Dumb question about utf-8 handling

Been using marc.pm for some time to do a "bunch-o-stuff," but haven't 
had to deal with unicode problems until now.  Couple of basic questions.

1.  If I use the "$record->encoding()" call, is it  simply checking for 
the "a" in leader position 9?  That is,  it is not doing anything more 
complicated, like checking the actual data.

(Reason for the question, if leader position 9 is wrong, and encoding 
depends on it, then encoding will not tell me anything more than marc.pm 
thinks that it is unicode.)

2.  If I want to update a record that is unicode do I have to make sure 
any text string that I send in say, append_fields(... has been encoded 
as utf8, i.e., use some perl function to ensure that the string(s) are 
utf8?  Alternatively, does marc.pm handle that automatically when it 
thinks it has a utf8 record?

(This problem comes from working with a set of records.  When I update 
those that have leader byte 9 set to "a," and I try to do something with 
the updated record, e.g., read it,  I crash with the  "does not map to 
..." message.  That leads me to assume that either I need to do 
something with the stuff I am adding to make it utf8 or I have records 
that say they are utf8 and are not.  Running some of the records through 
a little check program I have, usually finds a problem with the record 
length when it hits these records that were changed.)

-- 
Gene Damon
Director of Library Automation &
Learning Resources
Virginia Community College System
(804) 819-4981 (voice)
(804) 819-4773 (fax)
[email protected]
Address:
James Monroe Bldg.
15th Floor
101 N. 14th St.
Richmond, VA 23219

http://library.vccs.edu/reference