Abhinandan Ekande wrote:
> https://cr.opensolaris.org/action/browse/pkg/ae112802/7166082-rev1/webrev/
t_pkg_publisher.py:
- You could use self.cmdline_run() instead of calling subprocess
yourself.
- line 468: Can we do this test against one of the test publishers we set
up? For that matter, do we have to test against the output of "pkg
publisher", as opposed to almost any other pkg command? Or even
against an image without any configured publishers? I don't like the
idea of using a real URI in the testsuite.
- line 478: if re.search() returned None, then eucJP_locale simply
wouldn't exist, and you'd get a NameError exception raised here. Why
not simply assert_ on m, and skip setting eucJP_locale at all? You
only use it on on line 485, but isn't the value always going to be the
same, anyway?
- line 489ff: Why not simply compare the two unicode objects?
A high-level (though relatively minor) concern I have is the requirement
you've made on a 145MB package. True, it's only needed for development,
but it's still pretty big.
I don't have the package installed, but locale -a still lists jp_JP.eucJP
as one of the non-UTF-8 locales I have. Trying it seems to indicate it's
broken, which I suppose is why we need the extra package.
Are any of the non-UTF-8 locales locale -a tells me I have usable for this
test? If not, is there a facet configuration we can use to trim the
installation of system/local/extra down to just what we need for this test?
Thanks,
Danek
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
