Fix: Can't use curl to download a single manifest in one invocation (#5918)#6099
Fix: Can't use curl to download a single manifest in one invocation (#5918)#6099dsotirho-ucsc wants to merge 9 commits intodevelopfrom
Conversation
8b62dbc to
830a1e4
Compare
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #6099 +/- ##
===========================================
- Coverage 85.22% 85.20% -0.03%
===========================================
Files 156 156
Lines 22408 22451 +43
===========================================
+ Hits 19098 19130 +32
- Misses 3310 3321 +11 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
da860a5 to
b81d052
Compare
9fb4dbb to
391cedb
Compare
|
Successful use of curl to download a manifest in one invocation: (Unabridged version: 6099_manifest_invocation_unabridged.txt) |
lambdas/service/app.py
Outdated
|
|
||
| def manifest_route(*, fetch: bool, initiate: bool): | ||
| def manifest_route(*, fetch: bool, method: str): | ||
| initiate = method in ['PUT', 'POST'] |
There was a problem hiding this comment.
| initiate = method in ['PUT', 'POST'] | |
| initiate = method in {'PUT', 'POST'} |
lambdas/service/app.py
Outdated
|
|
||
| def _file_manifest(fetch: bool, token_or_key: Optional[str] = None): | ||
| request = app.current_request | ||
| require(request.method != 'POST' or request.raw_body.decode() == '', |
There was a problem hiding this comment.
This is the same as
| require(request.method != 'POST' or request.raw_body.decode() == '', | |
| require(request.method != 'POST' or request.raw_body == b'', |
Right?
|
Added #6003 as a blocker due to overlapping changes to the IT |
600eb8b to
81a4905
Compare
nadove-ucsc
left a comment
There was a problem hiding this comment.
Looks good but needs rebase
81a4905 to
ff39020
Compare
|
|
||
| def server_side_sleep(self, max_seconds: int | float) -> float: | ||
| """ | ||
| Sleep in the Lambda. |
There was a problem hiding this comment.
See my comment regarding the "Lambda" vs "Lambda Function" terminology on your other PR.
| Sleep in the Lambda. | ||
|
|
||
| :param max_seconds: The requested number of seconds to sleep. The actual | ||
| time slept will be less however if the requested |
There was a problem hiding this comment.
| time slept will be less however if the requested | |
| time slept will be less if the requested |
| number would cause the Lambda to exceed its | ||
| execution timeout. | ||
|
|
||
| :return: The actual number of seconds slept |
There was a problem hiding this comment.
"number" suggests an integral amount which is not what is being returned.
test/integration_test.py
Outdated
| first_fetch = bool(self.random.getrandbits(1)) | ||
| fetch_modes = [first_fetch, not first_fetch] | ||
| for fetch in fetch_modes: | ||
| with self.subTest('manifest', | ||
| catalog=catalog, | ||
| format=format, | ||
| curl=curl, | ||
| fetch=fetch, | ||
| wait=wait): | ||
| args = dict(catalog=catalog, | ||
| filters=json.dumps(filters), | ||
| **({} if wait is None else {'wait': wait})) | ||
| if format is None: | ||
| format = first(supported_formats) | ||
| else: | ||
| args['format'] = format.value | ||
|
|
||
| # Wrap self._get_url to collect all HTTP responses | ||
| _get_url = self._get_url | ||
| responses = list() | ||
|
|
||
| def get_url(*args, **kwargs): | ||
| response = _get_url(*args, **kwargs) | ||
| responses.append(response) | ||
| return response | ||
|
|
||
| with mock.patch.object(self, '_get_url', new=get_url): | ||
|
|
||
| # Make multiple identical concurrent requests to | ||
| # test the idempotence of manifest generation, | ||
| # and its resilience against DOS attacks. | ||
|
|
||
| def worker(_): | ||
| response = self._check_endpoint(POST if curl else PUT, | ||
| '/manifest/files', | ||
| args=args, | ||
| fetch=fetch) | ||
| self._manifest_validators[format](catalog, response) | ||
|
|
||
| num_workers = 3 | ||
| with ThreadPoolExecutor(max_workers=num_workers) as tpe: | ||
| results = list(tpe.map(worker, range(num_workers))) | ||
|
|
||
| self.assertEqual([None] * num_workers, results) | ||
| execution_ids = self._manifest_execution_ids(responses, fetch=fetch) | ||
| # The second iteration of the inner-most loop re- | ||
| # requests the manifest with only `fetch` being | ||
| # different. In that case, the manifest will already | ||
| # be cached and no step function execution is | ||
| # expected to have been started. | ||
| expect_execution = fetch == first_fetch | ||
| self.assertEqual(1 if expect_execution else 0, len(execution_ids)) |
There was a problem hiding this comment.
