diff options
author | Matthias Clasen <mclasen@redhat.com> | 2022-09-14 14:55:43 -0400 |
---|---|---|
committer | Marco Trevisan (TreviƱo) <mail@3v1n0.net> | 2022-09-15 03:43:04 +0200 |
commit | dcb459a0b0d3a1065735fed1d1a1cc9cec545d23 (patch) | |
tree | 6d650bf7135347374e9411237d092d661a5b8692 | |
parent | b3e32eda7ca282c6d429d076995eae470b02d707 (diff) | |
download | glib-dcb459a0b0d3a1065735fed1d1a1cc9cec545d23.tar.gz |
Fix g_unichar_iswide for unassigned codepoints
There are a few blocks in Unicode (mainly ideographs)
which default to wide. These blocks are defined in the
header comment of EastAsianWidth.txt.
We have some tests which check that unassigned codepoints
in those blocks get reported as wide, so make sure we handle
this correctly.
-rw-r--r-- | glib/guniprop.c | 21 |
1 files changed, 19 insertions, 2 deletions
diff --git a/glib/guniprop.c b/glib/guniprop.c index 3744cd1b6..d1363e546 100644 --- a/glib/guniprop.c +++ b/glib/guniprop.c @@ -477,6 +477,14 @@ g_unichar_iswide_bsearch (gunichar ch) return FALSE; } +static const struct Interval default_wide_blocks[] = { + { 0x3400, 0x4dbf }, + { 0x4e00, 0x9fff }, + { 0xf900, 0xfaff }, + { 0x20000, 0x2fffd }, + { 0x30000, 0x3fffd } +}; + /** * g_unichar_iswide: * @c: a Unicode character @@ -491,8 +499,17 @@ g_unichar_iswide (gunichar c) { if (c < g_unicode_width_table_wide[0].start) return FALSE; - else - return g_unichar_iswide_bsearch (c); + else if (g_unichar_iswide_bsearch (c)) + return TRUE; + else if (g_unichar_type (c) == G_UNICODE_UNASSIGNED && + bsearch (GUINT_TO_POINTER (c), + default_wide_blocks, + G_N_ELEMENTS (default_wide_blocks), + sizeof default_wide_blocks[0], + interval_compare)) + return TRUE; + + return FALSE; } |