X-Git-Url: http://lists.indexdata.com/cgi-bin?a=blobdiff_plain;ds=sidebyside;f=doc%2Fyaz-iconv-man.xml;h=d24dff7a01b57702d79b58cb1b6b01b1908b7725;hb=9c0e939209ea21db376876168faa4e6563d3ac16;hp=d8e6d70361e0357be7aee745a0be5fe2465290ba;hpb=49e8c1a3167e2ea965d1787b89e3ac731b590ad9;p=yaz-moved-to-github.git
diff --git a/doc/yaz-iconv-man.xml b/doc/yaz-iconv-man.xml
index d8e6d70..d24dff7 100644
--- a/doc/yaz-iconv-man.xml
+++ b/doc/yaz-iconv-man.xml
@@ -21,7 +21,7 @@
yaz-iconv
- YAZ Charcter set conversion utility
+ YAZ Character set conversion utility
@@ -80,72 +80,103 @@
-
+
ENCODINGS
The yaz-iconv command and the API as defined in
yaz/yaz-iconv.h is a wrapper for the
- library system call iconv. But YAZ' implementation also implements
- conversions on its own. So regardless of iconv, the following
- below are always supported by YAZ:
+ library system call iconv. But YAZ' iconv utility also implements
+ conversions on its own. The table below lists characters sets (or encodings).
+ that are supported by YAZ. Each character set is marked with either
+ encode or decode. If
+ an encoding is encode-enabled YAZ may convert to
+ to the designated encoding. If an encoding is decode-enabled, YAZ
+ may convert from the designated encoding.
marc8 (encode, decode)
+
The MARC8 encoding as defined by
the Library of Congress. Most MARC21/USMARC records usees
this encoding.
+
marc8s (encode, decode)
+
Like MARC8 but with conversion prefers non-combined characters
in the Latin-1 plane over combined characters.
+
marc8lossy (encode)
+
Lossy encoding of MARC-8.
+
marc8lossless (encode)
+
Lossless encoding of MARC8.
+
utf8 (encode, decode)
+
The most commonly used UNICODE encoding on the Internet.
+
iso8859-1 (encode, decode)
+
ISO-8859-1, AKA Latin-1.
+
iso5426 (decode)
+
ISO 5426. Some MARC records (UNIMARC) uses this encoding.
+
iso5428:1984 (encode, decode)
+
ISO 5428:1984.
+
advancegreek (encode, decode)
+
An encoding for Greek used by some vendors (Advance).
+
+
+
+ danmarc (decode)
+
+
+ Danmarc (in danish) is
+ an encoding based on UNICODE which is used for DanMARC2 records.
+
+