That is because it isn't done through scripts.
These things are done by first using the data dictionary to define your tables and such and then creating a .df file.
You can use a scripts to dump the .df file:
Code:
#!/bin/sh
#
# $1 = db name
# $2 = .df file name
# if no .df file is specified assume that is is $1.df
DFNAME=${2-${1}}.df
mpro ${1} -p dumpdf.p -param "${DFNAME}"
/* dumpdf.p
*
* dump schema
*
* 08/05/02 tom created
*
* August 5, 2002 */
define variable fname as character no-undo.
fname = session:parameter.
run prodict/dump_df.p ( "ALL", input fname, "" ).
quit.
... and to load the .df file.
Code:
#!/bin/sh
#
# $1 = db name
# $2 = .df file name
# if no .df file is specified assume that is is $1.df
DFNAME=${2-${1}}.df
# strip area assignments out of .df file
cat ${DFNAME} | grep -v " AREA " > ${DFNAME}.noarea
mpro ${1} -p loaddf.p -param "${DFNAME}"
/* loaddf.p
*
*/
run prodict/load_df ( input session:parameter ).
quit.
You might want to get fancy and create an incremental .df file. To do that you connect two databases and select " Create Incremental .df File..." from the admin menu.
Fun fact -- if you go through the incremental .df process twice (with the order of the databases reversed the second time) you can create a "backout" .df file in case your update fails and you need to be able to remove your changes.