Get desktop application:
View/edit binary Protocol Buffers messages
Info on a failing test row about the failure.
Used in:
,Number of results that have failed.
The build ID the test first failed at.
The time the test first failed at.
The test ID for the first test failure.
The build ID the test last passed at.
The time the test last passed at.
A snippet explaining the failure.
Link to search for build changes, internally a code-search link.
Text for option to search for build changes.
Text to display for link to search for build changes.
The build ID for the latest test failure. (Does not indicate the failure is 'over', just the latest test failure we found.)
The test ID for the latest test failure.
Maps (property name):(property value) for arbitrary alert properties.
A list of IDs for issue hotlists related to this failure.
Dynamic email list, route email alerts to these instead of the configured defaults.
Maps (custom column headers name):(custom column headers value) for arbitrary alert properties.
A cluster of failures grouped by test status and message for a test results table.
Used in:
Test status cluster grouped by.
Error message or testFailureClassification string cluster grouped by.
ClusterRows that belong to this cluster.
Cells in a TestRow that belong to a specific Cluster.
Used in:
Name of TestRow.
Index within row that belongs to Cluster (refer to columns of the row).
TestGrid column headers. Does not contain the individual cells.
Used in:
Unique instance of the job, typically BUILD_NUMBER from prow or a guid
Name associated with the column (such as the run/invocation ID).No two columns should have the same build_id and name. The name field allows the display of multiple columns with the same build_id.
Milliseconds since start of epoch (python time.time() * 1000) TODO(#683): Use a timestamp, not this double
Additional custom headers like commit, image used, etc.
Custom hotlist ids.
An optional hint for the updater.
Dynamic email list, route email alerts to these instead of the configured defaults.
Status totals for the column. Only written in tab state, if a broken threshold is defined for columns.
A single table of test results backing a dashboard tab.
A cycle of test results, not including the results. In the TestGrid client, the cycles define the columns.
A test case with test results. In the TestGrid client, the cases define the rows (and the results define the individual cells).
The latest configuration used to generate this test group.
Seconds since epoch for last time this cycle was updated.
Stored info on previous timing for parts of the update cycle.
Stored info on default test metadata.
Clusters of failures for a TestResultTable instance.
Most recent timestamp that clusters have processed.
A metric and its values for each test cycle.
Used in:
Name of metric, such as duration
Sparse encoding of values. Indices is a list of pairs of <index, count> that details columns with metric values. So given: Indices: [0, 2, 6, 4] Values: [0.1,0.2,6.1,6.2,6.3,6.4] Decoded 12-value equivalent is: [0.1, 0.2, nil, nil, nil, nil, 6.1, 6.2, 6.3, 6.4, nil, nil, ...]
n=index of first value, n+1=count of filled values
only present for columns with a metric value
Used in:
TestGrid rows
Used in:
Display name, which might process id to append/filter info.
raw id for the row, such as the bazel target or golang package.
Results for this row, run-length encoded to reduce size/improve performance. Thus (encoded -> decoded equivalent): [0, 3, 5, 4] -> [0, 0, 0, 5, 5, 5, 5] [5, 1] -> [5] [1, 5] -> [1, 1, 1, 1, 1] The decoded values are Result enums
Test IDs for each test result in this test case. Must be present on every column, regardless of status.
Short description of the result, displayed on mouseover. Present for any column with a non-empty status (not NO_RESULT).
Names of metrics associated with this test case. Stored separate from metric info (which may be omitted).
Numerical performance/timing data, etc.
Short string to place inside the cell (F for fail, etc) Present for any column with a non-empty status (not NO_RESULT).
IDs for issues associated with this row.
An alert for the failure if there's a recent failure for this row.
Values of a user-defined property found in cells for this row. TODO: Fold this into `properties` field.
General key-value pairs associated with cells in this row. Present for any column with a non-empty status (not NO_RESULT).
Used in:
True if this column has any in-progress runs.
True if a broken threshold is defined, and this column's fail/total ratio exceeds it.
Info on default test metadata for a dashboard tab.
Used in:
Name of the test with associated test metadata.
Default bug component.
Default owner.
Default list of cc's.
When present, only file a bug for failed tests with same error type. Otherwise, always file a bug.
Info on time taken to update test results during the last update cycle.
Used in:
Metrics for how long parts of the update cycle take.
Used in:
The name for a part of the update cycle.
Time taken for a part of the update cycle, in seconds.