mozilla
/
gecko-dev
зеркало из https://github.com/mozilla/gecko-dev.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
gecko-dev/testing/web-platform/test_metamerge.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

228 строки
3.8 KiB
Python
Исходник Обычный вид История

Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
from io import BytesIO
import metamerge
Bug 1790816 - Reformat testing/ with isort. r=linter-reviewers,webdriver-reviewers,perftest-reviewers,jgraham,gbrown,whimboo,sparky,ahal # ignore-this-changeset Differential Revision: https://phabricator.services.mozilla.com/D162657
2022-12-21 17:38:37 +03:00
import mozunit
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
ancestor = """
global-new-deleted: A
global-new-changed: A
global-current-deleted: A
global-current-changed: A
[failing-test.html]
[Unchanged subtest]
expected: FAIL
[New deleted subtest]
expected: FAIL
[New modified subtest]
expected: TIMEOUT
[Current deleted subtest]
expected: FAIL
[New modified current deleted]
expected: FAIL
[Ancestor no expected new expected]
bug: 1234
[new-deleted-test.html]
[Deleted subtest]
expected: FAIL
[current-deleted-test.html]
[Deleted subtest]
expected: FAIL
[test-modified.html]
expected: TIMEOUT
[new-modified-current-deleted.html]
expected:
if os == "linux": FAIL
TIMEOUT
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
[new-add-expected-with-subtest.html]
[subtest]
expected: FAIL
[current-add-expected-with-subtest.html]
[subtest]
expected: FAIL
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
"""
new = """
global-new-added: A
global-new-changed: B
global-current-deleted: A
global-current-changed: A
[failing-test.html]
[Unchanged subtest]
expected: FAIL
[New added subtest]
expected: FAIL
[New modified subtest]
expected:
if os == "linux": FAIL
TIMEOUT
[Current deleted subtest]
expected: FAIL
[New modified current deleted]
expected: TIMEOUT
[Ancestor no expected new expected]
bug: 1234
expected: FAIL
[new-added-test.html]
[Added subtest]
expected: FAIL
[current-deleted-test.html]
[Deleted subtest]
expected: FAIL
[test-modified.html]
expected:
if os == "linux": FAIL
[new-modified-current-deleted.html]
expected:
if os == "linux": FAIL
if os == "mac": FAIL
TIMEOUT
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
[new-add-expected-with-subtest.html]
expected: FAIL
[subtest]
expected: FAIL
[current-add-expected-with-subtest.html]
[subtest]
expected: FAIL
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
"""
current = """
global-new-deleted: A
global-new-changed: A
global-current-added: A
global-current-changed: B
[failing-test.html]
[Unchanged subtest]
expected: FAIL
[New deleted subtest]
expected: FAIL
[New modified subtest]
expected: TIMEOUT
[Current added subtest]
expected: FAIL
[Ancestor no expected new expected]
bug: 1234
[new-deleted-test.html]
[Deleted subtest]
expected: FAIL
[current-added-test.html]
[Added subtest]
expected: FAIL
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
[new-add-expected-with-subtest.html]
[subtest]
expected: FAIL
[current-add-expected-with-subtest.html]
expected: FAIL
[subtest]
expected: FAIL
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
"""
updated = """global-new-deleted: A
global-new-changed: A
global-current-added: A
global-current-changed: B
[failing-test.html]
[Unchanged subtest]
expected: FAIL
[New modified subtest]
expected:
if os == "linux": FAIL
TIMEOUT
[Current added subtest]
expected: FAIL
[Ancestor no expected new expected]
expected: FAIL
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
bug: 1234
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
[New added subtest]
expected: FAIL
[New modified current deleted]
expected: TIMEOUT
[current-added-test.html]
[Added subtest]
expected: FAIL
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
[new-add-expected-with-subtest.html]
expected: FAIL
[subtest]
expected: FAIL
[current-add-expected-with-subtest.html]
expected: FAIL
[subtest]
expected: FAIL
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
[new-added-test.html]
[Added subtest]
expected: FAIL
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
[test-modified.html]
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
