cpet/openbsd-ports
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
145,313 Commits 1 Branch 0 Tags
Commit Graph

5469 Commits

Author SHA1 Message Date
sthen
36eb147a1a oops, adjust previous to unbreak (needed _BUILDLIB_DEPENDS in addition 
to BUILD_DEPENDS rather than replacing)
 2019-12-02 12:02:05 +00:00
sthen
0660d1cde0 Change the gettext-tools poisoning to use _BUILDLIB_DEPENDS instead of 
BUILD_DEPENDS and simplify (it's always defined, so no need for a "is defined"
check).

Although it's not used often there is now an external-facing library
that is part of gettext-tools so it's valid to have it listed as a LIB_DEPENDS
rather than a BUILD_DEPENDS.

_BUILDLIB_DEPENDS suggested by espie@ naddy@ as an improvement to my more
complex and less complete "check BUILD_DEPENDS and LIB_DEPENDS" which ignored
multi-packages.
 2019-12-02 11:38:22 +00:00
espie
0ab1dbdef3 for the packaging step 
- don't use $@D so that we can put two targets in there
- set a variable to add the debug package to the targets

This fixes behavior in the very specific case the normal package was already
built without DEBUG_PACKAGES and you're repackaging with DEBUG_PACKAGES set.

okay landry@
 2019-12-02 10:36:51 +00:00
espie
c1024b8fcd update the internals with a few more details 2019-12-01 09:36:29 +00:00
kmos
03e2882de7 Add the common misspelling "MODPY_TEST" (should be "MODPY_PYTEST") to the 
poisoned misspellings to help point out when the typo happens.

ok phessler
 2019-11-30 18:11:46 +00:00
espie
542e93575d document the result 2019-11-30 10:20:53 +00:00
sthen
4c645f78fa oops, that should not have been part of previous commit 2019-11-29 02:43:16 +00:00
sthen
d13bf19794 reserve 846 for net/dhcpcd, the master branch upstream now has working privsep 2019-11-29 02:42:44 +00:00
espie
e8b70183b5 so figured out what to do with links. Fun! 
(there's still one corner case to detect, but it's unlikely to happen in
practice)
 2019-11-27 15:10:19 +00:00
sthen
d665f4f90d +snmp_exporter 2019-11-27 10:47:22 +00:00
espie
70af6987c7 a few more comments in the generated makefile 
fold abstraction that's no longer needed
turns out I don't need to worry about symlinks, egdb handles them
 2019-11-26 12:02:46 +00:00
espie
609569ab04 better if I build the correct package :/ 
sorry guys, broken debug packages
 2019-11-26 11:49:02 +00:00
espie
7eaccc69c8 fold trivial code 2019-11-25 15:35:05 +00:00
espie
afa1c62789 stop doing anything with @static-lib, my plan didn't work correctly 
(how do you link with a lib in a package that you can't really depend
upon)

There are few enough static libs that the difference won't be that much
anyway.

okay pirofti@
 2019-11-25 14:51:15 +00:00
espie
337eaeceb0 add a warning for the actual absence of debug-info 
okay pirofti@
 2019-11-25 14:35:06 +00:00
espie
1136f61008 have comment reflect reality 2019-11-25 13:57:11 +00:00
espie
efd273d5a8 rename p into pkgname, so I do not need an extra temp variable 
move checksum_package to the final loop, so *only* final packages
will get checksummed
 2019-11-24 12:54:12 +00:00
espie
703ae9ad93 restore logic to NOT register debug packages 
noticed by aja@, thx
 2019-11-24 12:23:22 +00:00
espie
eeedab813e use the exact same logic for FETCH_PACKAGES: grab all files. If we got 
them all we move them (and we can still fail if permissions are not okay)
Otherwise, we run the normal _internal-package-only again, with FETCH_PACKAGES
neutralized (as before)

This fixes the issue reported by semarie@ and kmos@
 2019-11-23 16:31:20 +00:00
espie
085528005c tweak the logic slightly: use one single variable _pkg${_S} to hold the 
packages we build (normal/debug) and put the code back inside the actual
target instead of fragments.

Stop doing tests (shell exits as usual and trap is enough to remove temp
files).

