Neoview Character Sets Administrator's Guide (R2.3)

Table C-1 How Pass-Through Mode and UTF16 Conversion are Implemented From the Neoview
Transporter Java Client
Summary of Character
Translation BehaviorsAdditional Guidelines
How to Enable and Disable
UTF16 Conversion for Java
Strings
How Pass-Through Mode
is Enabled and Disabled
for the Java Client
When the Java Client is enabled
for pass-through mode, the Java
Client loads and extracts user
data in its client locale character
encoding. When pass-through
mode is enabled, the Java Client
does not perform UTF16
translation for internal use.
When pass-through mode is
disabled, the Java Client first
converts all incoming character
data to UTF16, the native Java
String format used internally.
• The Java Client then converts
user data bound for client
locales from UTF16 to the
specified client locale
character encoding. If the
encoding is specified in the
Java Client's control file, that
encoding is used. If not, the
encoding defaults to the
character set that is currently
configured for the target
client locale.
• The Java Client converts user
data bound for ISO88591
columns in the Neoview
database from UTF16 to the
encoding required for the
database. For example, user
data for ISO88591 columns is
encoded in ISO8859-1 for the
ISO88591 configuration, SJIS
for the SJIS configuration, and
UTF8 for the Unicode
configuration. The Java Client
always preserves
UTF16-encoded user data
bound for UCS2 columns in
the UTF16 format.
• In the ISO88591
configuration, the
NVT.pass-through-mode
system property must
be set to TRUE to
disable UTF16
conversion.
• In the SJIS or Unicode
configuration, the
NVT.pass-through-mode
system property must
be set to FALSE to
enable UTF16
conversion. If you set it
to TRUE in the SJIS or
Unicode configuration,
the Neoview Client
generates a warning to
the log file and resets
the system property to
FALSE.
• If no value is specified
for the
NVT.pass-through-mode
system property, it
defaults to FALSE.
Before they run the Java
Client, users can edit the
properties stored in the
/conf/nvt.properties
file in the installation
directory. Users can edit
the
NVT.pass-through-mode
system property value to
enable or disable UTF16
conversion within the Java
Client:
NOTE: The specified
NVT.pass-through-mode
system property overrides
any encoding specified in
the control file.
• Set the
NVT.pass-through-mode
system property to
TRUE to disable
conversion of incoming
character data to UTF16
for native Java Strings.
• Set the
NVT.pass-through-mode
system property to
FALSE to enable Java
Client conversion of
incoming character data
to UTF16.
The JDBC connectivity
server communicates the
current ISO_MAPPING
value to the Java Client
so that it knows what
character set to store in
ISO88591 columns in each
of the three
configurations:
• If the value of
ISO_MAPPING is
ISO88591 (ISO88591
configuration), the
Java Client
automatically enables
pass-through mode.
All client locale
character data is
stored in ISO88591
columns in its original
client locale character
encoding but treated
as if it were encoded
in ISO8859-1.
• If the value of
ISO_MAPPING is
SJIS (SJIS
configuration), the
Java Client
automatically disables
pass-through mode.
Client locale SJIS
encoding is stored in
ISO88591 columns in
SJIS format.
• If the value of
ISO_MAPPING is
UTF8 (Unicode
configuration), the
Java Client
automatically disables
pass-through mode.
Client locale-encoded
data is stored in
ISO88591 columns in
UTF8 format.
Encoding Data Sources
Encoding for a data source or pipe source must be Java-compliant and can be specified in a global
option or a sources option of the control file. File and pipe sources that do not specify an encoding
at the source level inherit the encoding specified in the global options section. If an encoding is
not specified in the global options section or for the individual source, the default Java encoding
is used. Java determines the default encoding by the locale of the client machine. A syntax error
reads the control file if an encoding is specified for a JDBC or JMS data source.
On a load operation, the data file is read using the specified or default encoding and converted
to UTF16 Java strings, then encoded in the character set specified by ISO_MAPPING for ISO88591
columns or retained in UTF16 encoding for UCS2 columns. On an extract job, the reverse actions
44 Configuring Neoview Client Applications