We require python3 to get new library features and things and to cut ties with 2.7 since we are going to have to do it anyway EOY.
pip3 install --user virtualenv
python3 -m virtualenv venv
. venv/bin/activate
pip install -r requirements.txt$ ./activedata_compare_tests.py --help
usage: activedata_compare_tests.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID] [--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--treeherder-url TREEHERDER_URL]
[--activedata ACTIVEDATA]
[--combine-chunks]
[--output-push-differences-only]
ActiveData compare-tests
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--activedata ACTIVEDATA
ActiveData url. (default: https://activedata.allizom.org/query)
--combine-chunks Combine chunks (default: False)
--output-push-differences-only
When loading multiple pushes, only output keys which have different
values for sub_keys across the
pushes. (default: False)
You can save a set of arguments to a file and specify them later
using the @argfile syntax. The arguments contained in the file will
replace @argfile in the command line. Multiple files can be loaded
into the command line through the use of the @ syntax. Each argument
and its value must be on separate lines in the file.writes results of ActiveData query in json format to stdout.
$ ./activedata_query.py --help
usage: activedata_query.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--activedata ACTIVEDATA] --file FILE [--raw]
Query ActiveData tests and write the result as json to stdout.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--activedata ACTIVEDATA
ActiveData url. (default: https://activedata.allizom.org/query)
--file FILE File containing ActiveData query as json.. (default: None)
--raw Do not reformat/indent json. (default: False)
You can save a set of arguments to a file and specify them later
using the @argfile syntax. The arguments contained in the file will
replace @argfile in the command line. Multiple files can be loaded
into the command line through the use of the @ syntax. Each argument
and its value must be on separate lines in the file.$ ./activedata_query_tests.py --help
usage: activedata_query_tests.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID] [--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--treeherder-url TREEHERDER_URL]
[--activedata ACTIVEDATA]
[--include-passing-tests] [--raw]
ActiveData query tests.
Query ActiveData tests and write the result as json to stdout.
Errors will be written to stderr.
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--activedata ACTIVEDATA
ActiveData url. (default: https://activedata.allizom.org/query)
--include-passing-tests
Query tests against ActiveData. (default: False)
--raw Do not reformat/indent json. (default: False)
You can save a set of arguments to a file and specify them later
using the @argfile syntax. The arguments contained in the file will
replace @argfile in the command line. Multiple files can be loaded
into the command line through the use of the @ syntax. Each argument
and its value must be on separate lines in the file.$ ./create_log_summaries.py --help
usage: create_log_summaries.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
--path PATH [--filename FILENAME]
[--include-tests] [--dechunk] [--raw]
Analyze downloaded Test Log files producing json summaries..
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--path PATH Log. (default: None)
--filename FILENAME Base log filename suffix. (default: live_backing.log)
--include-tests Include TEST- lines. (default: False)
--dechunk Combine chunks. (default: False)
--raw Do not reformat/indent json. (default: False)
You can save a set of arguments to a file and specify them later using
the @argfile syntax. The arguments contained in the file will replace
@argfile in the command line. Multiple files can be loaded into the
command line through the use of the @ syntax.
Each argument and its value must be on separate lines in the file.$ ./combine_log_summaries.py --help
usage: combine_log_summaries.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--file FILES] [--alias ALIASES]
[--differences] [--ignore IGNORE]
[--munge-test-data]
Combine analyzed Test Log json files.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--file FILES
--alias ALIASES
--differences Output only differences in data. (default: False)
--ignore IGNORE Ignore keys matching regular expression when calculating differences. (default: None)
--munge-test-data Modify TEST- lines in output to improve comparibility. (default: False)
You can save a set of arguments to a file and specify them later
using the @argfile syntax. The arguments contained in the file will
replace @argfile in the command line. Multiple files can be loaded
into the command line through the use of the @ syntax. Each argument
and its value must be on separate lines in the file.$ ./download_treeherder_jobdetails.py --help
usage: download_treeherder_jobdetails.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID] [--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--add-bugzilla-suggestions]
[--test-failure-pattern TEST_FAILURE_PATTERN]
[--build-platform BUILD_PLATFORM]
[--job-group-name JOB_GROUP_NAME]
[--job-group-symbol JOB_GROUP_SYMBOL]
[--job-type-name JOB_TYPE_NAME]
[--job-type-symbol JOB_TYPE_SYMBOL]
[--machine-name MACHINE_NAME]
[--platform PLATFORM]
[--platform-option PLATFORM_OPTION]
[--result RESULT] [--state STATE]
[--tier TIER]
[--treeherder-url TREEHERDER_URL]
--download-job-details
DOWNLOAD_JOB_DETAILS
[--output OUTPUT] [--alias ALIAS]
Download Job Details files from Treeherder/Taskcluster.
