Clarion Community Help

| **Navigation:**  ISAM File Drivers > dBaseIV Database Driver >====== dBaseIV:Other ====== | [[dbaseiv supported commands and attributes.htm|{{btn_prev_n.gif|Previous page}}]][[coming future.htm|{{btn_home_n.gif|Return to chapter overview}}]][[dos database driver.htm|{{btn_next_n.gif|Next page}}]] |
|  ||

**International Sort Sequence**

dBase IV sorts as if there were no diacritics in a field, thus A is sorted the same as Ä. If two words are identical except for diacritic characters, then the words are sorted as though the diacritic character was greater than the normal character. For example Äa <; Ab <; Äb whereas a CLADIGRAPH of ÄAE will sort as Ab <; Äa <; Äb. Solution- if the same file is used in Clarion and dBase IV, issue a BUILD statement rebuild the keys before updating the file (reading the file causes no problems).

**Boolean Evaluation**

dBase IV allows a logical field to accept one of 11 possible values (1,0,y,Y,n,N,t,T,f,F or a space character). The space character is neither true nor false. When using a logical field from a preexisting database in a logical expression, account for all these possibilities. Remember that when a STRING field is used as an expression, it is true if it contains any data and false if it is equal to zero or blank. Therefore, to evaluate a Logical field's truth, the expression should be true if the field contains any of the "true" characters (T,t,Y, or y). For example, if a Logical field were used to specify a product as taxable or nontaxable, the expression to evaluate its truth would be:

//(If Condition)//:

**    Taxable='1' OR Taxable='T' OR Taxable='t' OR Taxable='Y' OR Taxable='y'**

**Large MEMOs**

Clarion supports MEMO fields up to a maximum of 64K. If you have an existing file which includes a memo greater than 64K, you can use the file but not modify the large MEMOs.

You can determine when your application encounters a large MEMO by detecting when the memo pointer variable is non-blank, but the memo appears to be blank. Error 47 (Bad Record Declaration) is posted, and any modification to the MEMO field is ignored.

**Long Field Names**

dBase IV supports a maximum of 10 characters in a field name. If you require more, use an External Name with 10 characters or less.

**Key Definition**

dBase IV supports the use of expressions to define keys. Within the Dictionary Editor, you can place the expression in the external name field in the //Key Properties //dialog. The general format of the external name is :

**'FileName=T[Expression]'**

Where **FileName**// //represents the name of the index file (which can contain a path and file extension), and **T**// //represents the type of the index. Valid types are: C = character, D = date, and N = numeric. If the type is D or N then **Expression**// //can name only one field.

Multiple-index (.MDX) files require the NAME() attribute on a KEY or INDEX to specify the storage type of the key and any expression used to generate the key values. The general format of the NAME() attribute on a KEY or INDEX is:

**    NAME('TagName|FileName[PageSize]=T[Expression],FOR[Expression]')**

The following documents the parameters for the NAME() attribute:

| **TagName** | Specifies the name of an index tag within a multiple index file. If omitted the driver creates a dBase IV style .NDX file using the name specified in FileName. |
| **FileName** | Specifies the name of the index file, which may contain a path and extension. |
| **PageSize** | Specifies that when creating a .MDX file, (if a TagName is specified), a number in the range 2-32 specifying the number of 512-byte blocks in each index page. This value is only used when creating the file. If you specify multiple values with declarations for different tags in the same .MDX file, the largest value will be selected. The default value is 2. |
| **T** | Specifies the type of index. Legal types are C = character, D = date, N = numeric. If the type is D or N then //Expression //may name //only one //field. |
| **Expression** | Specifies an expression to generate the index. It may refer to multiple fields and invoke multiple xBase functions. The functions currently supported are listed below. Square brackets must enclose the expression. |

Elements of the NAME() attribute may be omitted from the right. When specifying an Expression, you must also specify the type and name. If the Expression is omitted, the driver determines the Expression from the key fields when the file is created, or from the index file when opened.

If the type is omitted, the driver determines the index type from the first key component when the file is created, or from the index file when opened.

If the NAME() attribute is omitted altogether, the index file name is determined from the key label. The path defaults to the same location as the .DBF.

Tag names are limited to 9 characters in length. If the supplied name is too long it is automatically truncated.

Specify all field names in the NAME() attribute without a prefix.

dBase IV additionally supports the use of the Xbase FOR statement in expressions to define keys. The expressions supported in the FOR condition must be a simple condition of the form:

**expression comparison_op expression**

//comparison_op// may be <;, <;=, =<;, <;>, =, =>, >= or >.

The expression may refer to multiple fields in the record and contain xBase functions. Square brackets must enclose the expression. The currently supported functions appear below. If the driver encounters an unsupported Xbase function in a preexisting file, it posts error 76 'Invalid Index String' when the file is opened for keys and static indexes.

String expressions may use the '+' operator to concatenate multiple string arguments. Numeric expressions use the '+' or '-' operators with their conventional meanings. The maximum length of a dBase IV expression is 250 characters.

**Supported xBase Key Definition Functions**

| **ALLTRIM(string)** | Removes leading and trailing spaces. |
| **CTOD(string)** | Converts a string key to a date. The //string //must be in the format mm/dd/yy; the result takes the form 'yyyymmdd'. The yyyy element of the date defaults to the twentieth century. An invalid date results in a key containing blanks. |
| **DELETED()** | Returns TRUE if the record is deleted. |
| **DTOC(date)** | Converts a date key to string format 'mm/dd/yy.' |
| **DTOS(date)** | Converts a date key to string format 'yyyymmdd.' |
| **FIXED(float)** | Converts a float key to a numeric. |
| **FLOAT(numeric)** | Converts a numeric key to a float. |
| **IIF(bool,val1,val2)** | Returns val1 if the first parameter is TRUE, otherwise returns val2. |
| **LEFT(string, n)** | Returns the leftmost //n //characters of the string key as a string of length //n//. |
| **LOWER(string)** | Converts a string key to lower case. |
| **LTRIM(string)** | Removes spaces from the left of a string. |
| **RECNO()** | Returns the current record number. |

| **RIGHT(string, n)** | Returns the rightmost //n //characters of the string key as a string of length //n//. |

| **RTRIM(string)** | Removes spaces from the right of a string. |
| **STR(numeric[,length[,decimal places]])** | Converts a numeric to a string. The length of the string and the number of decimal places are optional. The default string length is 10, and the number of decimal places is 0. |
| **SUBSTR(string,offset,n)** | Returns a substring of the //string //key starting at //offset //and of //n //characters in length. |
| **TRIM(string)** | Removes spaces from the right of a string (identical to RTRIM). |
| **UPPER(string)** | Converts a string key to upper case. |
| **VAL(string)** | Converts a string key to a numeric. |