Issue 831574: Solaris term.h needs curses.h (original) (raw)

This issue has been migrated to GitHub: https://github.com/python/cpython/issues/39465

classification

Title: Solaris term.h needs curses.h
Type: behavior Stage: resolved
Components: Build Versions:

process

Status: closed Resolution: rejected
Dependencies: Superseder:
Assigned To: anthonybaxter Nosy List: ajaksu2, akuchling, anthonybaxter, loewis, nnorwitz
Priority: normal Keywords:

Created on 2003-10-28 06:53 by anthonybaxter, last changed 2022-04-11 14:56 by admin. This issue is now closed.

Files
File name Uploaded Description Edit
configure-solaris-term.h anthonybaxter,2003-10-28 06:53 configure.in and configure patch
Messages (7)
msg18778 - (view) Author: Anthony Baxter (anthonybaxter) (Python triager) Date: 2003-10-28 06:53
Solaris' term.h requires curses.h to be included first. This causes the configure script to emit lines about a bug in autoconf. From the autoconf mailing lists, their standard response is to fix the configure script, see e.g. http://mail.gnu.org/archive/html/bug-autoconf/2003-05/msg00118.html The following patch against 2.3 branch for configure and configure.in makes things a bit happier. Note that Include/py_curses.h already includes curses.h before term.h, this just fixes the breakage of configure.
msg18779 - (view) Author: Neal Norwitz (nnorwitz) * (Python committer) Date: 2003-10-28 13:08
Logged In: YES user_id=33168 Should the #if be an #ifdef ? Looks fine to me, but I don't know much about autoconf. :-) I think Martin is the expert. Martin do you have an opinion?
msg18780 - (view) Author: Anthony Baxter (anthonybaxter) (Python triager) Date: 2003-10-29 01:38
Logged In: YES user_id=29957 Dunno if #ifdef is better or not - I just worked from the example in the attached autoconf mailing list message.
msg18781 - (view) Author: Martin v. Löwis (loewis) * (Python committer) Date: 2003-10-31 15:22
Logged In: YES user_id=21627 I find it confusing that the test for curses.h already refers to HAVE_CURSES_H; I think you should first check for curses.h, and then use HAVE_CURSES_H in the test for term.h I also agree that #ifdef is better than #if, even though it should not matter in an ISO C compiler (which replaces undefined symbols by 0 in an #if).
msg18782 - (view) Author: A.M. Kuchling (akuchling) * (Python committer) Date: 2007-01-05 14:33
Is this bug still relevant to Python 2.5?
msg82017 - (view) Author: Daniel Diniz (ajaksu2) * (Python triager) Date: 2009-02-14 11:36
Is the configure script still emitting lines about a bug in autoconf on Solaris?
msg87641 - (view) Author: Daniel Diniz (ajaksu2) * (Python triager) Date: 2009-05-12 17:08
Closing on lack of feedback.
History
Date User Action Args
2022-04-11 14:56:00 admin set github: 39465
2009-05-12 17:08:40 ajaksu2 set status: pending -> closedresolution: rejectedmessages: + stage: resolved
2009-04-05 18:25:41 georg.brandl set status: open -> pending
2009-02-14 11:36:06 ajaksu2 set nosy: + ajaksu2type: behaviormessages: +
2003-10-28 06:53:44 anthonybaxter create