--download-job-details specifies a regular expression which will be matched
against the base file name of the url to the file to select the files to be
downloaded. This is not a shell glob pattern, but a full regular expression.
Files will be saved to the output directory using the path to the job detail
and a file name encoded with meta data as:
output/revision/job_guid/job_guid_run/path/platform,buildtype,job_name,job_type_symbol,filename
if --alias is specified, a soft link will be created from
output/revision to output/alias.
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
Job Related Arguments
Job related pattern objects are used to select the jobs which will be
returned. All specified patterns must match to return a job.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--add-bugzilla-suggestions
Add bugzilla suggestions to job objects. (default: False)
--test-failure-pattern TEST_FAILURE_PATTERN
Include failures from bugzilla suggestions matching this regular expression. (default: None)
--build-platform BUILD_PLATFORM
Match job build platform regular expression. (default: None)
--job-group-name JOB_GROUP_NAME
Match job group name regular expression. (default: None)
--job-group-symbol JOB_GROUP_SYMBOL
Match job group symbol regular expression (default: None)
--job-type-name JOB_TYPE_NAME
Match job type name regular expression. (default: None)
--job-type-symbol JOB_TYPE_SYMBOL
Match job type symbol regular expression. (default: None)
--machine-name MACHINE_NAME
Match job machine name regular expression. (default: None)
--platform PLATFORM Match job platform regular expression. (default: None)
--platform-option PLATFORM_OPTION
Match job platform option regular expression: opt, debug, pgo,... (default: None)
--result RESULT Match job result regular expression: unknown, success, testfailed, .... (default: None)
--state STATE Match job state regular expression: pending, running, completed. (default: None)
--tier TIER Match job tier regular expression. (default: None)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--download-job-details DOWNLOAD_JOB_DETAILS
Regular expression matching Job details url basenames to be
downloaded. Example:live_backing.log|logcat.*.log. Default
None. (default: None)
--output OUTPUT Directory where to save downloaded job details. (default: output)
--alias ALIAS Alias (soft link) to revision subdirectory where the downloaded job details were saved. (default: None)
You can save a set of arguments to a file and specify them later using
the @argfile syntax. The arguments contained in the file will replace
@argfile in the command line. Multiple files can be loaded into the
command line through the use of the @ syntax.
Each argument and its value must be on separate lines in the file.Compare opt and pgo tasks and flag duplicates.
./mach taskgraph tasks --json | ./duplicate-opt-pgo-tasks.py
./duplicate-opt-pgo-tasks.py tasks.json./get_pushes_jobs_job_details_json.py --help
usage: get_pushes_jobs_job_details_json.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID]
[--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--add-bugzilla-suggestions]
[--test-failure-pattern TEST_FAILURE_PATTERN]
[--build-platform BUILD_PLATFORM]
[--job-group-name JOB_GROUP_NAME]
[--job-group-symbol JOB_GROUP_SYMBOL]
[--job-type-name JOB_TYPE_NAME]
[--job-type-symbol JOB_TYPE_SYMBOL]
[--machine-name MACHINE_NAME]
[--platform PLATFORM]
[--platform-option PLATFORM_OPTION]
[--result RESULT] [--state STATE]
[--tier TIER]
[--treeherder-url TREEHERDER_URL]
[--add-resource-usage] [--raw]
Downloads pushes, jobs and job details data from Treeherder, writing results as
nested json to stdout.
