Add an extension for accessing Git-related meta data that can be mixed
in into a Pathname object (e.g. `HOMBREW_REPOSITORY` or the path of a
`Tap` instance). The goal here is to eliminate code duplication.
Beta versions of XQuartz have address sanitization enabled, which
breaks some software at runtime, including wine.
Closes Homebrew/homebrew-core#2481.
Closes#459.
Signed-off-by: Misty De Meo <mistydemeo@github.com>
Add SharedEnvExtension#gcc_with_cxx11_support? to centralise the
logic for checking whether a compiler is known to support C++11.
Update logic to accept GCC 4.8 and above (including 6). Thereby also
address oversight in #163 where support for GCC 6 was added without
updating the C++11 compiler whitelist.
Add tests for Superenv#cxx11.
Closes#346.
The fix changes behavior in same cases, but those cases were all either
broken or showed unexpected behavior. The new behavior is very simple:
- If an argument starts with `--<option-name>=`, return whatever comes
after the equals sign.
Prior to this change, `ARGV.value` showed some unexpected behavior:
- `ARGV.value("foo")` returned `nil` for `--foo=` because at least one
character needed to be present after the equals sign. (All other
option parser implementations I'm aware of allow for empty values.)
- `ARGV.value("bar")` returned `"baz"` for `--foo=--bar=baz` because the
regular expression was not anchored to the start of the argument.
- `ARGV.value("++")` raised an exception because the string wasn't
escaped for use in the regular expression. (An unlikely corner case.)
Closes#231.
Signed-off-by: Martin Afanasjew <martin@afanasjew.de>
Setting -mcpu and -mtune on Tiger with gcc-4.2 exposes a bug in one of the
system headers, causing certain builds to fail. This can be fixed by
adding -faltivec to CFLAGS.
See: http://trac.macports.org/ticket/34213
The new stage() signature introduced by #66 breaks back-compatibility
under Ruby 1.8.7. This fixes it by switching back to a one-argument
block signature and using a new class to wrap both the Resource and
Mktemp info for the staging context, in a signature-back-compatible
way.
Addresses homebrew/homebrew-core#529.
Closes#135.
Signed-off-by: Andrew Janke <andrew@apjanke.net>
Also enables sandbox for --interactive and --debug use of install
and test, using automatic retention.
Closes#66.
Signed-off-by: Andrew Janke <andrew@apjanke.net>
Old logic doesn't include revision. This fixes that, and passes the
whole formula prefix path to avoid duplicating the path-construction
logic.
Closes#80.
Signed-off-by: Andrew Janke <andrew@apjanke.net>
Previously, superenv did not try to filter -I or -L flags
based on the list of requested dependencies; as a result, buildsystems
which opportunistically discover Homebrew-installed libraries were able
to link against them even under superenv.
This adds a list of all requested dependencies to the superenv environment,
and compares all -I and -L flags against those; any Cellar and opt paths
found which resolve to unrequested dependencies are filtered out.
* Count .DS_Store disk usage but not file count.
* Count symlink's own disk usage instead of ignoring it.
* Count hardlinks disk usage only once.
* Add testcase.
ClosesHomebrew/homebrew#50563.
ClosesHomebrew/homebrew#50566.
Signed-off-by: Xu Cheng <xucheng@me.com>
SDK 10.10 isn't something that exists for Xcode 7, so stop looking for
it and rely on MACOSX_DEPLOYMENT_TARGET instead.
See PR Homebrew/homebrew#50137 Yosemite build failure
ClosesHomebrew/homebrew#50355.
Signed-off-by: ilovezfs <ilovezfs@icloud.com>
`Library/ENV` like `Library/Homebrew` is part of Homebrew basecode. It
should be able to be accessed during the `brew tests`.
By adding HOMEBREW_ENV_PATH variable, we allow test suit to locate these
codes.
HOMEBREW_MAKE_JOBS can be a multidigit number. The regex should match
the entire number not just the last digit.
ClosesHomebrew/homebrew#50016.
Signed-off-by: Tim D. Smith <git@tim-smith.us>
FileUtils::fu_get_gid only started doing the conversion of the group
to_s automatically from OS X 10.6.8 (ruby-1.8.7-p358) forward.
OS X 10.6.7 (ruby-1.8.7-p174) would fail in brew's FileUtils::mktemp with the
error "Error: can't convert Fixnum into String."
FixesHomebrew/homebrew#49045FixesHomebrew/homebrew#49348ClosesHomebrew/homebrew#49369.
Signed-off-by: Mike McQuaid <mike@mikemcquaid.com>
* Blocks writing of new files via accidental typos, etc, which the normal open("blah", "a") doesn't.
* Where files don't exist they should ideally be using `(buildpath/"dog").write` instead of open("blah", "a") already.
* It's a bit less cluttered looking if you need several writes to different files in the formula, IMO.
In theory, letting these be set once wasn't an awful idea because it allows
users choice on where Perl modules end up.
In reality, as we've moved closer to sandboxing and at the same time accepted more
perl binding options into formulae it has been exposed as something of a hellraiser.
It's not that uncommon for Perl users to set a custom `INSTALL_BASE` in the ENV which
means we end up with quite a few formula that fail hard with this sort of error:
```
Only one of PREFIX or INSTALL_BASE can be given. Not both.
```
The other common error, which we discovered via sandboxing, is that the Perl modules
end up outside Homebrew's control which means we don't automatically remove them
when the package is removed. Sandboxing blocks this, which means when we eventually
move to enable that by default for users as well as CI it'll quickly become a
prominent issue.
