in reply to Character issue when pulling data out of MSSQL using DBI

I'm with apl: relying on font glyphs to see and show the data reminds me of the old story about blind men describing an elephant. At the very least, use some non-perl method to store the database field value in question into a plain-text file (without interpretation of any kind), and get a hex dump of the file.

Then, when you get the database field via a DBI query in your perl script, print it like this:

print join(" ", map { sprintf("%04x", ord()) } split //, $field_val +ue ), "\n";
There are other ways to dump a string as hex bytes; there's no reason to be picky about which method you use, so long as the output is complete and accurate.

If the two sets of hex numbers differ, you've got a strange problem (maybe the storage into a plain-text involved some interpretation that you were not aware of). If they are the same, your problem is to figure out why they are showing up differently in one display method vs. another.