLS configuration file. Changing the LANG setting sets all other parameters to valid values. for complete information about parameters in the NLSCFG file, see The NLS Configuration File.

   Note: The LANG setting should match the three-character language abbreviation (language ID) or the language name.

   You can find the appropriate language abbreviation for each code page in member CPCPALL in the alias library install_hlq.ERRNLS.DATA. The following shows a portion of the CPCPALL file, which includes the language setting for code page 280:

   * LANG=ITA                                        
   cp280                                                        
   cp850                                                        
   cp137                                                        
   cp65001                                                      
   BR                                                           
   * LANG=JPN                                                   
   cp939                                                        
   cp942                                                        
   cp65001                                                      
   BR                                                                     

   The following example illustrates how you can change the default language from English to Italian. Change:

   LANG = AMENGLISH

   to

   LANG = ITA

   Error messages are only translated for German, Spanish, French, Dutch, and Japanese. Error messages for all other languages are translated to English.

   Note: Swedish and some other languages may only be partially translated.

x

CPnnnnn b dbcs-id description

where:

CP

Is the code page prefix (always CP).

nnnnn

Is the code page number.

b

Is the character type. Possible values are:

A for ASCII.

E for EBCDIC.

dbcs-id

Is the DBCS identifier (DBCSID).

description

Is a description of the code page.

x

Note: As part of the basic initialization, monocasing tables are provided for most of the common European languages. You will only need to customize the monocasing tables if you require a special monocasing configuration.

NLS monocasing involves language-sensitive (code page sensitive) uppercase and lowercase conversion. You can customize the attributes of each character by completing the following steps:

1. Edit the code page definition file (CPnnnnn), which is named by the code page number. For example, CP00037 contains the monocasing information for US English code page 37. For more information on the code page definition file (CPnnnnn), see Use the Known Code Page File (CPXCPTBL) for z/OS.

   Note: You can reference the known code page file (CPXCPTBL) to find the name of the code page definition file.

   The code page definition file (CPnnnnn) contains the Code Point and Graphic Character Global Identifier (GCGID). Make the appropriate changes to GCGID uppercase in the third column and Character type in the fifth column.

   The following is a chart of a sample code page definition file layout:

   

2. Edit the code page generation list file (CPCODEPG) and add the code page definition information as described in Configuring New or Alternate Code Pages for FOCUS. The updated code page generation list (CPCODEPG) is used to regenerate the custom monocasing table file (CASETBL).
3. Execute the TSGU with the parameter CASE.

   The TSGU generates the updated NLS monocasing table file install_hlq.ERRNLS.DATA (CASETBL).

Note: For additional information on modifying monocasing values in the code page definition file, refer to the IBM CDRA Library SC09-9391-00 Level 1 Registry or contact your local Information Builders representative.

x

Note: As part of the basic initialization, sorting tables are provided for most of the common European languages. You will only need to customize the sorting tables if you require a special sorting sequence.

If you want to use a weighted sort that accounts for characters that are out of binary sequence, you can customize the sort tables by completing the following steps.

1. Edit the code page definition file install_hlq.ERRNLS.DATA(CPnnnnn), which is named by the code page number. For example, CP00037 contains the sorting information for US English code page 37. For more information on the code page definition file (CPnnnnn), see Use the Code Page Definition File (CPnnnnn) for z/OS.

   Note: You can reference the known code page file (CPXCPTBL) to find the name of the code page definition file.

   The code page definition file (CPnnnnn) contains the Code Point and Graphic Character Global Identifier (GCGID). Make the appropriate changes to the Sort weight listed in the fourth column. The following is a chart of a sample code page definition file layout:

   

2. Edit the code page generation list file (CPCODEPG) and add the code page definition information as described in Configuring New or Alternate Code Pages for FOCUS. The updated code page generation list (CPCODEPG) is used to regenerate the custom sorting table file (SORTTBL).
3. Execute the TSGU from FOCUS with the parameter SORT.

   The TSGU generates the updated NLS sorting table file install_hlq.ERRNLS.DATA (SORTTBL).

Note: For additional information on modifying monocasing values in the code page definition file, refer to the IBM CDRA Library SC09-9391-00 Level 1 Registry or contact your local Information Builders representative.

x

The code page definition file install_hlq.ERRNLA.DATA(CPnnnnn) contains information on the characters for each code point value in the code page.

dd aaaaaaaa aaaaaaaa xx h  

where:

dd

Is the hexadecimal code point value (00 through FF).

aaaaaaaa

Is the Graphic Character Global IDentifier (GCGID).

xx

Is the Sort weight. Consists of a hexadecimal code point value (00 through FF).

h

Is the character type. Possible values are:

L for Lower-case alphabet.

U for Upper-case alphabet.

A for Asian (non-alphabet) character.

D for Digit.

S for Special character.

C for Control (non-printable) character.

x

TSGU
TSGU GEN
TSGU INFO info-command 
TSGU CASE
TSGU SORT

where:

no parameters

Generates the NLS transcoding table(TRANTBL).

GEN

Generates all the tables (TRANTBL, CASETBL and SORTTBL).

INFO

Displays information about the current NLS configuration.

CASE

Generates the custom monocasing table (CASETBL) binary file.

SORT

Generates the custom sort sequence table (SORTTBL) binary file.

x

TSGU INFO CP [DBCS] [nnn [nnn [...]]] [MAT [DES]]
TSGU INFO CP [SBCS] [nnn [nnn [...]]] [MAT [DES]]
TSGU INFO TRAN [idx [idx [...]]] [M] [V]
TSGU INFO CASE [nnn [nnn [...]]] [M] [V]
TSGU INFO SORT [nnn [nnn [...]]] [M] [V]
TSGU INFO SET

where:

CP

Lists transcoding tables.

