On Oct 8, 2017, at 06:50, Marius Schamschula wrote:
> On Oct 7, 2017, at 7:15 PM, Ryan Schmidt wrote:
>
>> On Oct 6, 2017, at 20:34, Marius Schamschula wrote:
>>
>>> -- a/science/gmt5/Portfile
>>> +++ b/science/gmt5/Portfile
>>> @@ -5,9 +5,11 @@ PortGroup cmake 1.0
>>>
>>> name gmt5
>>> version 5.4.2
>>> +revision 1
>>> categories science
>>> platforms darwin
>>> -maintainers takeshi
>>> +maintainers eprofs.de:florian \
>>> + takeshi
>>> license GPL-3
>>> description The Generic Mapping Tools
>>> long_description GMT is an open source collection of ~120 tools \
>>> @@ -28,6 +30,10 @@ master_sites ftp://ftp.soest.hawaii.edu/gmt
>>> \
>>> use_xz yes
>>> distname gmt-${version}
>>> distfiles ${distname}-src${extract.suffix}
>>> +
>>> +# stealth update
>>> +dist_subdir ${name}/${version}_1
>>> +
>>> checksums rmd160 ed90b3b46aeb98b7d5ed9bb0c68fbf7ebf7df929 \
>>> sha256
>>> ddcd63094aeda5a60f541626ed7ab4a78538d52dea24ba915f168e4606e587f5
>>
>> This doesn't look intentional. It re-adds Florain as maintainer (you
>> deliberately removed him in
>> https://github.com/macports/macports-ports/commit/7d2e2673c27b3cbbf9266dd1a4e49f77682ef33f)
>> and changes the dist_subdir even though the checksums were not changed, so
>> there wasn't a stealth update.
>
> Fixed in
> https://github.com/macports/macports-ports/commit/c26a72b368f08c70053af3414c6406537bda1738
I don't understand "Leave unneeded stealth update in place, as it forced
rebuild for ghostscript 8.22". You can and should leave the "revision 1" line
in place, but there's no reason to leave the "dist_subdir ${name}/${version}_1"
line in the portfile.
