From acfebb41ddea04ded300bc1c1aedbf7ebebcf276 Mon Sep 17 00:00:00 2001 From: naruse Date: Sun, 12 Mar 2017 17:22:20 +0000 Subject: merge revision(s) 57816,57817: [Backport #13292] fix UTF-32 valid_encoding? * enc/utf_32be.c (utf32be_mbc_enc_len): check arguments precisely. [ruby-core:79966] [Bug #13292] * enc/utf_32le.c (utf32le_mbc_enc_len): ditto. * regenc.h (UNICODE_VALID_CODEPOINT_P): predicate for valid Unicode codepoints. fix UTF-32 valid_encoding? * test/ruby/test_io_m17n.rb (TestIO_M17N#test_puts_widechar): do not use invalid codepoint. [ruby-core:79966] [Bug #13292] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_4@57935 b2dd03c8-39d4-4d8f-98ff-823fe69b080e --- enc/utf_32be.c | 18 +++++++++++++++--- enc/utf_32le.c | 18 +++++++++++++++--- 2 files changed, 30 insertions(+), 6 deletions(-) (limited to 'enc') diff --git a/enc/utf_32be.c b/enc/utf_32be.c index 995c9d8ed5..17841e52a4 100644 --- a/enc/utf_32be.c +++ b/enc/utf_32be.c @@ -30,11 +30,23 @@ #include "regenc.h" #include "iso_8859.h" +static OnigCodePoint utf32be_mbc_to_code(const UChar* p, const UChar* end, OnigEncoding enc); static int -utf32be_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e ARG_UNUSED, - OnigEncoding enc ARG_UNUSED) +utf32be_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e, + OnigEncoding enc) { - return 4; + if (e < p) { + return ONIGENC_CONSTRUCT_MBCLEN_INVALID(); + } + else if (e-p < 4) { + return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(4-(int)(e-p)); + } + else { + OnigCodePoint c = utf32be_mbc_to_code(p, e, enc); + if (!UNICODE_VALID_CODEPOINT_P(c)) + return ONIGENC_CONSTRUCT_MBCLEN_INVALID(); + return ONIGENC_CONSTRUCT_MBCLEN_CHARFOUND(4); + } } static int diff --git a/enc/utf_32le.c b/enc/utf_32le.c index e255f0e246..18b798f102 100644 --- a/enc/utf_32le.c +++ b/enc/utf_32le.c @@ -30,11 +30,23 @@ #include "regenc.h" #include "iso_8859.h" +static OnigCodePoint utf32le_mbc_to_code(const UChar* p, const UChar* end, OnigEncoding enc); static int -utf32le_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e ARG_UNUSED, - OnigEncoding enc ARG_UNUSED) +utf32le_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e, + OnigEncoding enc) { - return 4; + if (e < p) { + return ONIGENC_CONSTRUCT_MBCLEN_INVALID(); + } + else if (e-p < 4) { + return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(4-(int)(e-p)); + } + else { + OnigCodePoint c = utf32le_mbc_to_code(p, e, enc); + if (!UNICODE_VALID_CODEPOINT_P(c)) + return ONIGENC_CONSTRUCT_MBCLEN_INVALID(); + return ONIGENC_CONSTRUCT_MBCLEN_CHARFOUND(4); + } } static int -- cgit v1.2.3