Due to the added indention, this is getting too narrow. Please refactor in an early commit.
test/service/test_manifest.py
Outdated
| def test(*, | ||
| format: ManifestFormat, | ||
| fetch: bool, | ||
| curl: Optional[bool] = False, |
There was a problem hiding this comment.
What would curl=None signify?
As always, assert that no invalid combinations are passed.
There was a problem hiding this comment.
What would curl=None signify?
Good point, this should not be optional.
test/integration_test.py
Outdated
| with self.subTest('manifest', | ||
| catalog=catalog, | ||
| format=format, | ||
| fetch=fetch, | ||
| curl=curl, | ||
| wait=wait): | ||
| execution_ids = self._test_manifest_execution(catalog, | ||
| format, | ||
| filters, | ||
| fetch, | ||
| curl, | ||
| wait) | ||
| # The second iteration of the inner-most loop re- | ||
| # requests the manifest with only `fetch` being | ||
| # different. In that case, the manifest will already | ||
| # be cached and no step function execution is | ||
| # expected to have been started. | ||
| expect_execution = fetch == first_fetch | ||
| self.assertEqual(1 if expect_execution else 0, len(execution_ids)) |
There was a problem hiding this comment.
This refactoring didn't have the desired effect of widening the horizontal space for this comment. The invocation of the introduced method also looks awkward. We typically switch to keyword arguments for a function/method with this many arguments. I don't think extracting a method is the solution. See if you can come up with some other solution to reduce the nesting.
hannes-ucsc
left a comment
There was a problem hiding this comment.
As always, the refactoring should be in a separate commit.
The refactoring only applied to my addition as doesn't work in this case as a separate prior commit. Instead of nesting the existing |
test/integration_test.py
Outdated
| for curl, wait, format in [(c, w, f) | ||
| for c in [False, True] | ||
| for w in ([None, 0, 1] if c else [None]) | ||
| for f in [None, *supported_formats]]: |
There was a problem hiding this comment.
This can be done more succinctly with itertools.product
test/integration_test.py
Outdated
| with self.subTest('manifest', | ||
| catalog=catalog, | ||
| format=format, | ||
| fetch=fetch, | ||
| curl=curl, | ||
| wait=wait): |
There was a problem hiding this comment.
Let's include subTest in the list of exceptions to https://github.com/DataBiosphere/azul/blob/develop/CONTRIBUTING.rst#line-wrapping-and-indentation so that this can be formatted as
| with self.subTest('manifest', | |
| catalog=catalog, | |
| format=format, | |
| fetch=fetch, | |
| curl=curl, | |
| wait=wait): | |
| with self.subTest('manifest', catalog=catalog, format=format, | |
| fetch=fetch, curl=curl, wait=wait): |
The update to CONTRIBUTING.rst should of course be in a separate commit.
test/service/test_manifest.py
Outdated
| fetch: bool, | ||
| curl: bool = False, | ||
| url: furl | None = None): | ||
| assert not fetch or not curl |
There was a problem hiding this comment.
| assert not fetch or not curl | |
| assert not (fetch and curl) |
is shorter and much more comprehensible
hannes-ucsc
left a comment
There was a problem hiding this comment.
Please separate the wait changes from the POST changes, most likely as a split commit.
CONTRIBUTING.rst
Outdated
| Only if the second and subsequent arguments won't fit on one line, do we | ||
| wrap all arguments, one line per argument. |
There was a problem hiding this comment.
Don't understand why you exclude the first argument.
