Word 2000 merge problem - field name starting with underscore

[Nick Kasoff]

I developed a set of merge documents which work with an Access
database, using a small text file to pass data. The project was
originally developed in 97, but was later deployed on another system
with 2000. On one of the systems with 2000, all the field names which
started with an underscore are prepended with an alphabetic character
(I think it was a W, but not positive). So of course, the merge fails,
because those fields don't exist. On the other system, everything
works fine.

On the system which doesn't work, I uninstalled Office 2000 and
installed Office 97, but it still doesn't work. Since there are
hundreds of Word documents, changing the merge fields isn't a good
option. So how can I get rid of this problem?

 

[Peter Jamieson]

When I started to try to replicate this, one guess was that the difference
between the machines that work and the one that does not could be that
a. Word is using ODBC to open the data source on all the machines except
one (which would use Word's internal text converter), or vice versa, or
b. Word is using ODBC on all the machines but for some reason the ODBC
version or configuration is different on one of the machines.

Since I haven't been able to replicate the problem at all using a text file
as data source, (a) seems to be ruled out. (b) still seems possible but it
is a difficult hypothesis to test - maybe you could print the value of
ActiveDocument.MailMerge.DataSource.ConnectString and .QueryString, and
check the ODBC driver versions.

But in any case that idea does not really sound right to me - I don't
remember seeing the text ODBC driver prepending a "w" or anything else,
although it can get confused with tab-delimited files if you have not told
ODBC that the file is tab-delimited.

What is the actual format of the data - ANSI/OEM? Comma-delimited,
tab-delimited or what? And are the headers actually in the data file or in a
separate header file?

 

[Nick Kasoff]

What is the actual format of the data - ANSI/OEM? Comma-delimited,
tab-delimited or what? And are the headers actually in the data file or in a
separate header file?

Data is comma delimited, with headers in the data file.

Also, though probably irrelevant, both systems are XP Professional.

 

[Cindy M -WordMVP-]

Hi Nick,

I developed a set of merge documents which work with an Access
database, using a small text file to pass data. The project was
originally developed in 97, but was later deployed on another system
with 2000. On one of the systems with 2000, all the field names which
started with an underscore are prepended with an alphabetic character
(I think it was a W, but not positive)

I, too, have never encountered anything like this; neither personally,
nor in support.

Another angle to consider is: how is this text file being generated?

And one other thing that could, perhaps, have an influence: on this one
particular system is there any chance that Word 2000 was installed
differently? Such as being part of Works, or some other special kind of
Suite?

Cindy Meister
INTER-Solutions, Switzerland
http://homepage.swissonline.ch/cindymeister (last update Sep 30 2003)
http://www.word.mvps.org

This reply is posted in the Newsgroup; please post any follow question
or reply in the newsgroup and not by e-mail

 

[Nick Kasoff]

The text file was generated as an export from Outlook. Word was
installed on both systems as part of Office 2000 Professional.

 

[Cindy M -WordMVP-]

Hi Nick,

The text file was generated as an export from Outlook. Word was
installed on both systems as part of Office 2000 Professional.

OK, now I'm confused. How does Access enter into all this?

And have you researched Peter's thoughts on the matter?

Cindy Meister
INTER-Solutions, Switzerland
http://homepage.swissonline.ch/cindymeister (last update Sep 30 2003)
http://www.word.mvps.org

This reply is posted in the Newsgroup; please post any follow question or reply in the newsgroup and
not by e-mail