On Tue, Apr 19, 2016 at 10:57:25AM -0400, John Ferlan wrote:
On 04/19/2016 10:48 AM, Michal Privoznik wrote:
> On 19.04.2016 16:38, John Ferlan wrote:
>>
>>
>> On 04/19/2016 09:50 AM, Michal Privoznik wrote:
>>> Our uninstall script is not exact counterpart of install one.
>>> Therefore we are leaving couple of files behind. This should not
>>> happen.
>>>
>>> Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com>
>>> ---
>>> docs/Makefile.am | 6 ++++++
>>> 1 file changed, 6 insertions(+)
>>>
>>
>> At 'install-data-local:', there's a :
>>
>> $(mkinstalldirs) $(DESTDIR)$(HTML_DIR)
>>
>> why not just the far more all encompassing:
>>
>> rm -rf $(DESTDIR)$(HTML_DIR)
>>
>> and
>>
>> rm -rf $(DESTDIR)$(DEVHELP_DIR)
>>
>> Rather than picking each part we install to uninstall? and missing
>> something in the future or even now. Do the 'html' or
'internals'
>> directories gets removed? And then of course the toplevel directory
>> which we created.
>>
>> IOW: There's no corollary for the:
>>
>> $(mkinstalldirs) $(DESTDIR)$(HTML_DIR)
>> $(mkinstalldirs) $(DESTDIR)$(HTML_DIR)/html
>> $(mkinstalldirs) $(DESTDIR)$(HTML_DIR)/internals
>> $(mkinstalldirs) $(DESTDIR)$(DEVHELP_DIR)
>>
>>
>
> Yeah. That's the other way of doing that. It's just that if users put
> anything in $(DESTDIR)$(HTML_DIR) it will be removed by uninstall. But I
> can propose v2 if you want.
>
I see there are other 'rf -rm' usages in other "clean" labels...
I don't have a strong feeling either way - perhaps there's other
opinionated folks that would like to chime in. If no one chimes in,
then I'm OK with what's here...
rm -rf is fine with me and I believe with others as well, so ACK from me.
I also now see there's "-rm " usages - so looks like it
makes my comment
in 1/4 unnecessary. It just looked strange to me...
I think '-' means "don't error out if this fails", similarly to
'@'
meaning "Don't print the command being ran" IIRC.
John
--
libvir-list mailing list
libvir-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list