summaryrefslogtreecommitdiffstats
path: root/taskcluster/taskgraph/task/transform.py
diff options
context:
space:
mode:
authorMatt A. Tobin <mattatobin@localhost.localdomain>2018-02-02 04:16:08 -0500
committerMatt A. Tobin <mattatobin@localhost.localdomain>2018-02-02 04:16:08 -0500
commit5f8de423f190bbb79a62f804151bc24824fa32d8 (patch)
tree10027f336435511475e392454359edea8e25895d /taskcluster/taskgraph/task/transform.py
parent49ee0794b5d912db1f95dce6eb52d781dc210db5 (diff)
downloadUXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.gz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.lz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.xz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.zip
Add m-esr52 at 52.6.0
Diffstat (limited to 'taskcluster/taskgraph/task/transform.py')
-rw-r--r--taskcluster/taskgraph/task/transform.py109
1 files changed, 109 insertions, 0 deletions
diff --git a/taskcluster/taskgraph/task/transform.py b/taskcluster/taskgraph/task/transform.py
new file mode 100644
index 000000000..8183254a0
--- /dev/null
+++ b/taskcluster/taskgraph/task/transform.py
@@ -0,0 +1,109 @@
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+from __future__ import absolute_import, print_function, unicode_literals
+
+import logging
+import itertools
+
+from . import base
+from .. import files_changed
+from ..util.python_path import find_object
+from ..util.templates import merge
+from ..util.yaml import load_yaml
+from ..util.seta import is_low_value_task
+
+from ..transforms.base import TransformSequence, TransformConfig
+
+logger = logging.getLogger(__name__)
+
+
+class TransformTask(base.Task):
+ """
+ Tasks of this class are generated by applying transformations to a sequence
+ of input entities. By default, it gets those inputs from YAML data in the
+ kind directory, but subclasses may override `get_inputs` to produce them in
+ some other way.
+ """
+
+ @classmethod
+ def get_inputs(cls, kind, path, config, params, loaded_tasks):
+ """
+ Get the input elements that will be transformed into tasks. The
+ elements themselves are free-form, and become the input to the first
+ transform.
+
+ By default, this reads jobs from the `jobs` key, or from yaml files
+ named by `jobs-from`. The entities are read from mappings, and the
+ keys to those mappings are added in the `name` key of each entity.
+
+ If there is a `job-defaults` config, then every job is merged with it.
+ This provides a simple way to set default values for all jobs of a
+ kind. More complex defaults should be implemented with custom
+ transforms.
+
+ This method can be overridden in subclasses that need to perform more
+ complex calculations to generate the list of inputs.
+ """
+ def jobs():
+ defaults = config.get('job-defaults')
+ jobs = config.get('jobs', {}).iteritems()
+ jobs_from = itertools.chain.from_iterable(
+ load_yaml(path, filename).iteritems()
+ for filename in config.get('jobs-from', {}))
+ for name, job in itertools.chain(jobs, jobs_from):
+ if defaults:
+ job = merge(defaults, job)
+ yield name, job
+
+ for name, job in jobs():
+ job['name'] = name
+ logger.debug("Generating tasks for {} {}".format(kind, name))
+ yield job
+
+ @classmethod
+ def load_tasks(cls, kind, path, config, params, loaded_tasks):
+ inputs = cls.get_inputs(kind, path, config, params, loaded_tasks)
+
+ transforms = TransformSequence()
+ for xform_path in config['transforms']:
+ transform = find_object(xform_path)
+ transforms.add(transform)
+
+ # perform the transformations
+ trans_config = TransformConfig(kind, path, config, params)
+ tasks = [cls(kind, t) for t in transforms(trans_config, inputs)]
+ return tasks
+
+ def __init__(self, kind, task):
+ self.dependencies = task['dependencies']
+ self.when = task['when']
+ super(TransformTask, self).__init__(kind, task['label'],
+ task['attributes'], task['task'])
+
+ def get_dependencies(self, taskgraph):
+ return [(label, name) for name, label in self.dependencies.items()]
+
+ def optimize(self, params):
+ if 'files-changed' in self.when:
+ changed = files_changed.check(
+ params, self.when['files-changed'])
+ if not changed:
+ logger.debug('no files found matching a pattern in `when.files-changed` for ' +
+ self.label)
+ return True, None
+
+ # we would like to return 'False, None' while it's high_value_task
+ # and we wouldn't optimize it. Otherwise, it will return 'True, None'
+ if is_low_value_task(self.label, params.get('project')):
+ # Always optimize away low-value tasks
+ return True, None
+ else:
+ return False, None
+
+ @classmethod
+ def from_json(cls, task_dict):
+ # when reading back from JSON, we lose the "when" information
+ task_dict['when'] = {}
+ return cls(task_dict['attributes']['kind'], task_dict)