of working haphazardly half the time.
NOTE: you *must* fix your sudoers for ports to keep working, make sure you
are able to pass thru SUDO_PORT_V1 ! (and the rest that's already there)
okay matthieu@, sthen@
stylistically, I tend to prefer ``, but since we use a lot of fragments,
the parenthetical nature of $() is better.
- give a more complete name to portstree-/inst- depends lists, so that several
"make package" can run in the same cache.
- create temp dependency files atomically by using mktemp/chmod/mv
(global cache)
(to be referenced by documentation, to stop confusing crypto signatures
with other signatures).
(old name kept over release, to be scraped completely afterwards, user-visible
change only)
okay sthen@
lang/icon/interp): needs more information than just "we're solving deps"
to know whether to clean things or not.
-> similar problems as locking oneself, similar solution, replace
_SOLVING_DEP with _DEPENDENCY_STACK that tracks the dependencies.
okay mbalmer^Waja
default (for production/bulks), setting it to 'yes' disables the check
as it was before (for slow archs), and setting it to 'warn' will now
show the differences between wantlib lists, but as a warning and not an
error (which was the behaviour introduced in previous commit for 'yes'
case)
ok espie@
before comparing them. In case there's a former error, there's no point
in adding to the confusion by having diff chime in one of the files doesn't
exist.
glib-gettext.m4 and intltool.m4 have been fixed ages ago but upstream
developers seem to get stuck on old versions or do not update their
macro to follow this change.
It makes no sense to keep patching the whole ports tree for it, so for
now we hardcode it here.
"why not" espie@
work.
also, warn for MASTER_SITES that don't end with /, as this is usually
an issue (but not always), as discussed with rpe@
In case successful fetch yields wrong size, always remove the partial for
later MASTER_SITES to pick it up, the 'cute' code that keeps partials over
30K is not really all that useful...
shebangs like "#!/usr/bin/env perl5" and "#!/usr/bin/perl5.12" will be
normalized altogether.
No bumps are required for ports already using MODPERL_BIN_ADJ.
Requested by and okay dcoppa@, okay sthen@
chromium was one of the ports that failed to build without PORTSDIR
being set and /usr/ports being a symlink -- but this is not the case
anymore.
If any other port fails because of this, I'll fix it.
ok espie@ todd@ sthen@
to the user-definable FETCH_CMD variable itself, allowing this to be used with
curl again (FETCH_CMD = /usr/local/bin/curl -C -) for those that need/want it
(e.g. some less common cases with proxies, or for better https support).
"go ahead" espie@, makes sense to naddy.
.for v1 v2 in list
techniques, which make it clearer.
Add new syntax to distfiles, which is not painful to do now, and support it
for dpb fetch as well: a distfile can now be file{url} to specify a different
result from the url (this allows really weird naming scheme on http sites
to give us consistent files, which has become a nagging problem).
This still doesn't mean you shouldn't encourage upstream to do the right thing
and provide nice archive urls.
extend the fetch process in bsd.port.mk to work more like dpb does, namely
download to intermediate "part" files that don't vanish and use -C if
necessary.
Both those mean that FETCH_CMD has to work like ftp, including the -C and
-o support...
discussed with various people.
