This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Ada] Correct failure to recognize UTF_32 BOM


Tested on i686-linux, committed on trunk

As mentioned on gcc-patches after another patch from g-byorma.adb,
The order of tests in GNAT.Byte_Order_Mark was incorrect,
which resulted in a situation where a UTF_32 little endian
sequence was always mis-interprted as UTF-16 (little-endian).

The following is a test program:

1234    package bom32 is end;

Here the first four characters of the file are a UTF-32
(little-endian) BOM sequence 16#FF# 16#FE# 16#00# 16#00#.
Compiling this example should give:

UTF-32 encoding format not recognized
compilation abandoned

Before this patch, the output was:

UTF-16 encoding format not recognized
compilation abandoned
16#

2008-03-26  Robert Dewar  <dewar@adacore.com>

	* g-byorma.adb (Read_BOM): Reorder tests so that UTF_32 is recognized

Attachment: difs
Description: Text document


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]