DBCS

Lists only DBCS transcodings.

SBCS

Lists only SBCS transcodings.

nnn

Code page which only shows given code page transcodings (no leading zeros necessary).

idx

Index which only shows given index transcodings.

MAT

Creates a matrix report for transcoding tables.

DES

Reverses an order of code pages in a matrix report.

TRAN

Shows transcoding table contents (trantbl.err).

CASE

Shows uppercase/lowercase table contents (casetbl.err).

SORT

Shows sort table contents (sorttbl.err).

M

Masks null transcode values in tables.

V

Shows tables in vertical layout.

SET

Shows the list of available languages.

x

To see the online help information for the TSGU command, issue the following command:

TSGU ?

The output is:

TSGU  : GEN   : cpcodepg-id ::                                       
      : TRAN  : cpcodepg-id :: (default)                             
      : CASE  : cpcodepg-id ::                                       
      : SORT  : cpcodepg-id ::                                       
      : INFO  : info-command ::                                      
      : XLAT  : infile  : outfile  :::                               
      : XLAT  TEMPLATE : cp1 cp2 ::                                  
      : KTBL  : ktbl-command : : xxxx : xxxx : ... ::::              
      : ERR   err-command :                                          

 GEN  : generate all necessary tables (binary files)                 
 TRAN : generate a TRANTBL and KTBL binary files                     
 CASE : generate a CASETBL binary file                               
 SORT : generate a SORTTBL binary file                               
 INFO : display info of the current Translation Services (? LANG)    
XLAT : convert a file by specified Code Pages in config file (tsgu.nls) 
       TEMPLATE shows a sample batch file/unix shell script for multi-  
       files to be XLATed. (tsgult only)                                
KTBL : generate individual KTBL binary files                            
ERR  : convert EDAHOME/nls/*.err file (UNIX and Windows NT only)        
xxxx : DBCS translation table identifier                               

cpcodepg-id:                                                            
    Max 4 characters. To use customized and saved CPCODEPG.NLS          
    If specified, a file named 'cpcp:cpcodepg-id:.nls' is used instead. 
    cpcpall.nls and cpcpucs.nls are provided as samples.               

info-command:                                                           
    CP   :| DBCS |: : nnn | nnn | ... : : MAT : DES ::                  
          | SBCS |                                                      
    TRAN : idx : idx : ... :: : M : : V ::                              
    CASE : nnn : nnn : ... :: : M : : V ::                              
    SORT : nnn : nnn : ... :: : M : : V ::                              
    ALL                                                                 
                                                                        
    CP   : Show TRTIDX                                                  
    DBCS : Show only DBCS translations                                  
    SBCS : Show only SBCS translations                                  
      idx  : Index, show only given indexes translations              
      nnn  : Code Page, show only given Code Pages translations       
      MAT  : Make a matrix report for TRTIDX                          
      DES  : Reverses an order of Code Pages in a matrix report       
      TRAN : Show translation tables (TRANTBL)                        
      CASE : Show upcase/lowcase tables (CASETBL)                     
      SORT : Show sort tables (SORTTBL)                               
      M    : Mask null translate values in tables                     
      V    : Show tables vertical layout against horizontal           
      ALL  : Show intlcm info for debugging                          

ktbl-command:                                                       
      CHK                                                             
      FMT                                                             
      CNV : UDC : : nn | CPU | MEM :                                  
      RVS : UDC : : nn | CPU | MEM :                                  
      LST : FFFF :                                                    
      ADD                                                             
                                                                      
      CHK : Check the KTBLs translation integrity                     
      FMT : Format the KTBLs                                          
      CNV : Convert the KTBLs                                         
      RVS : Make a reverse translation table for the KTBLs            
     LST : List the DBCS translation                              
     ADD : Make a new DBCS translation from two KTBLs             
     UDC : Remove Unused UDC code translation                     
     nn  : Threshold level for indexing the ranges                
     CPU : nn = 65536, to make only one index range               
     MEM : nn = 4, to make the smallest size format for the KTBLs 
     FFFF: Produce entries for source code even they are translated   
           to invalid codes                                       

err-command:   
     GEN  aaa 
     UCS  aaa 
                                                                  
     GEN : Create *:aaa:.err files in the directory EDACONF/etc.   
           (JPE and GE5 only)   
     UCS : Create *(aaa).err files in Unicode (utf-8) into the  
           directory EDACONF/etc.   
     aaa : Language abbreviation 

x

The following is a reference of keywords in the NLSCFG file and their associated values:

LANG=

Is the language. The list of known languages file (LANGTBL), located in install_hlq.ERRNLS.DATA, contains a list of valid values. The value of LANG determines the defaults for other parameters in FOCUS. The default value is AMENGLISH (American English).

CODE_PAGE=

Is the FOCUS code page. For z/OS, the hard-coded default value is 37.

CDN=

Is Continental Decimal Notation. Possible values are a comma (,) or a period (.). The default value is a period (.).

DATETIME=

Is the DATE/TIME format for SQL. Possible values are:

EDA for EDA type (default). On z/OS, the default is HYYMDs.

ISO for yyyy-mm-dd hh.mm.ss.

USA for mm/dd/yyyy hh:mm AM or PM.

EUR for dd.mm.yyyy hh.mm.ss.

JIS for yyyy-mm-dd hh:mm:ss.

LOCAL for any site-defined form.