Data Compare Delimiter Issue

[Gibbers]

I'm comparing pipe delimited files. I have filesets in different languages, English, Portuguese, and Spanish. When I get to the Spanish fileset, the column breakup just doesn't work. One file will be broken up into columns. The other will have all the data in column1.

I've tried forcing it to be UTF8 using iconv before I pull the files, but it still breaks.

[Aaron]

Could you email a pair of sample files to [email protected] ? Please include a link back to this forum post for our reference. A full screen screenshot of the comparison would also be useful.

[Gibbers]

I can't easily because the data is confidential.

It looks like it always happens on a semicolon. The left file splits fine by |.

The right file throws everything in column 1, column 2 is anything after a semicolon. Example:

....elephant; loudest.... is the first place it breaks.


Everything before ; is in column 1. Everything after ; is in column 2.

[Aaron]

Hello,

In the specific Format (click the Format button, or go to Tools menu -> File Formats, and select the specific format), in the Type tab, is it selected to "Detect" delimiter? You might need to configure or create a New Data format that is manually set to delimit on |, instead of Detect.

[Gibbers]

That fixed it. Thanks. Odd that the left file delimited by pipe and the right one by ;.

[Aaron]

The detection algorithm looks through the line and tries to determine what the delimiter is. Since left and right files are potentially different, each is scanned uniquely unless otherwise configured. We try out best to auto-detect, but some cases need to be defined manually if this auto-detection fails.

[ingo.schaa]

I tried to create a New Data format that is manually set to delimit on a special character (ASCII code = 135) using windows mechanism : pressing ALT key and entering ASCII code at numerical keypad = 135.
This works fine with character "p" = 112 but fails for my required delimiter = "Double Dagger" = code 135.
Also Copy+Paste of my special character does not work.
How can I tell Beyond Compare to delimit on special ASCII codes?

(I am using commerical version of Beyond Compare V3.3.4)

[Aaron]

Hello,

BC3 does not currently support this type of delimiter. I've added this to our Customer Wishlist. In the meantime, as a workaround you would need to perform a find/replace conversion and switch the delimiter character to one that is supported.