![OPEN statement [ESQL] [SP]](arrow-left.gif){kind=icon}
![PARAMETERS statement [DBISQL]](arrow-right.gif){kind=icon}
Writes the current query results to a file.
OUTPUT TO filename
[ APPEND ] [ VERBOSE ]
[ FORMAT output-format ]
[ ESCAPE CHARACTER character ]
[ DELIMITED BY string ]
[ QUOTE string [ ALL ] ]
[ COLUMN WIDTHS ( integer, … ) ]
[ HEXADECIMAL { ON | OFF | ASIS } ]
[ ENCODING encoding ]
ASCII| DBASEII | DBASEIII | EXCEL | FIXED | FOXPRO | HTML | LOTUS | SQL | XML
string or identifier
Places the contents of the Employees table in a file in ASCII format:
SELECT * FROM Employees; OUTPUT TO employee.txt FORMAT ASCII
Places the contents of the Employees table at the end of an existing file, and includes any messages about the query in this file as well:
SELECT * FROM Employees; OUTPUT TO employee.txt APPEND VERBOSE
Suppose you need to export a value that contains an embedded line feed character. A line feed character has the numeric value 10, which you can represent as the string '\x0a' in a SQL statement. You could execute the following statement, with HEXADECIMAL ON:
SELECT 'line1\x0aline2'; OUTPUT TO file.txt HEXADECIMAL ON
You get a file with one line in it, containing the following text:
line10x0aline2
If you execute the same statement with HEXADECIMAL OFF, you get the following:
line1\x0aline2
Finally, if you set HEXADECIMAL to ASIS, you get a file with two lines:
'line1 line2'
Using ASIS generates two lines because the embedded line feed character has been exported without being converted to a two-digit hex representation, and without a prefix.
The OUTPUT statement copies the information retrieved by the current query to a file.
You can specify the output format with the optional FORMAT clause. If no FORMAT clause is specified, the Interactive SQL OUTPUT_FORMAT option setting is used.
The current query is the SELECT or LOAD TABLE statement that generated the information that appears on the Results tab in the Results pane. The OUTPUT statement reports an error if there is no current query.
OUTPUT is especially useful in making
the results of a query or report available to another application,
but it is not recommended for bulk operations. For high-volume data
movement, use the ASCII and BINARY data
extraction functionality with the SELECT statement.
The extraction functionality provides much better performance for
large-scale data movement, and creates an output file you can use
for loads.
APPEND This optional keyword is used to append the results of the query to the end of an existing output file without overwriting the previous contents of the file. If the APPEND clause is not used, the OUTPUT statement overwrites the contents of the output file by default. The APPEND keyword is valid if the output format is ASCII, FIXED, or SQL.
VERBOSE When the optional VERBOSE keyword is included, error messages about the query, the SQL statement used to select the data, and the data itself are written to the output file. If VERBOSE is omitted (the default), only the data is written to the file. The VERBOSE keyword is valid if the output format is ASCII, FIXED, or SQL.
FORMAT Allowable output formats are:
ASCII The output is an ASCII format file with one row per line in the file. All values are separated by commas, and strings are enclosed in apostrophes (single quotes). The delimiter and quote strings can be changed using the DELIMITED BY and QUOTE clauses. If ALL is specified in the QUOTE clause, all values (not just strings) are quoted.
Three other special sequences are also used. The two characters \n represent a newline character, \\ represents a single \, and the sequence \xDD represents the character with hexadecimal code DD. This is the default output format.
If you are exporting Java methods that have string return values, you must use the HEXADECIMAL OFF clause.
DBASEII The output is a dBASE II format file with the column definitions at the top of the file. Note that a maximum of 32 columns can be output. Column names are truncated to 11 characters, and each row of data in each column is truncated to 255 characters.
DBASEIII The output is a dBASE III format file with the column definitions at the top of the file. Note that a maximum of 128 columns can be output. Column names are truncated to 11 characters, and each row of data in each column is truncated to 255 characters.
EXCEL The output is an Excel 2.1 worksheet. The first row of the worksheet contains column labels (or names, if there are no labels defined). Subsequent worksheet rows contain the actual table data.
FIXED The output is fixed format with each column having a fixed width. The width for each column can be specified using the COLUMN WIDTHS clause. No column headings are output in this format.
If COLUMN WIDTHS is omitted, the width for each column is computed from the data type for the column, and is large enough to hold any value of that data type. The exception is that LONG VARCHAR and LONG BINARY data defaults to 32KB.
FOXPRO The output is a FoxPro format file (the FoxPro memo field is different than the dBASE memo field) with the column definitions at the top of the file. Note that a maximum of 128 columns can be output. Column names are truncated to 11 characters. Column names are truncated to 11 characters, and each row of data in each column is truncated to 255 characters.
HTML The output is in the Hyper Text Markup Language format.
LOTUS The output is a Lotus WKS format worksheet. Column names are put as the first row in the worksheet. Note that there are certain restrictions on the maximum size of Lotus WKS format worksheets that other software (such as Lotus 1-2-3) can load. There is no limit to the size of file Interactive SQL can produce.
SQL The output is an Interactive SQL INPUT statement required to recreate the information in the table.
Sybase IQ does not support the INPUT statement.
You would need to edit this statement to a valid LOAD TABLE (or INSERT)
statement to use it to load data back in.
XML The output is an XML file encoded in UTF-8 and containing an embedded DTD. Binary values are encoded in CDATA blocks with the binary data rendered as 2-hex-digit strings. The LOAD TABLE statement does not accept XML as a file format.
ESCAPE CHARACTER The default escape character for characters stored as hexadecimal codes and symbols is a backslash (\), so \x0A is the line feed character, for example.
This default can be changed using the ESCAPE CHARACTER clause. For example, to use the exclamation mark as the escape character, you would enter:
... ESCAPE CHARACTER '!'
DELIMITED BY The DELIMITED BY clause is for the ASCII output format only. The delimiter string is placed between columns (default comma).
QUOTE The QUOTE clause is for the ASCII output format only. The quote string is placed around string values. The default is a single quote character. If ALL is specified in the QUOTE clause, the quote string is placed around all values, not just around strings.
COLUMN WIDTHS The COLUMN WIDTHS clause is used to specify the column widths for the FIXED format output.
HEXADECIMAL The HEXADECIMAL clause specifies how binary data is to be unloaded for the ASCII format only. When set to ON, binary data is unloaded in the format 0xabcd. When set to OFF, binary data is escaped when unloaded (\xab\xcd). When set to ASIS, values are written as is, that is, without any escaping—even if the value contains control characters. ASIS is useful for text that contains formatting characters such as tabs or carriage returns.
ENCODING The encoding argument lets you specify the encoding that is used to write the file. The ENCODING clause can be used only with the ASCII format.
If encoding is not specified, Interactive SQL determines the code page that is used to write the file as follows, where code page values occurring earlier in the list take precedence over those occurring later:
The code page specified with the DEFAULT_ISQL_ENCODING option (if this option is set)
The code page specified with the -codepage option when Interactive SQL was started
The default code page for the computer Interactive SQL is running on
In Interactive SQL, the Results tab displays only the results of the current query. All previous query results are replaced with the current query results.
None