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
Job Related Arguments
Job related pattern objects are used to select the jobs which will be
returned. All specified patterns must match to return a job.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--add-bugzilla-suggestions
Add bugzilla suggestions to job objects. (default: False)
--test-failure-pattern TEST_FAILURE_PATTERN
Include failures from bugzilla suggestions matching this regular expression. (default: None)
--build-platform BUILD_PLATFORM
Match job build platform regular expression. (default: None)
--job-group-name JOB_GROUP_NAME
Match job group name regular expression. (default: None)
--job-group-symbol JOB_GROUP_SYMBOL
Match job group symbol regular expression (default: None)
--job-type-name JOB_TYPE_NAME
Match job type name regular expression. (default: None)
--job-type-symbol JOB_TYPE_SYMBOL
Match job type symbol regular expression. (default: None)
--machine-name MACHINE_NAME
Match job machine name regular expression. (default: None)
--platform PLATFORM Match job platform regular expression. (default: None)
--platform-option PLATFORM_OPTION
Match job platform option regular expression: opt, debug, pgo,... (default: None)
--result RESULT Match job result regular expression: unknown, success, testfailed, .... (default: None)
--state STATE Match job state regular expression: pending, running, completed. (default: None)
--tier TIER Match job tier regular expression. (default: None)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--add-resource-usage Download resource-usage.json job detail and add to job object. (default: False)
--raw Do not reformat/indent json. (default: False)
You can save a set of arguments to a file and specify them later using
the @argfile syntax. The arguments contained in the file will replace
@argfile in the command line. Multiple files can be loaded into the
command line through the use of the @ syntax.
Each argument and its value must be on separate lines in the file../get_pushes_json.py --help
usage: get_pushes_json.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID] [--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--treeherder-url TREEHERDER_URL] [--raw]
Downloads pushes data from Treeherder, writing results as
json to stdout.
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--raw Do not reformat/indent json. (default: False)
You can save a set of arguments to a file and specify them later using
the @argfile syntax. The arguments contained in the file will replace
@argfile in the command line. Multiple files can be loaded into the
command line through the use of the @ syntax.
Each argument and its value must be on separate lines in the file../get_pushes_jobs_json.py --help
usage: get_pushes_jobs_json.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--treeherder-url TREEHERDER_URL]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID] [--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--add-bugzilla-suggestions]
[--test-failure-pattern TEST_FAILURE_PATTERN]
[--build-platform BUILD_PLATFORM]
[--job-group-name JOB_GROUP_NAME]
[--job-group-symbol JOB_GROUP_SYMBOL]
[--job-type-name JOB_TYPE_NAME]
[--job-type-symbol JOB_TYPE_SYMBOL]
[--machine-name MACHINE_NAME]
[--platform PLATFORM]
[--platform-option PLATFORM_OPTION]
[--result RESULT] [--state STATE] [--tier TIER]
[--raw]
Downloads pushes and jobs data from Treeherder, writing results as nested json to
stdout.
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
Job Related Arguments
Job related pattern objects are used to select the jobs which will be
returned. All specified patterns must match to return a job.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--add-bugzilla-suggestions
Add bugzilla suggestions to job objects. (default: False)
--test-failure-pattern TEST_FAILURE_PATTERN
Include failures from bugzilla suggestions matching this regular expression. (default: None)
--build-platform BUILD_PLATFORM
Match job build platform regular expression. (default: None)
--job-group-name JOB_GROUP_NAME
Match job group name regular expression. (default: None)
--job-group-symbol JOB_GROUP_SYMBOL
Match job group symbol regular expression (default: None)
--job-type-name JOB_TYPE_NAME
Match job type name regular expression. (default: None)
--job-type-symbol JOB_TYPE_SYMBOL
Match job type symbol regular expression. (default: None)
--machine-name MACHINE_NAME
Match job machine name regular expression. (default: None)
--platform PLATFORM Match job platform regular expression. (default: None)
--platform-option PLATFORM_OPTION
Match job platform option regular expression: opt, debug, pgo,... (default: None)
--result RESULT Match job result regular expression: unknown, success, testfailed, .... (default: None)
--state STATE Match job state regular expression: pending, running, completed. (default: None)
--tier TIER Match job tier regular expression. (default: None)
--raw Do not reformat/indent json. (default: False)
You can save a set of arguments to a file and specify them later using
the @argfile syntax. The arguments contained in the file will replace
@argfile in the command line. Multiple files can be loaded into the
command line through the use of the @ syntax.
