[prev in list] [next in list] [prev in thread] [next in thread] 

List:       macports-dev
Subject:    Re: Just another heads up / RFC --> Re: new naming scheme for llvm/clang will cause wreckage fyi
From:       Christopher Jones <jonesc () hep ! phy ! cam ! ac ! uk>
Date:       2020-04-23 19:26:00
Message-ID: F88217AF-1F61-45B5-9B36-670DECBFA891 () hep ! phy ! cam ! ac ! uk
[Download RAW message or body]

[Attachment #2 (multipart/alternative)]


Hi,

> On 23 Apr 2020, at 8:15 pm, Ken Cunningham <ken.cunningham.webuse@gmail.com> wrote:
> 
> The suggested renaming of the clang/llvm/lldb ports to 
> 
> clang-10
> 
> instead of 
> 
> clang-10.0
> 
> going forward from clang-10 onwards will have repercussions in a number of places, \
> should it proceed, as per the PR. 
> For one, this does not work any more:
> 
> configure.compiler=macports-clang-10


That is not at all unexpected. The PR

https://github.com/macports/macports-ports/pull/6902 \
<https://github.com/macports/macports-ports/pull/6902>

just adds the new compiler as a new port, so does not update any of the port groups \
to add the new compiler to the required lists, to make it available as an option, \
e.g.

https://github.com/macports/macports-ports/blob/master/_resources/port1.0/group/compilers-1.0.tcl \
<https://github.com/macports/macports-ports/blob/master/_resources/port1.0/group/compilers-1.0.tcl>


Not that I am suggesting the above PR should change the above PG. I would suggest \
making those changes as a separate change, following the commit of the compiler \
itself. Just adding the new compiler cannot cause any wreckage, as nothing will use \
it, in the first instance.

Note adding a new compiler to the various places it needs to be done in the port \
groups is not hard. Its a little more complicated this time due to the chance in \
numbering scheme, but still should be possible.

So I would propose first concentrating on just committing the llvm-10 suite of ports, \
and worry about the integrating into the macports build system later on.

cheers Chris

> 
> and that is no doubt just the beginning.
> 
> It was mentioned that clang-10.0 would also cause wreckage due to the 2-digit \
> version number, and that could be so -- I am not completely sure how base currently \
> sets that up. 
> Best,
> 
> Ken


[Attachment #5 (unknown)]

<html><head><meta http-equiv="Content-Type" content="text/html; \
charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: \
space; line-break: after-white-space;" class="">Hi,<br class=""><div><br \
class=""><blockquote type="cite" class=""><div class="">On 23 Apr 2020, at 8:15 pm, \
Ken Cunningham &lt;<a href="mailto:ken.cunningham.webuse@gmail.com" \
class="">ken.cunningham.webuse@gmail.com</a>&gt; wrote:</div><br \
class="Apple-interchange-newline"><div class=""><div class="">The suggested renaming \
of the clang/llvm/lldb ports to <br class=""><br class="">clang-10<br class=""><br \
class="">instead of <br class=""><br class="">clang-10.0<br class=""><br \
class="">going forward from clang-10 onwards will have repercussions in a number of \
places, should it proceed, as per the PR.<br class=""><br class="">For one, this does \
not work any more:<br class=""><br class="">configure.compiler=macports-clang-10<br \
class=""></div></div></blockquote><div><br class=""></div><div><br \
class=""></div>That is not at all unexpected. The PR</div><div><br \
class=""></div><div><a href="https://github.com/macports/macports-ports/pull/6902" \
class="">https://github.com/macports/macports-ports/pull/6902</a></div><div><br \
class=""></div><div>just adds the new compiler as a new port, so does not update any \
of the port groups to add the new compiler to the required lists, to make it \
available as an option, e.g.</div><div><br class=""></div><div><a \
href="https://github.com/macports/macports-ports/blob/master/_resources/port1.0/group/compilers-1.0.tcl" \
class="">https://github.com/macports/macports-ports/blob/master/_resources/port1.0/group/compilers-1.0.tcl</a></div><div><br \
class=""></div><div>Not that I am suggesting the above PR should change the above PG. \
I would suggest making those changes as a separate change, following the commit of \
the compiler itself. Just adding the new compiler cannot cause any wreckage, as \
nothing will use it, in the first instance.</div><div><br class=""></div><div>Note \
adding a new compiler to the various places it needs to be done in the port groups is \
not hard. Its a little more complicated this time due to the chance in numbering \
scheme, but still should be possible.</div><div><br class=""></div><div>So I would \
propose first concentrating on just committing the llvm-10 suite of ports, and worry \
about the integrating into the macports build system later on.</div><div><br \
class=""></div><div>cheers Chris</div><div><br class=""></div><div><blockquote \
type="cite" class=""><div class=""><div class=""><br class="">and that is no doubt \
just the beginning.<br class=""><br class="">It was mentioned that clang-10.0 would \
also cause wreckage due to the 2-digit version number, and that could be so -- I am \
not completely sure how base currently sets that up.<br class=""><br \
class="">Best,<br class=""><br class="">Ken</div></div></blockquote></div><br \
class=""></body></html>


["smime.p7s" (smime.p7s)]

0	*H
 010
	`He0	*H
 ~0z0b :0
	*H
0S10	UUK10U

eScienceCA10U	Authority10UUK e-Science CA 2B0
190819155650Z
200917155650Z0e10	UUK10U
eScience10U	Cambridge10
UUCS1!0Uchristopher robert jones0"0
	*H
0
ĺ]nغos&Xen*zݫi1bg\_,JPOvrTJ*_Lnk% \
{L)joqy=W	0!ou~f-oZa}9XKI&.KM55QEՐo;Mz \
Qh^Z`M$ceX}$(?!F*3Ȁ'e#'V3J4}\>h	pub=Q) \
C0?0U00U0U%0++0U \
؝0q9鍐[270U#0j\	Y6U0#U0jonesc@hep.phy.cam.ac.uk0%U0support@grid-support.ac.uk07U \
00.0 +/0
*HL0
*HL0;U40200 . ,*http://crl.ca.ngs.ac.uk/crl/escience2b.crl0
	*H
MlFX
lfo\h^C^Qs
Bncq<U>M>o|V2cX#
߹󫕍%ױq[H6V<ޱKwJY5k!j5Y`7i,&&ق|g-JшGmz'Dsڌ \
3#3h #|`>ٶjGm~+򧧗K\:T$#\EULlJ[p#$(H100Z0S10	UUK10U
 
eScienceCA10U	Authority10UUK e-Science CA 2B:0
	`He A0	*H
	1	*H
0	*H
	1
200423192600Z0/	*H
	1" 7E>αG`~B-
t%0i	+71\0Z0S10	UUK10U

eScienceCA10U	Authority10UUK e-Science CA 2B:0k*H
	1\ Z0S10	UUK10U

eScienceCA10U	Authority10UUK e-Science CA 2B:0
	*H
c4:UFajuG8%СxNЀR(Za]];JUŤxv%W!B5@BVC$ \
T2cŮVi@%]'ǜWx)bI\j)*ܾI	wE'{&}uP/	oL-S \
&ݜ (!uΞxsn^8wK6\{}>E \
x3'1M#B1Ճ?եnnP!



[prev in list] [next in list] [prev in thread] [next in thread]

Configure | About | News | Add a list | Sponsored by KoreLogic