Furthermore this rule doesn't just apply to calls. Note the bullet uses the term "element", not "argument".
CONTRIBUTING.rst
Outdated
| The exception to this rule are logging method invocations and calls to | ||
| reject(), require(), or the integration test context manager subTest() :: |
There was a problem hiding this comment.
Confusing mixing of conjunctions "and" and "or".
Also, subTest is not just used in IT. It would be arbitrary and confusing to restrict the exception to only the IT usages of subTest.
lambdas/service/app.py
Outdated
| raise BRE(f'The {name!r} parameter is not valid JSON') | ||
|
|
||
|
|
||
| def validate_wait(wait: str | None) -> Optional[int]: |
There was a problem hiding this comment.
Why is this returning a value? None of the other validators do.
lambdas/service/app.py
Outdated
| def wait_parameter_spec(default: int | None = None) -> JSON: | ||
| return params.query( | ||
| 'wait', | ||
| schema.optional(int |
There was a problem hiding this comment.
This would be better described as an enum.
lambdas/service/app.py
Outdated
| '`application/x-www-form-urlencoded` to this endpoint') | ||
| query_params = request.query_params or {} | ||
| _hoist_parameters(query_params, request) | ||
| if is_post and 'wait' not in query_params: |
There was a problem hiding this comment.
Two lines down a different idiom is used to set the default. Be consistent with precedent.
Are you sure the default is needed at this level? You should avoid the case where a default is injected when the parameter is actually disallowed. I'm a bit confused by the fact that the wait parameter is conditional on is_post in one place and on fetch in another. I don't remember: Did I ask you to allow wait in use cases other than the one we actually need it for, i.e., with curl?
There was a problem hiding this comment.
Are you sure the default is needed at this level? You should avoid the case where a default is injected when the parameter is actually disallowed.
Yes, the wait parameter only applies if the initial request was POST, in which case we want to set a default (1) if the param was not specified.
I'm a bit confused by the fact that the
waitparameter is conditional onis_postin one place and onfetchin another. I don't remember: Did I ask you to allowwaitin use cases other than the one we actually need it for, i.e., withcurl?
If an initial POST request to the non-fetch endpoint returns a 301 with a token, the wait parameter is carried over from the initial request, so that each subsequent request to the non-fetch endpoint will have the same wait parameter. This is why the param is conditional on is_post in one section (when setting a default), and is not fetch in another section (when processing a non-initial request). I've added a brief comment to the latter to help clarify.
CONTRIBUTING.rst
Outdated
| The one exception to this rule are logging method invocations and calls to | ||
| reject() and require():: | ||
| The exception to this rule are logging method invocations, calls to | ||
| reject(), require(), or the context manager subTest() :: |
There was a problem hiding this comment.
| reject(), require(), or the context manager subTest() :: | |
| reject(), require(), or uses of TestCase.subTest() :: |
| Only if the second and subsequent arguments won't fit on one line, do we | ||
| wrap all arguments, one line per argument. |
There was a problem hiding this comment.
This should not be deleted.
lambdas/service/app.py
Outdated
|
|
||
| def manifest_route(*, fetch: bool, initiate: bool): | ||
| def wait_parameter_spec(default: str | None = None) -> JSON: | ||
| possible_values = ('0', '1') |
There was a problem hiding this comment.
| possible_values = ('0', '1') | |
| valid_values = ['0', '1'] |
| Requests to this endpoint are idempotent, so PUT would be | ||
| the more standards-compliant method to use. POST is offered | ||
| as a convenience for `curl` users, exploiting the fact that | ||
| `curl` drops to GET when following a redirect in response to | ||
| a POST, but not a PUT request. This is the only reason for | ||
| the deprecation of this endpoint and there are currently no | ||
| plans to remove it. | ||
|
|
||
| To use this endpoint with `curl`, pass the `--location` and | ||
| `--data` options. This makes `curl` automatically follow the | ||
| intermediate redirects to the GET /manifest/files endpoint, | ||
| and ultimately to the URL that yields the manifest. Example: | ||
|
|
||
| ``` | ||
| curl --data "" --location {post_manifest_example_url} | ||
| ``` | ||
|
|
||
| In order to facilitate this, a POST request to this endpoint | ||
| may have a `Content-Type` header of | ||
| `application/x-www-form-urlencoded`, which is what the | ||
| `--data` option sends. The body must be empty in that case | ||
| and parameters cannot be hoisted as described above. |
There was a problem hiding this comment.