Each argument and its value must be on separate lines in the file.Summarize job json for Test Isolation
$ ./summarize_isolation_pushes_jobs_json.py --help
usage: summarize_isolation_pushes_jobs_json.py [-h]
[--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
[--treeherder-url TREEHERDER_URL]
[--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}]
[--push_id PUSH_ID]
[--author AUTHOR]
[--comments COMMENTS]
[--date-range DATE_RANGE | --revision REVISION | --commit-revision COMMIT_REVISION | --revision-url REVISION_URL | --revision-range REVISION_RANGE]
[--add-bugzilla-suggestions]
[--test-failure-pattern TEST_FAILURE_PATTERN]
[--build-platform BUILD_PLATFORM]
[--job-group-name JOB_GROUP_NAME]
[--job-group-symbol JOB_GROUP_SYMBOL]
[--job-type-name JOB_TYPE_NAME]
[--job-type-symbol JOB_TYPE_SYMBOL]
[--machine-name MACHINE_NAME]
[--platform PLATFORM]
[--platform-option PLATFORM_OPTION]
[--result RESULT]
[--state STATE] [--tier TIER]
[--whiteboard WHITEBOARD]
[--override-bug-summary OVERRIDE_BUG_SUMMARY]
[--cache CACHE]
[--update-cache]
[--dump-cache-stats]
[--bug-creation-time BUG_CREATION_TIME]
[--bugs-after BUGS_AFTER]
[--bug BUGS] [--raw]
[--csv-summary] [--csv-results]
[--include-failures]
[--include-tests]
Analyze pushes from bugs marked with whiteboard [test isolation] or a
value specified from the command options.
Queries Bugzilla for bugs marked with [test isolation] in the whiteboard,
determines the bug number, bug summary and revision from the bug then reads
push and job data from Treeherder and produces a summary of runtimes and
test failures, writing results as either csv text or json to stdout. By
default, output is writtenas formatted json.
Intermediate results are stored in a cache directory to re-used on subsequent
runs. When changing options, it is safest to delete the cache directory and
start over.
Push Related Arguments
If a push isn't selected, the most recent push will be returned.
Job Related Arguments
Job related pattern objects are used to select the jobs which will be
returned. All specified patterns must match to return a job.
optional arguments:
-h, --help show this help message and exit
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
Logging level. (default: INFO)
--treeherder-url TREEHERDER_URL
Treeherder url. (default: https://treeherder.mozilla.org)
--repo {mozilla-central,autoland,mozilla-inbound,try,mozilla-beta,mozilla-release,mozilla-esr68}
repository name to query. (default: mozilla-central)
--push_id PUSH_ID Push id. (default: None)
--author AUTHOR Push author email. Should be specified if --repo is try and more
than one revision is selected. (default: None)
--comments COMMENTS Push comments pattern. (default: None)
--date-range DATE_RANGE
Push date range startdate enddate CCYY-MM-DD CCYY-MM-DD. (default: None)
--revision REVISION Push Revision. (default: None)
--commit-revision COMMIT_REVISION
Either Push Revision or any commit referenced in the push. (default: None)
--revision-url REVISION_URL
Url to push revision which can be used in place of --repo and --revision. (default: None)
--revision-range REVISION_RANGE
Push revision range fromchange-tochange. (default: None)
--add-bugzilla-suggestions
Add bugzilla suggestions to job objects. (default: False)
--test-failure-pattern TEST_FAILURE_PATTERN
Include failures from bugzilla suggestions matching this regular expression. (default: None)
--build-platform BUILD_PLATFORM
Match job build platform regular expression. (default: None)
--job-group-name JOB_GROUP_NAME
Match job group name regular expression. (default: None)
--job-group-symbol JOB_GROUP_SYMBOL
Match job group symbol regular expression (default: None)
--job-type-name JOB_TYPE_NAME
Match job type name regular expression. (default: None)
--job-type-symbol JOB_TYPE_SYMBOL
Match job type symbol regular expression. (default: None)
--machine-name MACHINE_NAME
Match job machine name regular expression. (default: None)
--platform PLATFORM Match job platform regular expression. (default: None)
--platform-option PLATFORM_OPTION
