Better interface for selection of encoding.

diff --git a/encoding.h b/encoding.h
index 287e478e9bd636747debf02f06d736e713227b7a..12473d241e902b38146bcfc15d61387ece001f23 100644 (file)
--- a/encoding.h
+++ b/encoding.h
@@ -1,5 +1,17 @@
+/* Basic file encoding */
+#ifndef __ENCODING_H
+#define __ENCODING_H
+
+typedef enum _ENC {
+  ONE_BYTE = 0,
+  TWO_BYTE_HILO = 1,
+  TWO_BYTE_LOHI = 2
+} ENCODING;
+
 int open_conv_to_internal(char* fromcode);
 void close_conv_to_internal();
 char* to_internal(char* str, size_t len);
 void init_encodings();
-char* get_encoding(char* gedcom_name);
+void set_encoding_width(ENCODING enc);
+
+#endif /* __ENCODING_H */

diff --git a/gedcom.enc b/gedcom.enc
index 8c2a56b1d0b7de84c0b6a570518375bed22ec160..b47dc565fde1b977f0f6281ff4c8995ab83f5a43 100644 (file)
--- a/gedcom.enc
+++ b/gedcom.enc
@@ -1,8 +1,11 @@
 # Mapping of charsets for gedcom parsing
-# Each line contains the gedcom name, appended with (LOHI) or (HILO)
-# for 2 byte encodings, and the iconv name of the charset, separated
-# by whitespace
-UNICODE(LOHI)  UTF16LE
-UNICODE(HILO)  UTF16BE
-ASCII          ASCII
-ANSI           CP1252
+# Each line contains (separated by whitespace):
+#   - the gedcom name
+#   - a token identifying the width of characters and the ordering;
+#     currently supported values: 1, 2_LOHI, 2_HILO
+#   - the iconv name of the charset
+UNICODE                2_LOHI  UTF16LE
+UNICODE                2_HILO  UTF16BE
+ASCII          1       ASCII
+ANSI           1       CP1252
+ANSEL          1       ANSEL