Re: [Tinycc-devel] Better support of universal-character-names in strings

[Christian Jullien]

Hi Petr,

Is there a reason why you removed \n in 97 tests?
Without it this test fails on some systems we are working on, like OpenBSD.

Can you please confirm and apply this patch which fix the issue on OpenBSD?

diff --git a/tests/tests2/97_utf8_string_literal.c 
b/tests/tests2/97_utf8_string_literal.c
index c7e8998..c8d7f6e 100644
--- a/tests/tests2/97_utf8_string_literal.c
+++ b/tests/tests2/97_utf8_string_literal.c
@@ -15,5 +15,6 @@ int main()
     wchar_t s[] = 
L"hello$$<E4><BD><A0><E5><A5><BD><C2><A2><C2><A2><E4><B8><96><E7><95><8C><E2><82><AC><E2><82><AC>world";
     wchar_t *p;
     for (p = s; *p; p++) printf("%04X ", (unsigned) *p);
+    printf("\n");
     return 0;
 }

-----Original Message-----
From: Tinycc-devel [mailto:tinycc-devel-bounces+eligis=orange.fr@nongnu.org] On 
Behalf Of Petr Skocik
Sent: Monday, January 18, 2021 01:14
To: tinycc-devel@nongnu.org
Subject: [Tinycc-devel] Better support of universal-character-names in strings

Hi, I've pushed a patch improving handling of universal character names
in strings.

char cau_svete_ucns[]="\u010dau_sv\u011bte";

is now equal to

char cau_svete_utf8[]="čau_světe";

C>=C99 also requires support for such escape sequences even in
identifiers (the support is indeed present in gcc and clang), requiring
compilability  even for things like `char \u010dau_sv\u011bte[];`. The
last bit I haven't done.

Cheers,

Petr Skocik


_______________________________________________
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
https://lists.nongnu.org/mailman/listinfo/tinycc-devel