Commit Graph

37 Commits

Author SHA1 Message Date
Justin M. Keyes
6c65f8f182 CI/AppVeyor: revert whitelist
revert 6b028ec5f2 #10746

This seems to skip tags also, which breaks our release automation.
From the AppVeyor logs:

    2019-09-15 17:25:00.232 Warning Commit "1f76c4af" of branch "master" skipped as commit message contains either "[skip ci]" or "[ci skip]" or "[skip appveyor]"
    2019-09-15 17:20:24.152 Warning Commit "baa5263b" skipped as branch "stable" is not in white-list
    2019-09-15 17:20:03.163 Warning Commit "6cb5ffc0" skipped as branch "release-0.4" is not in white-list
    2019-09-15 17:20:02.977 Warning Commit "baa5263b" skipped as branch "v0.4.1" is not in white-list
    2019-09-15 16:35:38.810 Warning Commit "31956372" of branch "master" skipped as commit message contains either "[skip ci]" or "[ci skip]" or "[skip appveyor]"
    2019-09-15 16:28:35.237 Warning Commit "e2cc5fe0" skipped as branch "stable" is not in white-list
    2019-09-15 16:28:17.814 Information Build version 31489 created
    2019-09-15 16:28:16.364 Warning Commit "e2cc5fe0" skipped as branch "v0.4.0" is not in white-list
2019-09-15 17:39:07 -07:00
Justin M. Keyes
ce9367c254
CI/AppVeyor: revert "skip MSVC_32 for non-PR" [skip travis] #10805
Revert commit bfbc1a7872.

The nightly release job is failing:
https://travis-ci.org/neovim/bot-ci/builds/569687794

- We need the MSVC artifact on master so that our nightly job can publish it.
- Saving time on master is low-priority because most CI activity is from PRs.
2019-08-18 17:26:20 +02:00
Daniel Hahler
bfbc1a7872 CI/AppVeyor: skip MSVC_32 for non-PR builds [skip travis] #10786
It was run for the PR already, and we can save the ~25min it takes with
merged PRs then - given that there is only one parallel job on AppVeyor
this should help to reduce the queue size.
2019-08-16 20:13:31 +02:00
Daniel Hahler
6b028ec5f2
ci: AppVeyor: branches: only: master (#10746)
Only build PRs for master, and not other branches on the main repo, e.g.
when reverting via GitHub's UI.
2019-08-11 20:30:27 +02:00
Daniel Hahler
b59a1f7b52
ci: restore accidentally removed config (#10592) 2019-07-24 02:16:09 +02:00
Daniel Hahler
f08d10a0df
tests: re-enable "tab drag in tabline to the left moves tab left" (#10588)
Ref: https://github.com/neovim/neovim/pull/4874
2019-07-24 02:12:19 +02:00
Daniel Hahler
28a86608a8 CI: improve gcov handling #10404
- Move __gcov_flush to process_spawn, for more reliable coverage
  tracking of subprocesses
- Travis: use GCOV_ERROR_FILE
- codecov: use "-X fix" to skip "fixing" uploaded coverage data; it
  should be handled by codecov's backend instead.
- AppVeyor: no $PATH mangling, which breaks with the improved coverage tracking
  due to missing .dll in PATH.
2019-07-07 21:09:37 +02:00
Daniel Hahler
04f559dc05
ci: AppVeyor: GCOV_ERROR_FILE: head/tail [skip ci] (#10335) 2019-06-27 14:38:33 +02:00
Daniel Hahler
755e94801f ci: AppVeyor: use fast_finish=true 2019-06-17 02:00:58 +02:00
Daniel Hahler
93f8c2793c
ci: AppVeyor: fix cov job, remove duplicate non-cov one (#10217)
* ci: AppVeyor: set GCOV_ERROR_FILE

This prevents the warnings/errors to be spilled into test results,
causing them to fail them, e.g.:

