summaryrefslogtreecommitdiffstats
path: root/taskcluster/gecko_taskgraph/transforms/l10n.py
blob: e36c70246b87b52082a3faa1dfe1f3d26fbac5b7 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
# 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/.
"""
Do transforms specific to l10n kind
"""


import json

from mozbuild.chunkify import chunkify
from taskgraph.transforms.base import TransformSequence
from taskgraph.util.schema import (
    optionally_keyed_by,
    resolve_keyed_by,
    taskref_or_string,
)
from taskgraph.util.taskcluster import get_artifact_prefix
from taskgraph.util.treeherder import add_suffix
from voluptuous import Any, Optional, Required

from gecko_taskgraph.loader.multi_dep import schema
from gecko_taskgraph.transforms.job import job_description_schema
from gecko_taskgraph.transforms.task import task_description_schema
from gecko_taskgraph.util.attributes import (
    copy_attributes_from_dependent_job,
    task_name,
)
from gecko_taskgraph.util.copy_task import copy_task


def _by_platform(arg):
    return optionally_keyed_by("build-platform", arg)


l10n_description_schema = schema.extend(
    {
        # Name for this job, inferred from the dependent job before validation
        Required("name"): str,
        # build-platform, inferred from dependent job before validation
        Required("build-platform"): str,
        # max run time of the task
        Required("run-time"): _by_platform(int),
        # Locales not to repack for
        Required("ignore-locales"): _by_platform([str]),
        # All l10n jobs use mozharness
        Required("mozharness"): {
            # Script to invoke for mozharness
            Required("script"): _by_platform(str),
            # Config files passed to the mozharness script
            Required("config"): _by_platform([str]),
            # Additional paths to look for mozharness configs in. These should be
            # relative to the base of the source checkout
            Optional("config-paths"): [str],
            # Options to pass to the mozharness script
            Optional("options"): _by_platform([str]),
            # Action commands to provide to mozharness script
            Required("actions"): _by_platform([str]),
            # if true, perform a checkout of a comm-central based branch inside the
            # gecko checkout
            Optional("comm-checkout"): bool,
        },
        # Items for the taskcluster index
        Optional("index"): {
            # Product to identify as in the taskcluster index
            Required("product"): _by_platform(str),
            # Job name to identify as in the taskcluster index
            Required("job-name"): _by_platform(str),
            # Type of index
            Optional("type"): _by_platform(str),
        },
        # Description of the localized task
        Required("description"): _by_platform(str),
        Optional("run-on-projects"): job_description_schema["run-on-projects"],
        # worker-type to utilize
        Required("worker-type"): _by_platform(str),
        # File which contains the used locales
        Required("locales-file"): _by_platform(str),
        # Tooltool visibility required for task.
        Required("tooltool"): _by_platform(Any("internal", "public")),
        # Docker image required for task.  We accept only in-tree images
        # -- generally desktop-build or android-build -- for now.
        Optional("docker-image"): _by_platform(
            # an in-tree generated docker image (from `taskcluster/docker/<name>`)
            {"in-tree": str},
        ),
        Optional("fetches"): {
            str: _by_platform([str]),
        },
        # The set of secret names to which the task has access; these are prefixed
        # with `project/releng/gecko/{treeherder.kind}/level-{level}/`.  Setting
        # this will enable any worker features required and set the task's scopes
        # appropriately.  `true` here means ['*'], all secrets.  Not supported on
        # Windows
        Optional("secrets"): _by_platform(Any(bool, [str])),
        # Information for treeherder
        Required("treeherder"): {
            # Platform to display the task on in treeherder
            Required("platform"): _by_platform(str),
            # Symbol to use
            Required("symbol"): str,
            # Tier this task is
            Required("tier"): _by_platform(int),
        },
        # Extra environment values to pass to the worker
        Optional("env"): _by_platform({str: taskref_or_string}),
        # Max number locales per chunk
        Optional("locales-per-chunk"): _by_platform(int),
        # Task deps to chain this task with, added in transforms from primary-dependency
        # if this is a shippable-style build
        Optional("dependencies"): {str: str},
        # Run the task when the listed files change (if present).
        Optional("when"): {"files-changed": [str]},
        # passed through directly to the job description
        Optional("attributes"): job_description_schema["attributes"],
        Optional("extra"): job_description_schema["extra"],
        # Shipping product and phase
        Optional("shipping-product"): task_description_schema["shipping-product"],
        Optional("shipping-phase"): task_description_schema["shipping-phase"],
    }
)