Match job platform option regular expression: opt, debug, pgo,... (default: None)
--result RESULT Match job result regular expression: unknown, success, testfailed, .... (default: None)
--state STATE Match job state regular expression: pending, running, completed. (default: None)
--tier TIER Match job tier regular expression. (default: None)
--whiteboard WHITEBOARD
Bugzilla whiteboard value used to select the appropriate bugs. Should only be used with --bug. (default: [test isolation])
--override-bug-summary OVERRIDE_BUG_SUMMARY
When reprocessing a bug with a problematic bug summary or when using --whiteboard to select a bug not filed by intermittent-bug-filer, specify an override bug summary to mimic an intermittent bug summary to be used to determine if a failure or test is reproduced. Otherwise the original bug summary will be used. Should only be used with --bug. (default: None)
--cache CACHE Directory used to store cached objects retrieved from Bugzilla and Treeherder. (default: ~/cia_tools_cache/)
--update-cache Recreate cached files with fresh data. (default: False)
--dump-cache-stats Dump cache statistics to stderr. (default: False)
--bug-creation-time BUG_CREATION_TIME
Starting creation time in YYYY-MM-DD or YYYY-MM-DDTHH:MM:SSTZ format. Example 2019-07-27T17:28:00PDT or 2019-07-28T00:28:00Z (default: 2019-06-14)
--bugs-after BUGS_AFTER
Only returns bugs whose id is greater than this integer. (default: None)
--bug BUGS Only returns results for bug the specified bug. (default: [])
--raw Do not reformat/indent json. (default: False)
--csv-summary Output summary data in csv format. Does not include individual failures or tests. (default: False)
--csv-results Output test data in csv format. Does not include individual failures. (default: False)
--include-failures Include individual failures in output. (default: False)
--include-tests Include individual tests in output. (default: False)
You can save a set of arguments to a file and specify them later using
the @argfile syntax. The arguments contained in the file will replace
@argfile in the command line. Multiple files can be loaded into the
command line through the use of the @ syntax.
Each argument and its value must be on separate lines in the file.$ ./tasks_costs.py --help
usage: tasks_costs.py [-h] --costs COSTS --tasks TASKS [--project PROJECTS]
[--tier TIERS] [--json] [--verbose]
Associate Task Labels to Task costs.
optional arguments:
-h, --help show this help message and exit
--costs COSTS Path to json file containing costs.
--tasks TASKS Path to json file containing tasks.
--project PROJECTS One or more of mozilla-central, autoland, ... If not
specified, returns all projects.
--tier TIERS One or more of 1, 2, 3. If not specified, returns all
projects.
--json Output results in json format.
--verbose Output no cost warnings.
This script matches a costs json file to a tasks json file to
associate a task label to costs and writes a new costs file with the
attached task label.
To obtain the costs file execute a BigQuery containing an sql query of
the form
select w.provisionerId, w.workerType, s.project, s.tier,
s.suite, s.groupSymbol, s.symbol, s.collection,
sum(s.execution*1000.0*w.cost_per_ms) as cost
from taskclusteretl.derived_task_summary as s,
taskclusteretl.derived_daily_cost_per_workertype as w
where s.workerType = w.workerType
and s.provisionerid = w.provisionerid
and s.date = w.date
and s.date between '2019-12-02' and '2020-01-06'
and s.project in ('autoland', 'try')
and s.tier in (2, 3)
group by provisionerId, workerType, project, tier, suite, groupSymbol, symbol, collection
order by provisionerId, workerType, project, tier, suite, groupSymbol, symbol, collection;
and save the results locally to a costs.json file.
Using a source checkout, obtain a tasks json file via
./mach taskgraph tasks --json > /tmp/tasks.json
Then execute
python tasks_costs.py --costs=costs.json --tasks=tasks.json 2> costs.err > costs-annotated.csv
or
python tasks_costs.py --costs=costs.json --tasks=tasks.json --json 2> costs.err > costs-annotated.json