expected:
if os == "linux": FAIL
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
[new-modified-current-deleted.html]
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
expected:
if os == "linux": FAIL
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
if os == "mac": FAIL
TIMEOUT
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
"""
def test_merge():
def get_manifest(str_data):
Bug 1622686 - Make testing/web-platform flake8 compliant. r=sylvestre,jgraham This also enables py3 linting for testing/web-platform. The external testing/web-platform/tests is excluded from linting. Differential Revision: https://phabricator.services.mozilla.com/D90744
2020-09-29 17:22:49 +03:00
bytes_io = BytesIO(str_data.encode("utf-8"))
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
return metamerge.compile(bytes_io, metamerge.data_cls_getter)
Bug 1654103: Standardize on Black for Python code in `mozilla-central`. Allow-list all Python code in tree for use with the black linter, and re-format all code in-tree accordingly. To produce this patch I did all of the following: 1. Make changes to tools/lint/black.yml to remove include: stanza and update list of source extensions. 2. Run ./mach lint --linter black --fix 3. Make some ad-hoc manual updates to python/mozbuild/mozbuild/test/configure/test_configure.py -- it has some hard-coded line numbers that the reformat breaks. 4. Make some ad-hoc manual updates to `testing/marionette/client/setup.py`, `testing/marionette/harness/setup.py`, and `testing/firefox-ui/harness/setup.py`, which have hard-coded regexes that break after the reformat. 5. Add a set of exclusions to black.yml. These will be deleted in a follow-up bug (1672023). # ignore-this-changeset Differential Revision: https://phabricator.services.mozilla.com/D94045
2020-10-26 21:34:53 +03:00
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
ancestor_manifest = get_manifest(ancestor)
current_manifest = get_manifest(current)
new_manifest = get_manifest(new)
Bug 1576991 - Ensure expected is correctly inserted merging wpt ini files - r=maja_zf We can't parse files where the key-value pairs are after the subtest data, so ensure that we add it before. Also fix a bug when metadata is added on one side of the merge but unchanged on the other; we were accidentially inserting the entire test rather than just the expected value. Differential Revision: https://phabricator.services.mozilla.com/D43658 --HG-- extra : moz-landing-system : lando
2019-08-28 16:37:58 +03:00
result = metamerge.make_changes(ancestor_manifest, current_manifest, new_manifest)
assert result == updated
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
Bug 1622686 - Make testing/web-platform flake8 compliant. r=sylvestre,jgraham This also enables py3 linting for testing/web-platform. The external testing/web-platform/tests is excluded from linting. Differential Revision: https://phabricator.services.mozilla.com/D90744
2020-09-29 17:22:49 +03:00
Bug 1571755 - Add a command for merging wpt metadata, r=maja_zf This command is inteded to be usable as a git mergetool for the specific case of merging ini files in a way that should always succeed, produces reasonably likely output, but isn't reliably correct. The main use case is for the sync bot where we update metadata on branches and experience conflicts when we also have changes on master. We don't necessarily need to resolve these perfectly but we do need to provide a resolution automatically since otherwise it blocks syncs. Since the ini parser is in-tree we want to make this an in-tree command that the sync can use to provide the resolution. The general strategy is that we want to prefer the "new" metadata where possible. This won't always be correct e.g. if a test got fixed on master and simultaneously got edited to go from TIMEOUT to FAIL in a browser without the fix. But it's not a bad approximation (and generally a human will struggle to do the merge better by hand, so we have to assume later try jobs will fix things). Differential Revision: https://phabricator.services.mozilla.com/D40832 --HG-- extra : moz-landing-system : lando
2019-08-09 20:48:53 +03:00
if __name__ == "__main__":
mozunit.main()