[SQL]
|
Description
|
Factory Value
|
Access Mode
|
---|---|---|---|
AllowDuplicateIndex
|
If set to yes, allows duplicate index definitions. This parameter provides compatibility with earlier versions. In versions preceding 4.5, it was possible to create duplicate indexes.
|
no
|
RO
|
AuditTrailEnabled
|
If set to yes, audit trail is enabled. Possible values are yes and no.
|
no
|
RO
|
CharPadding
|
When set to yes, solidDB® enforces SQL standard padding of CHAR values with blanks (right-filled) to the length defined for the column. With the default setting (no), the blanks are discarded. The value of the parameter does not affect comparisons (where blanks are always discarded).
Note
▪This parameter is effective only when using ODBC or JDBC drivers, not when using solidDB® SQL Editor (solsql).
▪This parameter affects the ODBC and JDBC driver behavior.
▪This parameter is not effective in Unicode databases (General.InternalCharEncoding=UTF8).
|
no
|
RO
|
ConvertOrsToUnionsCount
|
This parameter specifies the maximum number of OR operations that may be converted to UNION operations.
Note This parameter does not force the optimizer to convert OR operations to UNION operations; it only sets a maximum limit on the number of OR operations that the server may convert to UNION operations.
|
10
|
RO
|
CursorCloseAtTransEnd
|
By default, the solidDB® ODBC driver closes all the cursors opened from the user connection when a commit is called with SqlTransact from this connection. If this parameter is set to no, the cursors are kept open.
|
yes
|
RO
|
DecFloatPrecision16
|
If set to yes, the precision of the decimal float data type is limited to 16 (same as in version 4.5 and earlier).
In storage, the decimal float type is inflicted by the column type specification 'DECIMAL' (without scale and precision).
Also, expressions involving DECIMAL or NUMERIC data types may produce decimal float values.
By default (no), the precision of the decimal float data type is 52.
Possible values are yes and no.
|
no
|
RO
|
EmulateOldTimestampDiff
|
If included in the solid.ini file and set to yes, the old TIMESTAMPDIFF behavior is emulated by the server. This old behavior returns the integer number of intervals of type interval by which timestamp_exp2 is greater than timestamp_exp1. Otherwise, the default is the new behavior which returns the integer number of interval as the amount of full units between timestamp_exp1 and timestamp_exp2.
|
no
|
RW/
Startup |
EnableHints
|
If set to no, hints are disabled.
For details on hints, see Using Optimizer hints in solidDB® SQL Guide.
Sometimes hints in queries may produce undesirable effects. They may be disabled by setting this parameter to no.
|
yes
|
RW/
Startup |
ExecuteNodataODBC3Behaviour
|
By default, when the execution of a DELETE or UPDATE statement does not affect any rows, the statement returns SQL_SUCCESS. This is the ODBC v.2 behavior. By setting this parameter to yes, the SQLSTATE returned in those cases is SQL_NO_DATA, which conforms to ODBC v.3.
|
no
|
RO
|
Info
|
Sets the level of informational messages [0-8] printed from the server (0=no info, 8=all info); information is written into the file defined by parameter InfoFileName, or into soltrace.out if InfoFileName is not defined.
|
0
|
RW/
Startup |
InfoFileFlush
|
If set to yes, flushes info file after every write operation
|
yes
|
RW/
Startup |
InfoFileName
|
Default info file name. The default name is soltrace.out. Because the soltrace.out file can contain information from several sources, set the InfoFileName to another name if you set the Info or SQLInfo parameters to a number larger than 0.
|
soltrace.out
|
RW/
Startup |
InfoFileSize
|
Sets the maximum size of the info file.
|
100 MB
|
RW/
Startup |
IsolationLevel
|
Possible values:
3 (SERIALIZABLE)
2 (REPEATABLE READ)
1 (READ COMMITTED)
For more information about transaction isolation levels, see SET TRANSACTION ISOLATION in the solidDB® SQL Guide and Choosing transaction isolation levels. Important: In-memory tables support only the READ COMMITTED and REPEATABLE READ isolation levels.
|
1 (Read Committed)
|
RW
|
Latin1CaseSemantics
|
If set to no, uppercase/lowercase conversions are disabled for characters with decimal value between 126 and 256.
|
yes
|
RW/
Startup |
MaxBlobExpressionSize
|
Certain string operations use only the first N bytes of a character value, not the entire value. For example, the LOCATE() operation checks only the first N bytes of the string. If you want to tell the server to check further into (or less far into) long strings, you may set this parameter.
By default, the units are kilobytes — for example, "64" means 64KB You may specify "MB" if you want to express the units in megabytes.
This parameter applies to all the character data types, including CHAR, VARCHAR, LONG VARCHAR, WCHAR, WVARCHAR, and LONG WVARCHAR. Since the Wide character data types use 2 bytes per character, the number of characters searched is half the number of bytes.
For example, if you set MaxBlobExpressionSize to 64K bytes, then the first 32K characters of Wide character data types will be searched.
|
1024KB (1MB) Unit: 1 KB m=MB
|
RW/
Startup |
MaxNestedProcedures
|
Sets the maximum number of allowed nested procedures. If this parameter is defined too high, the server stack may become insufficient depending on the operating system.
|
16
|
RW/
Startup |
MaxNestedTriggers
|
Sets the maximum number of allowed nested triggers. This maximum number includes both direct and indirect nesting, so both A → A → A and A → B → A are counted as three nested triggers.
|
16
|
RW/
Startup |
NumericPadding
|
If set to yes, causes output of DECIMAL and NUMERIC to be zero-right-padded up to the specified scale.
|
no
|
RO
|
PreferExactNumericFunctions
|
Controls the precision of SUM() and AVG() type functions.
When set to yes:
▪If the argument of SUM() or AVG() function is of exact numeric datatype (TINYINT, SMALLINT, INTEGER, BIGINT, NUMERIC or DECIMAL), the function returns the result in DECIMAL data type. The default precision of DECIMAL is 52 and scale is floating.
▪Functions FLOOR() CEILING() and ABS() return their result in the same data type as the argument.
Note The SQL.PreferExactNumericFunctions parameter can be set only by editing the solid.ini file.
|
no
|
RW/
Startup |
ProcedureCache
|
Specifies the number of procedures which set the size of cache memory for parsed procedures.
|
10
|
RW/
Startup |
SimpleOptimizerRules
|
When set to yes, simple optimization rules are used instead of using full optimization rules.
|
no
|
RO
|
SortArraySize
|
This parameter sets the size of the array that SQL uses when ordering the result set of a query.
The units are "rows" — for example, if you specify a value of 1000, the server will create an array big enough to sort 1000 rows of data.
|
4000
|
RW
|
SQLInfo
|
Sets the level of informational SQL level messages [0-8] (0=no info, 8=all info); information is written into a file defined by parameter InfoFileName, or into soltrace.out if InfoFileName is not defined.
|
0
|
RW/
Startup |
TimestampDisplaySize19
|
If set to yes, the precision (maximum number of digits) of data type timestamp is set to 19. In this case, the timestamp is presented as yyyy-mm-dd hh:mm:ss.
|
no
|
RO
|
TriggerCache
|
Specifies the number of triggers which set the size of cache memory that each user has for triggers.
|
20
|
RW/
Startup |
UpCaseQuotedIdentifiers
|
If set to yes, the SQL identifiers given in double quotation marks are converted to upper case when reaching the solidDB® server. If set to no, the upper/lower case distinction is preserved whereby uniqueness of names incorporates the case too.
|
yes
|
RW/
Startup |