Add identify filter for ISO-8859-6 (Latin/Arabic)

Note that some text encoding conversion libraries, such as Solaris iconv
and FreeBSD iconv, map 0x30-0x39 to the Arabic script numerals rather than
the 'regular' Roman numerals. (That is, to Unicode codepoints 0x660-0x669.)

Further, Windows CP28596 adds more mappings to use the unused bytes in
ISO-8859-6.

diff --git a/ext/mbstring/libmbfl/filters/mbfilter_iso8859_6.c b/ext/mbstring/libmbfl/filters/mbfilter_iso8859_6.c
index c010908689ee4355b492c9e3c3a083eeac1d4da8..ead2b49e0b75a2f5d64266b1b4d91484f3484799 100644 (file)
--- a/ext/mbstring/libmbfl/filters/mbfilter_iso8859_6.c
+++ b/ext/mbstring/libmbfl/filters/mbfilter_iso8859_6.c
@@ -31,6 +31,8 @@
 #include "mbfilter_iso8859_6.h"
 #include "unicode_table_iso8859_6.h"
 
+static int mbfl_filt_ident_iso8859_6(int c, mbfl_identify_filter *filter);
+
 static const char *mbfl_encoding_8859_6_aliases[] = {"ISO8859-6", "arabic", NULL};
 
 const mbfl_encoding mbfl_encoding_8859_6 = {
@@ -47,7 +49,7 @@ const mbfl_encoding mbfl_encoding_8859_6 = {
 const struct mbfl_identify_vtbl vtbl_identify_8859_6 = {
        mbfl_no_encoding_8859_6,
        mbfl_filt_ident_common_ctor,
-       mbfl_filt_ident_true
+       mbfl_filt_ident_iso8859_6
 };
 
 const struct mbfl_convert_vtbl vtbl_8859_6_wchar = {
@@ -132,3 +134,11 @@ int mbfl_filt_conv_wchar_8859_6(int c, mbfl_convert_filter *filter)
 
        return c;
 }
+
+static int mbfl_filt_ident_iso8859_6(int c, mbfl_identify_filter *filter)
+{
+       if (c >= 0xA0 && !iso8859_6_ucs_table[c - 0xA0]) {
+               filter->status = 1;
+       }
+       return c;
+}