Hi,

I am taking over some code that someone has written.  I am getting an error when I try to update the table.

"Invalid delimited data:  text qualifier must be followed by a column delimiter (except the last column)."

The file is a tab delimited file and when I look at everything it looks fine.  Any quick info on this?

Thanks

Elena

EBOUGHEY,

 I take it this is a file data import from a text file?

Logicalman

Yes it is.

EBOUGHEY,
 
 Check the file, and see if there are any spaces in the first row (Column headers). For safety, there should never be any spaces in column names, use underscores (_) instead. IN SQL to reference a column name that has spaces (e.g. First Name), it would be referenced [First Name].

Logicalman

the file doesn't have a header row.  It imports as col1-col014. The dts package renames the columns after they are imported. I don't have the file here (wish i did now).

I recall this happening before.  The programmer told me he had to open it up in excel and do something to one of the columns.  I just can't figure it out though.

EBOUGHEY,

 Unfortunately, I think you need to get the file and import it into XL to see the offending column.

 By the sound of what the last programmer stated, it sounds like there may be one or more text data columns, and one contains either a control code or an extra comma.

 Sorry, I can't be of use without the file.

Logicalman
 

No, thanks a lot for what you've given me.  I had no idea where to even start and now i do.

Thanks again,

Elena

Elena,

 Please post the result of what you find, I've got an interest in this now. You never know, I may get the same thing sooner, rather than later!

Logicalman

Here is a sample of the data and the actual code the programmer used.  I hope this can help alleviate the problem.  I noticed he did not use a substr(ltrim on columns 1-3:

"17730454"    "1GNEK13Z83R129014"    "385005"    " RONALD CLYDE* GRAHAM"    "415 STILL FOREST TER"    "SANFORD"    "FL"    "32771-8367"    "4073284982"    "2003-03-07 00:00:00"    "2"    "1"    "9"    "2003"
"17767368"    "3GNFK16Z22G357492"    "385005"    " DIRK GLEN PRUSIA"    "6715 S SYLVAN LAKE DR"    "SANFORD"    "FL"    "32771-9052"    "4073228390"    "2003-04-25 00:00:00"    "2"    "1"    "9"    "2003"



insert into mastermailhistory(id, vin, clientnumber, buyername, address1, city, state, zip, phone, lastrepairorderstamp, dropweek, dropmonth, dropyear, droptype)
select col001, col002, col003, substring(ltrim(col004), 1, 60), substring(ltrim(col005), 1, 40), substring(ltrim(col006), 1, 30), substring(ltrim(col007), 1, 2), substring(ltrim(col008), 1, 10), substring(ltrim(col009), 1, 9),  substring(ltrim(col010), 1, 10), substring(ltrim(col011), 1, 1), substring(ltrim(col012), 1, 1), substring(ltrim(col013), 1, 2), substring(ltrim(col014), 1, 4)
from post where col002 <> vin and col013 <> dropmonth

A quick trick, if you have MS Access, is to import it using Access. In most cases it will import the good records and subsequently push the bad records, w/ a reason, to an error table. Look at the error table for hints and the known offending records. Had to use Access to clean Web Log recs I later imported into SQL Server.

Thanks

J. Kusch

Some problems that we had where we got this type of error converting were that the data had the actual delimiter character as part of the data or had too many or not enough delimters at the end of the line (I don't rememebr which it was) for it to recognize the end of a line. Sometimes we had line feeds in the data ina field and we had to strip those out or the datatypes didn't match up. In the long run to fix these problems with our import data, my VB.net programmers wrote a parser to clean up the records. I run that first, then my DTS package.

Access brought the file right in.  No errors.

I'm really not that well versed in SQL but I wonder if having just col001 instead of:

substr(ltrim(col001),1,17

makes a difference.  Would It?

Elena

Does Anyone have any other suggestions?  Can I import this differently?  Especially with me being such a novice?

I'm desperate for some help now.  I am running so far behind because I can't post back to the file.

Thanks for any info you may have,

Elena

If it loaded into Access correctly, can you then port it from Access to SQL Server.

Thanks

J. Kusch

I've not done that before but I can try that.

Would it work if I added a header column to the text file with col001, col002 etc?

Elena

Adding a column header would only gain you a more user readable column for your commands. There would be no difference in how it is parsed and processed.

Thanks

J. Kusch