Column name line

[xnappo]

Hello,

I deal with a table type where the 3rd row is the column headers and the first two lines are 'don't cares'.

I would like to request that the 'First line contains' drop-down be change to 'Row containing column headers'...?

Alternatively, is there a way to strip the first two lines from comparison altogether such that the third line because the 'column header' line?

Thanks!
xnappo

[Aaron]

Hello,

Have you had a chance to give our Data Compare a try? Assuming the delimiters are auto-detected for your columns, it will then sort your data based on the defined key (default: column 1). You can disable this sorting or disable/change the key.

Alternatively, you can use our Text Compare, and create grammar items that define the data you wish to mark as Unimportant. This would depend specifically on the exact structure of your files. From your description, it sounds like the Data Compare would better serve your needs.

http://www.scootersoftware.com/support.php?zz=kb_unimportantv3

If you have any questions, please let us know. Also, if you are having any trouble, please email us a pair of sample files and your Support package (Help menu -> Support; Export) to [email protected]. If you email us, please include a link back to this forum post for reference. Thanks.

[Aaron]

For stripping off the first two lines, you can perform an external conversion that removes the first two lines (or any other data), and then output that information into a temporary file. We can plug that external conversion into either the Data Compare or Text Compare as part of a File Format. The File Format would perform the conversion every time it is called.

Here is a KB article that is an example case of defining a file format and creating an external conversion:

http://www.scootersoftware.com/support.php?zz=kb_externalconversion

[xnappo]

Hi,

Thanks a lot for the quick reply.

I am definitely using Data mode - that is the primary reason I am evaluating your tool! It is critical when diff'ing my data to be able to turn off columns that may have been inserted between revisions (or move them to the end). I think only data mode can do this easily?

As I mentioned, the data has the headers in row 3, not 1, and it would just be nice for it to detect that, though not critical by any means.

Anyhow - I will look into the pre-processing. I assume it also has some means by which that can be un-done at save/merge time... Will read the article

Thanks,
xnappo

[Aaron]

Hello,

There are two processings: one pre-process and one post-process. We have methods of reorganizing, ignoring, or completely hiding columns built into the graphical interface of BC3, but not rows.

For the pre/post-processing, what row data do you have that needs to be stripped out or ignored? One strategy could be to move it to the bottom of the file, and place it in columns "on the far right", then mark those columns as always unimportant.