NetBSD curses support improvements

[0-wiz-0]

BPO	21457
Nosy	@jcea, @0-wiz-0, @serhiy-storchaka
Superseder	bpo-31891: Make curses compiling on NetBSD 7.1 and tests passing
Files	curses.diff: Patch fixing curses module build on NetBSD

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2017-11-04.08:54:41.909>
created_at = <Date 2014-05-09.08:58:00.456>
labels = ['extension-modules', 'build']
title = 'NetBSD curses support improvements'
updated_at = <Date 2017-11-04.08:54:41.907>
user = 'https://github.com/0-wiz-0'

bugs.python.org fields:

activity = <Date 2017-11-04.08:54:41.907>
actor = 'serhiy.storchaka'
assignee = 'none'
closed = True
closed_date = <Date 2017-11-04.08:54:41.909>
closer = 'serhiy.storchaka'
components = ['Extension Modules']
creation = <Date 2014-05-09.08:58:00.456>
creator = 'wiz'
dependencies = []
files = ['35193']
hgrepos = []
issue_num = 21457
keywords = ['patch']
message_count = 6.0
messages = ['218140', '218150', '218201', '305393', '305399', '305546']
nosy_count = 3.0
nosy_names = ['jcea', 'wiz', 'serhiy.storchaka']
pr_nums = []
priority = 'normal'
resolution = 'out of date'
stage = 'resolved'
status = 'closed'
superseder = '31891'
type = 'compile error'
url = 'https://bugs.python.org/issue21457'
versions = ['Python 3.5']

[0-wiz-0]

NetBSD's curses headers have different include guards than ncurses.
Also, the NetBSD curses library has been improved and some workaround are no longer necessary.
Diff against hg attached.

[jcea]

Thomas, What happens if Python is compiled in an old version of NetBSD?

[0-wiz-0]

Thanks for your reply!
I've checked:
keypad in NetBSD curses was fixed in 2009. All supported NetBSD release (5.x and 6.x) have the fix.
nodelay and keyname were fixed even earlier, in 2003.

So this is no problem for older NetBSD releases.

[serhiy-storchaka]

Guards around functions have been removed in bpo-31891.

But could you please explain why changes in py_curses.h are needed?

[0-wiz-0]

Thanks for looking at this.
I looked at the patch again, and I can't make sense of the py_curses.h part either - I've removed it from pkgsrc with a request for information if it's still needed.

Btw, thanks for working on NetBSD curses support in Python. You should really try out on a NetBSD 8 (or -current) snapshot as well, since the curses library was improved for better support especially for python curses in the last year.

[serhiy-storchaka]

I see, the curses library takes a large progress. On NetBSD 8 just two previously tests should be fixed (bpo-31924). Many previously skipped tests are passed now.