The same logic will be used for debug_packages
 2019-11-23 16:19:48 +00:00
espie
bdfceef887 add extra variables to make the definition of PACKAGE_COOKIE more regular 
to be used for DEBUG_PACKAGES
 2019-11-23 16:00:34 +00:00
espie
5bc08af6fd zap useless tests (dates back when this governed _PACKAGE_LINKS as well, 
before the MULTI_PACKAGES/BUILD_PACKAGES separation)
 2019-11-23 15:50:14 +00:00
espie
58cccbe929 fix bug in FETCH_PACKAGES logic: do NOT run _internal-package-only if 
FETCH_PACKAGES succeeded. Also: error out if we can't copy/link the
cached package, as this is probably some permission error.
 2019-11-23 15:43:20 +00:00
sthen
593e1a38f9 portgen sqlports improvements, changes suggested by espie in response to my 
not-so-good change to just using the PortsQ "cache":

- use _paths/_ports directly rather than the "ports" view and remove bogus
flavour-clearing regex subst

- open sqlports ReadOnly
 2019-11-23 14:59:39 +00:00
abieber
e0f2a305ba Remove unused NPM bits. 
OK afresh1@
 2019-11-22 00:01:10 +00:00
sthen
273bc91140 - comment-out net/cvsync uid 
- drop description of "noauto", not used since 2006
 2019-11-20 15:37:52 +00:00
afresh1
45800df741 Remove sqlports-compact support from portgen 
Ok, initial idea, and initial patch from cwen@
 2019-11-19 22:43:31 +00:00
espie
31c7e58902 do the destdir dance using File::Spec for simpler, canonical paths 2019-11-19 16:35:20 +00:00
espie
0b8672b605 rename generic "out" into "mk" since we're writing to a makefile 
(makes it more obvious compared to the plist fh)

start preparing for actually handling links
 2019-11-19 15:49:30 +00:00
espie
cf1e5fa4f3 repair 2019-11-19 15:06:19 +00:00
espie
d189070fb1 document a design choice 2019-11-19 14:55:42 +00:00
espie
2a18d1de91 ... and the argument parsing proper is now identical as well! 2019-11-19 14:54:01 +00:00
espie
a5d9dff07c inherit from PlistReader so that differences are located in a better location 2019-11-19 14:51:32 +00:00
espie
f1c3d659c4 move the common state things as well 2019-11-19 14:45:28 +00:00
espie
978d83893b make the state code identical in preparation for lifting. 2019-11-19 14:41:29 +00:00
espie
cea793235b start moving common code into its own file 2019-11-19 14:38:56 +00:00
espie
289bc15185 add DEBUG_* glue 
okay sthen@
 2019-11-19 12:39:04 +00:00
espie
5c2dd6bbc8 move process_subpackage logic into the PlistReader class in order to 
share the code
 2019-11-18 16:07:17 +00:00
espie
37b5d2c0a2 make things closer to update-plist 2019-11-18 15:56:29 +00:00
espie
9ffc3dddaf tweak code order slightly to make things identical to build-debug-info 2019-11-18 15:54:48 +00:00
espie
28daf77466 I'm an idiot, of course conflict requires some tweaks as well 2019-11-14 12:47:03 +00:00
espie
aaff77ddbc temporary fix: if we don't know how to generate debug for symlinks, don't 
bother registering them
 2019-11-14 08:48:44 +00:00
espie
6fdf092c90 clean-up (put stuff into self, not state) 
inherit from BaseFS, so that dest/undest/resolve_link can happen
 2019-11-13 16:16:51 +00:00
espie
ac67505216 tweak call, some options are not needed, passing PORTSDIR will be 
necessary to grab other code.
also error out when file writing breaks.
 2019-11-13 15:22:41 +00:00
espie
029ecfd382 move the dest/undest/resolve_link code to its own class, so it can be 
reused in other WRKINST contexts (build-debug-info)

Also make use of state to display errors, like duh
 2019-11-13 11:38:48 +00:00
espie
ed96787dac tweak it yet again to always run build-debug-info, it's cheap enough 2019-11-13 10:08:57 +00:00
rsadowski
e2593bc802 update MASTER_SITE_KDE and prefer https over http/ftp. 2019-11-12 21:24:14 +00:00
espie
dbc7492568 don't error out on symlinks for now, this allows things to proceed 2019-11-12 20:06:54 +00:00
espie
10152e79fb copy remaining annotations with special semantics (cwd but first, pkgpath 
with debug/ prepended)

put a bit more boilerplate in the generated Makefile, so that the targets
themselves are much shorter
 2019-11-12 16:41:36 +00:00
espie
e9cee4d943 minor tweaks: 
- remove PKG_ARCH=* packages from DEBUG_PACKAGES, to help with
introspection
- ditch DEBUG_FILES entirely, the automated mechanism appears to be
good enough
- tweak show-debug-info to show the file names
 2019-11-12 16:40:26 +00:00