The build expression got quiet large over time and to make it a bit
easier to grasp the different scripts involved in the build are now
separated from the nix file.
Cargo has a few odd (old) ways of picking source files if the `bin.path`
attribute isn't given in the Cargo.toml. This commit adds support for
some of those. The previous behaviour always defaulted to `src/main.rs`
which was not always the right choice.
Since there is look-ahead into the unpacked sources before running the
actual builder the path selection logic has to be embedded within the
build script.
`buildRustCrate` currently supports two ways of running building
binaries when processing a crate:
- Explicit definition of all the binaries (& optionally the paths to
their respective `main.rs`) and,
- if not binary was explictly configured all files matching the patterns
`src/main.rs`, `src/bin/*.rs`.
When the explicit list is given without path information paths are now
being picked from a list of candidates. The first match wins. The order
is the same as within the cargo compatibility code.
If the crate does not provide any libraries the path `src/{bin_name}.rs`
is also considered.
All underscores within the binary names are translated into dashes (`-`)
before the lookups are made. This seems to be a common convention.
Build failed because of dependency Flask-PyMongo, which actually
isn't a dependency of eve as they have their own flask_pymongo code.
- Drop incorrect dependency flask-pymongo
- Drop redunant dependencies already propageted by flask:
itsdangerous, werkzeug, jinja2
- Drop dependency markupsafe, not required any more
See https://hydra.nixos.org/build/81125645
`tamarin-prover' upstream has a patch to fix GHC 8.4 compilation (and
uses stack lts-12.1 now), but it's not released yet:
a08f6e4007
The build is divided in several derivations, therefore the patch had to
be splitted and rebased for `lib/term', `lib/theory' and `lib/utils' to
ensure that the patch applies properly during the `patchPhase'.
Addresses #45960
Previously the Release.xz URL would show up with a new hash whenever
debian releases an update. By using archive.org we should have a stable
source for those. I wasn't able to find the equivalent in the debian
world. Maybe they don't keep all the different Release files around..
- add missing checkInputs
- apply upstream patch to fix some tests that failed due to
expired test ssl certs
- re-enable a previously disabled test case
Loading olm.lua as weechat script with `/script load olm.lua' causes
errors like this:
```
/nix/store/43jbh7yxh8j4gjfzbvpd9clncah5dip1-weechat-matrix-bridge-2018-05-29/lib/ffi.so: undefined symbol: lua_tointeger
```
As `olm.lua' is loaded by `matrix.lua' it doesn't need to be included
manually by the weechat configuration.
Could be caused by our older 10.10.5 CoreFoundation.
# github.com/segmentio/aws-okta/vendor/github.com/keybase/go-keychain
go/src/github.com/segmentio/aws-okta/vendor/github.com/keybase/go-keychain/corefoundation_go110.go:35:33: cannot use nil as type _Ctype_CFAllocatorRef in argument to _Cfunc_CFDataCreate
go/src/github.com/segmentio/aws-okta/vendor/github.com/keybase/go-keychain/corefoundation_go110.go:61: cannot use nil as type _Ctype_CFAllocatorRef in argument to func literal
go/src/github.com/segmentio/aws-okta/vendor/github.com/keybase/go-keychain/corefoundation_go110.go:98:41: cannot use nil as type _Ctype_CFAllocatorRef in argument to _Cfunc_CFStringCreateWithBytes
go/src/github.com/segmentio/aws-okta/vendor/github.com/keybase/go-keychain/corefoundation_go110.go:133: cannot use nil as type _Ctype_CFAllocatorRef in argument to func literal
/cc ZHF #45961
In file included from qt/bitcoin.cpp:9:
./qt/bitcoingui.h:14:10: fatal error: 'QLabel' file not found
#include <QLabel>
^~~~~~~~
1 error generated.
/cc ZHF #45961
Could be caused by our older 10.10.5 CoreFoundation.
go/src/github.com/ethereum/go-ethereum/vendor/github.com/rjeczalik/notify/watcher_fsevents_cgo.go:51:216:
cannot use nil as type _Ctype_CFAllocatorRef in argument to func literal
go/src/github.com/ethereum/go-ethereum/vendor/github.com/rjeczalik/notify/watcher_fsevents_cgo.go:162:47:
cannot use nil as type _Ctype_CFAllocatorRef in argument to
_Cfunc_CFStringCreateWithCStringNoCopy
go/src/github.com/ethereum/go-ethereum/vendor/github.com/rjeczalik/notify/watcher_fsevents_cgo.go:163:225:
cannot use nil as type _Ctype_CFAllocatorRef in argument to func literal
/cc ZHF #45961
Started to fail since 1.12.1.
Undefined symbols for architecture x86_64:
"boost::chrono::steady_clock::now()", referenced from:
pinger::start_send() in ping.o
asio::detail::timer_queue<asio::detail::chrono_time_traits<boost::chrono::steady_clock, asio::wait_traits<boost::chrono::steady_clock> > >::wait_duration_msec(long) const in ping.o
asio::detail::timer_queue<asio::detail::chrono_time_traits<boost::chrono::steady_clock, asio::wait_traits<boost::chrono::steady_clock> > >::wait_duration_usec(long) const in ping.o
asio::detail::timer_queue<asio::detail::chrono_time_traits<boost::chrono::steady_clock, asio::wait_traits<boost::chrono::steady_clock> > >::get_ready_timers(asio::detail::op_queue<asio::detail::scheduler_operation>&) in ping.o
pinger::handle_receive(unsigned long) in p
/cc ZHF #45961
The darwin build fails and it's probably not particularly useful there.
utils.c:33:19: error: use of undeclared identifier 'CLOCK_MONOTONIC'
clock_gettime(CLOCK_MONOTONIC, &t);
/cc ZHF #45961
Fixed minor issue where kube-addon manager complaints about
/opt/namespace.yaml missing.
Added release notes with reference to Kubernetes 1.11 release notes.
closes#43882
Added a wrapper package that creates symlinks to OpenSSL libraries with
the actual versions, because we have 1.0.2 with .so-suffix .1.0.0 and
cl-async-ssl is unhappy because of that.
I continue to dislike cl-postgres and simple-date upstream packaging.
The doctests for termonad fail to build only with nix. When building
without nix, the doctests run correctly:
https://github.com/cdepillabout/termonad/issues/15
This PR disables the tests for termonad, as well as removing it from
dont-distribute-packages.
`all-functions` corresponds to `-fprof-auto` which places an SCC on every binding. It is well known that SCCs hinder GHC from doing its optimization magic and really slows down profiled code to a point where the profiling reports are completely skewed towards things that were completely optimized away in production settings. Concretely this shows up with things like lenses which do not carry runtime overhead when properly simplified.
`exported-functions` corresponds to GHCs `-fprof-auto-exported` which doesn't put SCCs on `INLINE`d code and in turn doesn't influence simplification of this basic but important stuff.