* Bug 1723963 - Eliminate duplicated name for suite and test
* Bug 1723963 - Add unit tests
* Bug 1723963 - Fix unit tests
* Bug 1723963 - Move TextualSummary class, fix duplication test name for bugzilla
* Bug 1723963 - Reset testAlert signature's values
* Bug 1723963 - Rename Suite & Test column to Test for bug description
* Compute the noise profile of the data when generating alerts.
* Fix linting issues.
* Use choices in noise_profile, and unpack tuple on return.
* Fix linting issues
* Logging error to New Relic
Co-authored-by: bacasandrei <beatrice.raluca.acasandrei@gmail.com>
There are short or common test file names like 001.html. When Treeherder tries
to generate bug suggestions for that, it searches the intermittent bugs for
001.html in the summary which also matches other-test-001.html. If too many
bugs are returned (>20), Treeherder won't suggest any bugs.
By matching on path boundaries (/, \), whitespace () and list separators (,),
the other test files won't be matched. Because adding these rules to the SQL
query yields a slower method than filtering the wrong positive bugs out
afterwards, the latter method gets applied. This keeps the risk the SQL query
will not return all matches (limited at 50 lines) and has to be reevaluated if
it turns into an issue.
* Enable pagination in Compare View
* Moved implementation to CompareTableControls
* WIP updating the url parameters
* Keep filtered text in the input box when refreshing page
* Compact data in AlertTable
* Add sort to Magnitude of Change
* Abbreviate next/previous value, fix spacing tags&options
* Add unit tests for new/previous value
* Rename AlertTableMagnitude to Magnitude
* Update docker-shared-user for pulse_url and add PROJECTS_TO_INGEST to backend container
* Update docs to make them clearer
* Fix exception caught in pytest.raises
* Switch test logic to use mozci for regressions
* bump timeout on web node to give leeway for longer-running queries
* make changes to mozci config file
* fix duplicate jobs in unstructured logs bug
* Use mozci labels to increase accuracy of build failures
* Combine duplicate health and health_summary API logic into helper function
* Expand mozci to query groups and labels
* clean up get_commit_history and CommitHistory component
* refactor lints and builds to use same jobs
* get_test_failure_jobs refactor to remove extra query
* refactor get_test_failures to utilize existing data
* only return what we need from Job query
* fix JS tests and cleanup console warnings
fix python push health lint, test and build tests
* expand exception handling and add logging in get_commit_history
add logging to _get_failure_data
* Add separate mozci config file for local development
* upgrade mozci to 2.0.0
Co-authored-by: Cameron Dawson <cdawson@mozilla.com>
Co-authored-by: Sarah Clements <sclements313@gmail.com>
* Add author query string to my pushes page
* Add support for medium screen sizes to My Pushes page and fix a few UI elements
* Jobs that are completed should be green
* Health should stop querying API if status is completed
* Fix push health status widget in jobs view
* change backend logic with a default for when no results have been found
* change UI logic so that each tab is only shown if we have results and default value is in progress
* replace MetricCount use with StatusButton to ensure consistency
* Refactor PlatformConfig into two components
* Remove irrelevant tests
* fix email' s initial description
* fix push range hyperlink
* set job symbol instead of job type
* relocated method for setting job symbol
* provide searchStr param similar to that computed by frontend
* add partial unit tests
* define test stub
* provide better markdown values for alert summary & push range fields
Co-authored-by: Alexandru Irimovici <airimovici@mozilla.com>
* Re-enable caching of bug suggestions during ingestion
* Move search term sanitizing into its own function and add test cases
* do not search as a phrase
* temporarily disable test_for_empty_search_query
* Temporary bugscache fixes
Stop use of bugscache.search during log parsing until a proper
fix and tests for sanitizing strings in this query can be implemented
* add backslash to list of characters to clean from strings
* remove print and adjust comment
* rename backfill bot' s packages to Sherlock
* slight logs adjustments to Sherlock
* replace 'Perfsheriff bot' prefix with 'Sherlock'
* adjust comment referring to Sherlock
* rename SecretaryTool to Secretary
* also remove alerts older than a year
* slightly refactor main garbage collection method
* fix ORM filter
* provide test coverage
* reduce ambiguity around signature garbage collection
* parameterize ingestion test with job tier
* provide slight refactors on test code
* refactor complex alerting conditions
* split alert generation (test) scenarios in happy/unhappy paths
* address some PyCharm inspection warnings
* Bump python to 3.7.10 to try to fix issues with typing module
* Remove requirement on typing again
* Remove bigquery and activedata references (#7051)
* Remove activedata and bigquery again
* Add importlib-metadata to dev requirements
* Remove remnants of selenium tests from setup.cfg
* Don't call taskcluster.aio.createSession outside of async functions (it's not allowed)
* Disable newrelic linters to fix issue with imp module
See https://discuss.newrelic.com/t/python-warnings-during-pytest/114897
* Fix linting errors
* Use pytest-xdist to speed up tests
* Properly ignore newrelic warnings
* Disable django.contrib.staticfiles in tests, because it breaks everything
* Set TREEHERDER_DEBUG=False for unit tests, because it breaks some tests
* Fix linting issues
Sarah Clements
Sorin Toma
beatrice-acasandrei
esanuandra
Gregory Mierzwinski
Sebastian Hengst
ionutgoldan
Cameron Dawson
Marco Castelluccio
Alexandru Irimovici
Ben Hearsum (he/him)