<farkuhar>
SiFuh__: the prohibition on core ports in the dependency header only applies if the core port is not dynamically linked in. Otherwise it's okay to list them, since dynamic linking should be taken into account when sorting the targets of an update transaction. See Pkgfile(5) for examples where a dependency on zlib is okay, but a dependency on perl is not.
<farkuhar>
Hence the recent commit "bftpd: 6.2 -> 6.3" added back the "Depends on" header. In case somebody is performing a sysup on a system with out-of-date zlib *and* bftpd, we don't want their sysup to start working on bftpd until after zlib has been updated.
<farkuhar>
The appearance of libpcre2 in the dependency list for opt/fdm would also be permissible, if jue had bothered to pass --enable-pcre2 to the configure script. By omitting that flag, opt/fdm gets configured without any linking to libpcre2 (on CRUX-glibc), or suffers a build failure (on CRUX-musl).
SiFuh has quit [Remote host closed the connection]
SiFuh has joined #crux-musl
ukky has quit [Ping timeout: 252 seconds]
ukky has joined #crux-musl
dlcusa has quit [Remote host closed the connection]
<emmett1>
farkuhar: on my CRUX-MUSL-CLANG system theres no issue about 'prt-get dependent' and aliases as you said
<emmett1>
so far prt-get works fine
<emmett1>
same goes with 'prt-get {depends,deptree,quickdep}'
<farkuhar>
15:40 <emmett> yes, but it does not work for me. 'prt-get dependent <any pkg>' prints nothing
<emmett1>
yeah. its a false alarm.
<emmett1>
i guess i forgot how prt-get works
<emmett1>
'prt-get dependent' only applied to installed pkgs
zorz_ has joined #crux-musl
zorz_ has quit [Quit: leaving]
zorz_ has joined #crux-musl
<remiliascarlet>
farkuhar: My "suggestion" to replace GCC with TCC was just a joke, although it's a way more lightweight compiler, but lots of ports will be rendered unable to be compiled.
<SiFuh_>
remiliascarlet: I was thinking what if we made crux-musl an independent fork? No longer relies on CRUX itself?
<SiFuh_>
Complete restructuring of core and all the ports. And write a decent init system.
<remiliascarlet>
I'm actually all for it to be honest.
<SiFuh_>
Restructing would also mean, possible moving core to /, opt /usr and contrib /usr/local/
<SiFuh_>
And *-bin builds /usr/opt
<SiFuh_>
And a handbook emphasizing the use of seperate partitions for each.
<remiliascarlet>
And also add in properly written manpages.
<SiFuh_>
remiliascarlet: Oh yeah. What a job that would be.
<remiliascarlet>
Maybe also replace the GNU coreutils with ones that don't such.
<SiFuh_>
Well, I wouldn't mind a list in the future if I do attempt this
<SiFuh_>
remiliascarlet: farkuhar: could write the handbook and under ports it could say Rule 1) Do not beerman the system
<SiFuh_>
Hahaha
<remiliascarlet>
Rule 2) Don't lavaball the women.
<SiFuh_>
Heh
<SiFuh_>
He still hasn't got over the 'jap' that banned him
<SiFuh_>
Blames me for it to allowing you to be an op
<SiFuh_>
Bahaha
<SiFuh_>
Like I can police the future
zorz_ has quit [Quit: leaving]
dlcusa_ has joined #crux-musl
dlcusa_ is now known as dlcusa
zorz_ has joined #crux-musl
<zorz_>
falken folker fucker farkuhar
<SiFuh_>
You forgot one
<SiFuh_>
fuck_u_hard
<zorz_>
hahahah
<zorz_>
fuck_u_hurd
<SiFuh_>
fakQ
<farkuhar>
SiFuh_: remiliascarlet: An independent fork will require an investment in infrastructure, to host the new ports collections and to communicate our goals. We'd probably have to obtain a new domain name. Does either of you have a VPS that can be used to serve up such content?
<SiFuh_>
farkuhar: No, just use github
<SiFuh_>
Or as I would, gitlab
<farkuhar>
Well okay, if that's the shortcut you want to take. No objections here.
<ukky>
I can upload a few files onto my VPS. My space is limited though.