Non-basic-source-character-set characters conflict with MSVC

[nobody]

There exist high-ascii characters in the Boost sources
that cannot be converted to the local region for a
given computer because the context of the originating
region of the source is not known. Thus, these
high-ascii characters may not have a proper mapping,
and a warning will be emitted by MSVC 2003 and 2005.
This should be considered a failure case because,
though low, it is still emitting an alert. Thus,
without the correct region set on your system, the
Boost sources fail to build.

[Gennaro Prota]

Logged In: YES 
user_id=624384

I'm surprised that VC8 (2005) still gets this wrong. I've
changed the data-type to "Support Requests" as this is a
compiler problem: non basic source-set characters should be
trivially handled in translation phase 1 :-( Anyhow, could
you please tell us:

* your system region
* the number (Cxxxx) of the warning you get

?

[sf-robot]

Logged In: YES 
user_id=1312539

This Tracker item was closed automatically by the system. It was
previously set to a Pending status, and the original submitter
did not respond within 14 days (the time period specified by
the administrator of this Tracker).

[akalin]

Logged In: YES 
user_id=1578182

I get this same problem.  I'm using the Japanese region and
the warning I get is 4819.  Per the discussion in
http://archives.free.net.ph/message/20060403.042624.2a5fe361.en.html
, I'm sure it's the copyright symbols that are causing this
problem as they don't show up as copyright symbols for me.
(a half-width katakana 'u', if you're curious)

[anonymous]

Same problem here. My system codepage is Chinese(codepage 936), and some header files can't be displayed properly. for example, in utility/enable_if.hpp, the copyright mark and the author's name are not right. This make vc2005 emit lots of warnings(c4819).

I have to surpress the warning manually or change the boost header file myself for successful building. I believe this happens on all Chinese, Japanese, Korean windows system.

I suggest use only basic source set characters in header files. It's simply not possible to display every name correctly unless the file is in unicode.

[Marshall Clow]

This is a duplicate of #1736, and if you look there, you can see progress on this front. We are working to eliminate as many of these warning as possible, however, I am loathe to change author's names in the source. Things like changing © --> (c) are easy, and we're getting those.