transforms = TransformSequence()


def parse_locales_file(locales_file, platform=None):
    """Parse the passed locales file for a list of locales."""
    locales = []

    with open(locales_file, mode="r") as f:
        if locales_file.endswith("json"):
            all_locales = json.load(f)
            # XXX Only single locales are fetched
            locales = {
                locale: data["revision"]
                for locale, data in all_locales.items()
                if platform is None or platform in data["platforms"]
            }
        else:
            all_locales = f.read().split()
            # 'default' is the hg revision at the top of hg repo, in this context
            locales = {locale: "default" for locale in all_locales}
    return locales


def _remove_locales(locales, to_remove=None):
    # ja-JP-mac is a mac-only locale, but there are no mac builds being repacked,
    # so just omit it unconditionally
    return {
        locale: revision
        for locale, revision in locales.items()
        if locale not in to_remove
    }


@transforms.add
def setup_name(config, jobs):
    for job in jobs:
        dep = job["primary-dependency"]
        # Set the name to the same as the dep task, without kind name.
        # Label will get set automatically with this kinds name.
        job["name"] = job.get("name", task_name(dep))
        yield job


@transforms.add
def copy_in_useful_magic(config, jobs):
    for job in jobs:
        dep = job["primary-dependency"]
        attributes = copy_attributes_from_dependent_job(dep)
        attributes.update(job.get("attributes", {}))
        # build-platform is needed on `job` for by-build-platform
        job["build-platform"] = attributes.get("build_platform")
        job["attributes"] = attributes
        yield job


transforms.add_validate(l10n_description_schema)


@transforms.add
def setup_shippable_dependency(config, jobs):
    """Sets up a task dependency to the signing job this relates to"""
    for job in jobs:
        job["dependencies"] = {"build": job["dependent-tasks"]["build"].label}
        if job["attributes"]["build_platform"].startswith("win") or job["attributes"][
            "build_platform"
        ].startswith("linux"):
            job["dependencies"].update(
                {
                    "build-signing": job["dependent-tasks"]["build-signing"].label,
                }
            )
        if job["attributes"]["build_platform"].startswith("macosx"):
            job["dependencies"].update(
                {"repackage": job["dependent-tasks"]["repackage"].label}
            )
        yield job


@transforms.add
def handle_keyed_by(config, jobs):
    """Resolve fields that can be keyed by platform, etc."""
    fields = [
        "locales-file",
        "locales-per-chunk",
        "worker-type",
        "description",
        "run-time",
        "docker-image",
        "secrets",
        "fetches.toolchain",
        "fetches.fetch",
        "tooltool",
        "env",
        "ignore-locales",
        "mozharness.config",
        "mozharness.options",
        "mozharness.actions",
        "mozharness.script",
        "treeherder.tier",
        "treeherder.platform",
        "index.type",
        "index.product",
        "index.job-name",
        "when.files-changed",
    ]
    for job in jobs:
        job = copy_task(job)  # don't overwrite dict values here
        for field in fields:
            resolve_keyed_by(item=job, field=field, item_name=job["name"])
        yield job


@transforms.add
def handle_artifact_prefix(config, jobs):
    """Resolve ``artifact_prefix`` in env vars"""
    for job in jobs:
        artifact_prefix = get_artifact_prefix(job)
        for k1, v1 in job.get("env", {}).items():
            if isinstance(v1, str):
                job["env"][k1] = v1.format(artifact_prefix=artifact_prefix)
            elif isinstance(v1, dict):
                for k2, v2 in v1.items():
                    job["env"][k1][k2] = v2.format(artifact_prefix=artifact_prefix)
        yield job


@transforms.add
def all_locales_attribute(config, jobs):
    for job in jobs:
        locales_platform = job["attributes"]["build_platform"].replace("-shippable", "")
        locales_platform = locales_platform.replace("-pgo", "")
        locales_with_changesets = parse_locales_file(
            job["locales-file"], platform=locales_platform
        )
        locales_with_changesets = _remove_locales(
            locales_with_changesets, to_remove=job["ignore-locales"]
        )

        locales = sorted(locales_with_changesets.keys())
        attributes = job.setdefault("attributes", {})
        attributes["all_locales"] = locales
        attributes["all_locales_with_changesets"] = locales_with_changesets
        if job.get("shipping-product"):
            attributes["shipping_product"] = job["shipping-product"]
        yield job


