community/pdns-recursor: upgrade to 4.5.6

community/pdns-recursor/APKBUILD

 # Maintainer: Peter van Dijk <peter.van.dijk@powerdns.com>
 _pkgname=pdns-recursor # upstream package name
 pkgname=pdns-recursor
-pkgver=4.5.5
-pkgrel=1
+pkgver=4.5.6
+pkgrel=0
 pkgdesc="PowerDNS Recursive Server"
 url="https://www.powerdns.com/"
 arch="all !s390x !mips64" # broken context
@@ -14,7+14,7 @@
 pkgusers="recursor"
 pkggroups="recursor"
 source="https://downloads.powerdns.com/releases/$_pkgname-$pkgver.tar.bz2
 	$pkgname-GPGKEYS::https://www.powerdns.com/powerdns-keyblock.asc
 	$_pkgname.initd
 	recursor.conf
 	"
@@ -82,7+82,7 @@
 	"
 
 sha512sums="
-7df2e89eea55e870724370753b39d867e2b32fc8bcbdb3efaefa824917b16b9d382817dce7d551dc16d8ad0b0c04b3021cba310b47de969fe54c395eafe7706b  pdns-recursor-4.5.5.tar.bz2
-7027c12c450ecdf4fa569e882eebdb437f166c71090f867928fddb08c558df452922ee3b33fa1dad3097b322bdb79a2096e07d9d2e9f15423464a64df6a9cde3  pdns-recursor-GPGKEYS
+43c866a29bd72175bad71450dc20678183f6b11f60d54a174627501be52c79dc91819196be63e6859af086f42cc7297d7a3a6100088a2efe3c99eb8f5fd73e71  pdns-recursor-4.5.6.tar.bz2
+90ec29026d5457bb420d2477a5a9ec8083e4f09b6924c451cd02679fbaf6ea0c7a67aa91999d0dc5c98f95d8adffd3c3dc51138ea7e330825032329f6c75fd4d  pdns-recursor-GPGKEYS
 71f7be51d3a9a78f22f3946e95534d4f5b37b3bcbd1891ad3a17095be545fdb0eaf239d58c9f40ccc642542150c0b35ed025bf69acabfa779c0ab10c67a5e796  pdns-recursor.initd
 9eb82b86ba75af71ef49a510f087bbefaaacafb4c06dde9124c0fa74b27456d59dd246cc8bffd35394fcf4840995375c9004827bbc3b527d419f54fd65bf632a  recursor.conf
 "

[Bart Ribbers]

The problem here is that the hash of this file changed, but the exact same file already existed on the builders. This is why files have version in their names normally, that way they can never interfere.

If you rename it to e.g. $pkgname-$pkgver-GPGKEYS then you wouldn't have this problem, but preferably upstream starts versioning that file rather than just replacing it for a new one every once and while.

[Bart Ribbers]

@Habbie please use the Gitlab threads functionality to respond and keep context obvious.

What is the key used for anyway? If it's just to check the source tarball hasn't been tampered with, then yes that's what the hashing check is for and you can just get rid of the file.

[Peter van Dijk]

Apologies, I'm not used to Gitlab. Will try to keep it in mind.

As for what it's used for, I'm unsure. I also noticed shellcheck complaining on the MR linter pipeline: _gpg_signature_extensions appears unused. Verify use (or export if used externally).

[Bart Ribbers]

Yeah I'm unsure why those variables are there.

The file doesn't seem to be used, let's just get rid of it (and also those gpg signature variables) and see if it builds as is. Please make a new MR for it if it does

[Peter van Dijk]

Will do!