git: 67f9a7c0620e - main - graphics/kimageannotator: fix plist

[Adriaan de Groot]

On Friday, 4 June 2021 12:04:44 CEST you wrote:
>     graphics/kimageannotator: fix plist
> 
>     PR:             256406
>     Reported by:    jbeich

I almost always build with `bulk -t` in poudriere, on the assumption that that 
applies testing to the ports that get built. For the update to kImageAnnotator 
in question, I ran

```
poudriere bulk -j 13amd64 -p area51 -t graphics/ksnip graphics/spectacle
```

.. because those were the two consumers that I bumped for the same commit 
which updated graphics/kimageannotator. I just tried it again (with `-C` to 
force those two consumers to be rebuilt; I needn't have bothered, because 
other updates lower in the chain triggered lots of rebuilding). As output from 
this particular build, I get

```
[00:21:41] [01] [00:01:38] Finished graphics/kimageannotator | 
kImageAnnotator-0.5.1: Success
```

If I re-do it, naming kimageannotator explicitly (again with `-C`) **then** it 
fails. Similarly, testport passes graphics/kimageannotator when it's a 
dependency, not when named on its own.

The manpage explains that `bulk -t` applies only to the named ports, not the 
dependencies (and also turns off parallel build and twiddles other things). So 
it comes down to my assumptions about what pou would find with `-t` didn't 
match reality.

[ade]
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.freebsd.org/pipermail/dev-commits-ports-all/attachments/20210604/05e0bc16/attachment.sig>