Very Computer

Dict Item

0001 S
0002 09
0003 TELEPHONE#
0004 09
0005
0006
0007
0008
0009 L
0010 11
0011 1

Thanks.

You might have a problem.  I don't know how ViaDuct works when importing
data, but I know what I did when I used ViaDuct to download the data.

Here's how it works.

The data has to be normalized to work in a .DBF file.  It can either be done
on the Pick side or the DBF side.  I've done it both ways.

I wrote a little program that would send the escape and command sequences to
ViaDuct to prepare it to capture the file.  Then I would send down the data
using Pick's LIST or SORT verbs.  I would get the standard Pick output, so I
had to set my TERM length to 0 so that it would not put out column headers
quite as frequently.  I think I still got them every 65536 lines or so.

This would look like a Pick printout, until I discovered the LIST or SORT
with TAB delimiter and that made it look worse.

There would be columns

NAME              TYPE    TELEPHONE
ME OR YOU   PHONE 714-444-4444
                         FAX       714-444-4445

Well - hopefully you get the idea.  There would be no name associated with
FAX.  The first attempt had the normalization done on the FoxPro side.  That
programmer wrote a quick and dirty FoxPro program to fill in the blanks from
above.  So, in the FoxPro database, the data appeared as follows

NAME              TYPE    TELEPHONE
ME OR YOU   PHONE 714-444-4444
ME OR YOU   FAX       714-444-4445

When I did the downloads for conversion to Progress (ugg 8(> phooey ), I had
to do the normalization from within Pick.  So, I had the program DCOUNT the
attributes that I knew were multi-value and then output the other values
that number of times.  It was a pain in the neck and I hope somebody else
has a neater solution.

Good Luck,

Larry Hazel

Quote:> I have been importing data from Pick using ViaDuct Terminal Software.
> I use the .dbf3 file extension to make it a .dbf file. The problem is
> some of the data is left dangling on a line of its own. Example -
> in a record the telephone and fax number are on the same line. The data
> looks like this in the Pick record: 407-8284563]407-8284564. The phone
> number part imports with the name but the fax number is on a line right
> below the phone number by itself with no name. When merging the .dbf
> file with other software packages, I lose the fax number. Is there a
> way I can import the fax number with the right name? Here is how the
> dict item is defined:

> Dict Item

> 0001 S
> 0002 09
> 0003 TELEPHONE#
> 0004 09
> 0005
> 0006
> 0007
> 0008
> 0009 L
> 0010 11
> 0011 1

> Thanks.

The FoxPro fix sounds like the ticket. I will try it and report to the group
the results.

Best Wishes,
Rafael

ViaDuct supports a "duplicate" option in dictionary-driven file transfers
that automatically repeats the non-multivalue field.  You can select the
duplicate option by specifying "D+" in the Options box, or by filling in a
"+" in the Duplicate text box if you press the SELECT button while in the
Options box.  For more information on field duplication, you can look up
"duplicate" or "duplicate fields" in the ViaDuct on-line help.

Robert Catalano
Via Systems, Inc.
Tel:(718) 788-6230
Fax:(718) 768-4661
Web: www.via.com

Quote:> I have been importing data from Pick using ViaDuct Terminal Software.
> I use the .dbf3 file extension to make it a .dbf file. The problem is
> some of the data is left dangling on a line of its own. Example -
> in a record the telephone and fax number are on the same line. The data
> looks like this in the Pick record: 407-8284563]407-8284564. The phone
> number part imports with the name but the fax number is on a line right
> below the phone number by itself with no name. When merging the .dbf
> file with other software packages, I lose the fax number. Is there a
> way I can import the fax number with the right name? Here is how the
> dict item is defined:

> Dict Item

> 0001 S
> 0002 09
> 0003 TELEPHONE#
> 0004 09
> 0005
> 0006
> 0007
> 0008
> 0009 L
> 0010 11
> 0011 1

> Thanks.

DICT PHONE      DICT FAX
001 S          
002 0      
003 PHONE  
004                
005        
006  
007
008 F;0;(TFILENAME;X1;9;9)
009 L
010 10

DICT FAX
001 A
002 0
003 FAX
004
005
006
007
008 F;0;(TFILENAME;X2;9;9)
009 L
010 11

I do not know why you have anything in attr 4, but I dont use it. Obviously
FILENAME must be replaced with the name of your file.
This translate will use your id to read your record attr 9 (X1 = first
multivalue  X2 = second multivalue... etc)

Now I'm at work, it's easy for me to email it to anyone who asks ... or
I'll post it if enough people shout.

I did some downloads, and the D+ option works perfectly. "You Da Man."

Thanks,
Rafael

Terry

Quote:> I have been importing data from Pick using ViaDuct Terminal Software.
> I use the .dbf3 file extension to make it a .dbf file. The problem is
> some of the data is left dangling on a line of its own. Example -
> in a record the telephone and fax number are on the same line. The data
> looks like this in the Pick record: 407-8284563]407-8284564. The phone
> number part imports with the name but the fax number is on a line right
> below the phone number by itself with no name. When merging the .dbf
> file with other software packages, I lose the fax number. Is there a
> way I can import the fax number with the right name? Here is how the
> dict item is defined:

> Dict Item

> 0001 S
> 0002 09
> 0003 TELEPHONE#
> 0004 09
> 0005
> 0006
> 0007
> 0008
> 0009 L
> 0010 11
> 0011 1

> Thanks.