@transforms.add
def chunk_locales(config, jobs):
    """Utilizes chunking for l10n stuff"""
    for job in jobs:
        locales_per_chunk = job.get("locales-per-chunk")
        locales_with_changesets = job["attributes"]["all_locales_with_changesets"]
        if locales_per_chunk:
            chunks, remainder = divmod(len(locales_with_changesets), locales_per_chunk)
            if remainder:
                chunks = int(chunks + 1)
            for this_chunk in range(1, chunks + 1):
                chunked = copy_task(job)
                chunked["name"] = chunked["name"].replace("/", f"-{this_chunk}/", 1)
                chunked["mozharness"]["options"] = chunked["mozharness"].get(
                    "options", []
                )
                # chunkify doesn't work with dicts
                locales_with_changesets_as_list = sorted(
                    locales_with_changesets.items()
                )
                chunked_locales = chunkify(
                    locales_with_changesets_as_list, this_chunk, chunks
                )
                chunked["mozharness"]["options"].extend(
                    [
                        f"locale={locale}:{changeset}"
                        for locale, changeset in chunked_locales
                    ]
                )
                chunked["attributes"]["l10n_chunk"] = str(this_chunk)
                # strip revision
                chunked["attributes"]["chunk_locales"] = [
                    locale for locale, _ in chunked_locales
                ]

                # add the chunk number to the TH symbol
                chunked["treeherder"]["symbol"] = add_suffix(
                    chunked["treeherder"]["symbol"], this_chunk
                )
                yield chunked
        else:
            job["mozharness"]["options"] = job["mozharness"].get("options", [])
            job["mozharness"]["options"].extend(
                [
                    f"locale={locale}:{changeset}"
                    for locale, changeset in sorted(locales_with_changesets.items())
                ]
            )
            yield job


transforms.add_validate(l10n_description_schema)


@transforms.add
def stub_installer(config, jobs):
    for job in jobs:
        job.setdefault("attributes", {})
        job.setdefault("env", {})
        if job["attributes"].get("stub-installer"):
            job["env"].update({"USE_STUB_INSTALLER": "1"})
        yield job


@transforms.add
def set_extra_config(config, jobs):
    for job in jobs:
        job["mozharness"].setdefault("extra-config", {})["branch"] = config.params[
            "project"
        ]
        if "update-channel" in job["attributes"]:
            job["mozharness"]["extra-config"]["update_channel"] = job["attributes"][
                "update-channel"
            ]
        yield job


@transforms.add
def make_job_description(config, jobs):
    for job in jobs:
        job["mozharness"].update(
            {
                "using": "mozharness",
                "job-script": "taskcluster/scripts/builder/build-l10n.sh",
                "secrets": job.get("secrets", False),
            }
        )
        job_description = {
            "name": job["name"],
            "worker-type": job["worker-type"],
            "description": job["description"],
            "run": job["mozharness"],
            "attributes": job["attributes"],
            "treeherder": {
                "kind": "build",
                "tier": job["treeherder"]["tier"],
                "symbol": job["treeherder"]["symbol"],
                "platform": job["treeherder"]["platform"],
            },
            "run-on-projects": job.get("run-on-projects")
            if job.get("run-on-projects")
            else [],
        }
        if job.get("extra"):
            job_description["extra"] = job["extra"]

        job_description["run"]["tooltool-downloads"] = job["tooltool"]

        job_description["worker"] = {
            "max-run-time": job["run-time"],
            "chain-of-trust": True,
        }
        if job["worker-type"] == "b-win2012":
            job_description["worker"]["os"] = "windows"
            job_description["run"]["use-simple-package"] = False
            job_description["run"]["use-magic-mh-args"] = False

        if job.get("docker-image"):
            job_description["worker"]["docker-image"] = job["docker-image"]

        if job.get("fetches"):
            job_description["fetches"] = job["fetches"]

        if job.get("index"):
            job_description["index"] = {
                "product": job["index"]["product"],
                "job-name": job["index"]["job-name"],
                "type": job["index"].get("type", "generic"),
            }

        if job.get("dependencies"):
            job_description["dependencies"] = job["dependencies"]
        if job.get("env"):
            job_description["worker"]["env"] = job["env"]
        if job.get("when", {}).get("files-changed"):
            job_description.setdefault("when", {})
            job_description["when"]["files-changed"] = [job["locales-file"]] + job[
                "when"
            ]["files-changed"]

        if "shipping-phase" in job:
            job_description["shipping-phase"] = job["shipping-phase"]

        if "shipping-product" in job:
            job_description["shipping-product"] = job["shipping-product"]

        yield job_description