    [  FAILED  ] C:/projects/neovim/test/functional\core\main_spec.lua @ 97: Command-line option -s errors out when trying to use nonexistent file with -s
    C:/projects/neovim/test/functional\core\main_spec.lua:98: Expected objects to be the same.
    Passed in:
    (string) 'Cannot open for reading: "Xtest-functional-core-main-s.nonexistent": no such file or directory
    profiling:C:\projects\neovim\build/src/nvim/CMakeFiles/nvim.dir/buffer.c.gcda:Data file mismatch - some data files may have been concurrently updated without locking support
    '
    Expected:
    (string) 'Cannot open for reading: "Xtest-functional-core-main-s.nonexistent": no such file or directory
    '

    stack traceback:
            C:/projects/neovim/test/functional\core\main_spec.lua:98: in function <C:/projects/neovim/test/functional\core\main_spec.lua:97>

For reference, the locking appears to have been reworked for gcc 9.1 [1].

1: https://github.com/gcc-mirror/gcc/commit/56621355b

helpers.clear: keep GCOV_ERROR_FILE in environment

* ci: AppVeyor: remove MINGW_64 config (used with cov now)

Also:

- run MINGW_64-gcov first, and with PRs, since it provides coverage.
2019-06-14 13:39:57 +02:00
Justin M. Keyes
ae2401621a
ci/build.ps1: Respect CMAKE_BUILD_TYPE if provided #9869 2019-04-09 00:17:07 +02:00
Justin M. Keyes
8a8213f6d4 CI/AppVeyor: remove redundant cache directive
Maybe AppVeyor gets confused by the redundant entry?

Anyway, build.ps1 generally does not affect the validity of the deps
cache.
2019-04-08 01:46:25 +02:00
Justin M. Keyes
45bf5bb666 CI/AppVeyor: do not cache pacman packages
pacman log says:
    Total Download Size:   17.06 MiB
So this takes very little of the quota. But it also takes only a few
seconds to install, so caching it saves little or no time.
2019-04-08 01:41:25 +02:00
Justin M. Keyes
02aa57afa5 CI/AppVeyor: build deps out-of-tree
appveyor.yml: set cache to an absolute path.

Desperate attempt to get AppVeyor cache to work.

My assumption in a7a56293aa #9852 that that different jobs were
overwriting each other's cache is probably wrong: AppVeyor
docs/discussions hint that the cache is per-config (though I haven't
found a clear, unambiguous statement as such).
2019-04-08 01:00:47 +02:00
Justin M. Keyes
c5e8924f4e
CI/AppVeyor: do skip-logic earlier #9854 2019-04-07 14:43:50 +02:00
Justin M. Keyes
a7a56293aa
CI/AppVeyor: per-compiler deps cache #9852
The deps cache does not work for MSVC builds because the MINGW builds
ovewrite it.  Cache saves 10+ minutes on the build.
2019-04-07 03:54:22 +02:00
Jan Edmund Lazo
a7b17d4b3e CI/AppVeyor: invalidate .deps if build.ps1 changes 2018-10-22 01:28:02 +02:00
b-r-o-c-k
393935c32d ci/AppVeyor: enable MSVC_32 build 2018-03-26 10:54:44 -05:00
b-r-o-c-k
e24e98534b ci/AppVeyor: use PowerShell (#8124) 2018-03-11 23:44:07 +01:00
Justin M. Keyes
ffad8d4c51 ci/AppVeyor: disable MSVC_32 build
The MSVC_32 currently hangs.  When MSVC becomes the primary Windows
target, we can enable MSVC_32 and retire one of the mingw builds.  In
the meantime it adds too much time.
2018-03-11 12:43:42 +01:00
b-r-o-c-k
e0d55f2871 build/msvc: Allow MSVC AppVeyor builds to fail 2018-03-05 11:01:47 -06:00
b-r-o-c-k
cfb713b5c6 build/msvc: Add Appveyor CI for MSVC 2018-03-04 17:44:23 -06:00
Jan Edmund Lazo
a1adfdc7d5 ci: nodejs client acceptance-test #7706
ci: install nodejs 8 in Appveyor, Travis

provider: check node version for debug support
Resolve https://github.com/neovim/neovim/pull/7577#issuecomment-350590592 for Unix.

provider: test if nodejs in ci supports --inspect-brk

nodejs host for neovim requires nodejs 6+ to work properly.
nodejs 6.12+ or 7.6+ is required for debug support via `node --inspect-brk`.

provider: run cli.js of nodejs host directly

npm shims are useless because the user cannot set node to debug mode via
--inspect-brk. This is problematic on Windows which use batchfiles and
shell scripts to compensate for not supporting shebang.

