I'm looking for a way to essentially "COPY" or "IMPORT" 100K+ records contained in 
files on CD-ROM media into database files on Universe (and if possible jBASE) systems.  I have the 
flexibility of choosing the format the files contained on the CD-ROM.  I'd prefer to avoid hand-crafting a 
program, but am open to using a freely distributable, (and free) third-party tool.  A tool that is the 
converse of DOWNLOAD would be excellent, or a command-line program that imports from a standard formatted 
file would be fine, too.

Any suggestions?

Tom,

Off the top of my head:

/DATA directory on CDROM holding each record in a single file (delimit record elements with CR or LF as appropriate for your O/S)

ED VOC IMPFILE
F Data Import
/dev/cdrom/DATA


I don't know if the dictionary element is required on line 3, if it is create a UV file to use as the dictionary (perhaps copy or use an existing dictionary?)


COPY FROM IMPFILE TO OTHERFILE ALL

This probably isn't the fastest solution, but it is pretty simple.

----------------------------

Alternately, have them delimit a single file with CR or LF (record elements separated by @VM)

1. OPENSEQ the file to DATAFILE

DONE = @FALSE
LCNT = 0
LOOP
        READSEQ DATALINE FROM DATAFILE ON ERROR
                * Some nasty error
                DONE = 2
        END THEN
                LCNT += 1
                GOSUB PROCESSDATALINE
        END ELSE
                * EOF
                DONE = 1
        END
UNTIL DONE DO REPEAT

PRINT LCNT:" Lines Process"
IF DONE GT 1 THEN
        PRINT "READSEQ Error ":STATUS()
END

-------------------------------

Craig
-------
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please visit http://listserver.u2ug.org/

Reply via email to