Bugs item #1738667, was opened at 2007-06-17 07:54 Message generated for change (Comment added) made by msapiro You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=1738667...

Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: configuring/installing Group: 2.1 (stable)

Status: Pending

Resolution: None Priority: 5 Private: No Submitted By: David Chase (dr2chase) Assigned to: Nobody/Anonymous (nobody) Summary: misc/KoreanCodecs-2.0.5/src/hangul.c does not compile

Initial Comment: SunOS ... 5.10 Generic_118855-36 i86pc i386 i86pc Python 2.3.5 (#1, Nov 30 2005, 10:43:01) [C] on sunos5 gcc -v yields: Reading specs from /usr/sfw/lib/gcc/i386-pc-solaris2.10/3.4.3/specs Configured with: /builds/sfw10-gate/usr/src/cmd/gcc/gcc-3.4.3/configure --prefix=/usr/sfw --with-as=/usr/sfw/bin/gas --with-gnu-as --with-ld=/usr/ccs/bin/ld --without-gnu-ld --enable-languages=c,c++ --enable-shared Thread model: posix gcc version 3.4.3 (csl-sol210-3_4-branch+sol_rpath)

--------------

Compilation first fails when it attempts to invoke "cc" despite being configured to use the default (which is gcc). Hacking around this with a symbolic link from cc to gcc yields:

src/hangul.c:33: error: syntax error before numeric constant src/hangul.c:34: error: syntax error before numeric constant

The cause of this is that two of the enums defined in hangul.c

enum { /* Jaeum Codes on U+3100 */ G = 0x3131, GG, GS, N, NJ, NH, D, DD, L, LG, LM, LB, LS, LT, LP, LH, M, B, BB, BS, S, SS, NG, J, JJ, C, K, T, P, H };

preprocess to integers (after gcc -E):

enum { G = 0x3131, GG, 0, N, NJ, NH, D, DD, L, LG, LM, LB, LS, LT, LP, LH, M, B, BB, BS, S, 18, NG, J, JJ, C, K, T, P, H };

Presumably this is specific to the python include file version.

Proposed hackaround, inserted before the enums:

#ifdef GS #undef GS #endif

#ifdef SS #undef SS #endif

It did at least finish the make install after this.

----------------------------------------------------------------------

Comment By: Mark Sapiro (msapiro)

Date: 2007-06-17 08:37

Message: Logged In: YES user_id=1123998 Originator: NO

I believe this is a Solaris bug, not a Mailman bug. See

http://www.google.com/search?hl=en&q=site%3Amail.python.org++inurl%3Amailman++solaris+hangul

including

http://mail.python.org/pipermail/mailman-users/2007-January/055235.html

from which I quote

This is due to a bug in Solaris 10 headers. <signal.h> includes <sys/regset.h> which pollutes the macro "namespace" (if you will) with definitions for x86 registers, like GS and SS. This is only on Solaris/x86; Solaris/sparc doesn't have this error.

Regarding the cc/gcc issue see

http://www.google.com/search?hl=en&safe=off&q=site%3Amail.python.org++inurl%3Amailman++solaris+gcc+cc

including

http://mail.python.org/pipermail/mailman-users/2007-January/055506.html

----------------------------------------------------------------------

You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=1738667...