The patch uses `npm root -g` to get the absolute path of the global npm
modules. If that fails, then the user did not install neovim npm package
globally. Use that absolute path to find `neovim/bin/cli.js`, which is
what the npm shim actually runs with node. glob() is for a simple file
check in case bin/ is removed because the npm shims are ignored now.
2017-12-17 16:09:18 +01:00
Justin M. Keyes
981387b7c8 ci/appveyor: modify compression options for cache
Attempt to workaround #7317 by using a different compression algorithm.
2017-10-01 12:17:26 +02:00
James McCoy
9a9d9a187f ci: Ignore MINGW_64-gcov failures in overall ci status 2017-08-16 11:24:12 -04:00
Justin M. Keyes
44dc8bbb13 ci/win: list build permutations explicitly (#7163)
This avoids changing the matrix permutations for the old non-gcov
builds, so that old URLs continue to work:

https://ci.appveyor.com/api/projects/neovim/neovim/artifacts/build/Neovim.zip?branch=master&job=Configuration%3A%20MINGW_32
https://ci.appveyor.com/api/projects/neovim/neovim/artifacts/build/Neovim.zip?branch=master&job=Configuration%3A%20MINGW_64
2017-08-15 11:06:51 +02:00
James McCoy
e88fc35429
ci: Collect coverage data for MINGW_64 builds on appveyor 2017-08-14 09:15:59 -04:00
Justin M. Keyes
de378477cc ci/appveyor: fix cache pattern 2017-04-11 02:37:39 +02:00
ZyX
60ee50ce4b appveyor: Widen .deps cache dependency 2017-04-10 16:17:02 +03:00
Nikolai Aleksandrovich Pavlov
a1c928e70c ci: Do not hide ci directory (#6410) 2017-03-31 14:32:58 +02:00
Justin M. Keyes
0095ad5693 win/CI: Cache dependencies. 2017-02-16 16:23:18 +01:00
Justin M. Keyes
880ce887ed ci: Update appveyor artifact path. 2016-12-09 17:09:17 +01:00
Rui Abreu Ferreira
eb2d547bd5 build: Refactor appveyor/Windows scripts (#5244)
- Join all msys build scripts into one
- Clean up appveyor.yml and generate installer for tagged builds

A side effect of the cosmetic changes is that appveyor artifact
permalinks are easier to use, e.g. for the latest zip in the
master branch

    https://ci.appveyor.com/api/projects/neovim/neovim/artifacts/build/Neovim.zip?branch=master&job=Configuration%3A%20MINGW_64
2016-08-26 01:38:46 -04:00
Rui Abreu Ferreira
6348864e29 Appveyor: Generate helptags and build artifacts
Build the default CMake target now that helptag generation is
working again.

For build artifacts create a zip file with an instalation of
Neovim (generated by cpack).
2016-08-09 13:14:43 +01:00
Rui Abreu Ferreira
f53c8258be MinGW Appveyor builds
- Build for MinGW x86/x86_64. Move build scripts out of the yml file into
  separate batch files.
- The MinGW builds use MSYS to get runtime dependencies, but they do not
  link against the POSIX adaptation layer.
- For now only build the nvim.exe binary, but not the helptags.
2016-08-01 12:57:58 +01:00
Seth Jackson
e12d371f9d Windows: Crank to MSVC 2015 on AppVeyor. #3254 2015-08-27 23:21:04 -04:00
Rui Abreu Ferreira
43c767ba78 AppVeyor: MSVC builds for third-party/ 2015-05-15 17:55:02 +03:00