This looks like something that was already there but it shows up as an addition without a corresponding deletion. IOW, it appears that this was copied and pasted.
There was a problem hiding this comment.
This originated from a patch you provided, and was extended and refined during review feedback.
|
|
||
| def server_side_sleep(self, max_seconds: int | float) -> float: | ||
| """ | ||
| Run a sleep in the current Lambda function. |
| return Filters(explicit=self._parse_filters(filters), | ||
| source_ids=self._list_source_ids(catalog, authentication)) | ||
|
|
||
| def server_side_sleep(self, max_seconds: int | float) -> float: |
There was a problem hiding this comment.
Why is this in SourceController? It has nothing to do with sources.
| # to the client. | ||
| actual_seconds = min(float(max_seconds), | ||
| remaining_time - config.api_gateway_timeout_padding - 3) | ||
| time.sleep(actual_seconds) |
There was a problem hiding this comment.
We should emit a log entry here.
test/integration_test.py
Outdated
| if retry_after is not None: | ||
| retry_after = float(retry_after) | ||
| if url.args.get('wait') == 1: | ||
| # The waiting should have happened server-side and been |
There was a problem hiding this comment.
| # The waiting should have happened server-side and been | |
| # The wait should have happened server-side and been |
hannes-ucsc
left a comment
There was a problem hiding this comment.
Index: lambdas/service/app.py
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/lambdas/service/app.py b/lambdas/service/app.py
--- a/lambdas/service/app.py (revision 41d9b4b892f0503d0e5e8d6b7129b8211f9590a6)
+++ b/lambdas/service/app.py (date 1723741147188)
@@ -1320,15 +1320,15 @@
authentication=request.authentication)
-def wait_parameter_spec(default: str | None = None) -> JSON:
+def wait_parameter_spec(*, default: str | None = None) -> JSON:
valid_values = ['0', '1']
assert default in (None, *valid_values), default
return params.query(
'wait',
schema.optional(
- schema.with_default(default, type_=schema.enum(*valid_values))
- if default else
schema.enum(*valid_values)
+ if default is None else
+ schema.with_default(default, type_=schema.enum(*valid_values))
),
description=fd('''
If 0, the client is responsible for honoring the waiting period
@@ -1368,8 +1368,8 @@
+ ('/manifest/files' if initiate else '/manifest/files/{token}'),
# The initial PUT request is idempotent.
methods=[method],
- # To support curl requests made with the `--data` option we accept the
- # `application/x-www-form-urlencoded` content-type.
+ # In order to support requests made with `curl` and its `--data` option,
+ # we accept the `application/x-www-form-urlencoded` content-type.
content_types=['application/json', 'application/x-www-form-urlencoded'],
interactive=fetch,
cors=True,
@@ -1378,7 +1378,7 @@
params.path('token', str, description=fd('''
An opaque string representing the manifest preparation job
''')),
- *((wait_parameter_spec(),) if not fetch else ())
+ *([] if fetch else [wait_parameter_spec()])
]
},
method_spec={
@@ -1496,7 +1496,7 @@
'''),
'parameters': [
catalog_param_spec,
- *((wait_parameter_spec('1'),) if curl else ()),
+ *([wait_parameter_spec(default='1')] if curl else []),
filters_param_spec,
params.query(
'format',
@@ -1562,6 +1562,7 @@
'Location': {
'description': fd('''
The URL of the manifest preparation job at
+ # REVIEW: Why was the link removed?
''') + fd('''
the `GET /manifest/files/{token}` endpoint.
''') if initiate else fd('''
@@ -1687,17 +1688,19 @@
def _file_manifest(fetch: bool, token_or_key: Optional[str] = None):
request = app.current_request
- is_post = request.method == 'POST'
+ # REVIEW: We don't us `is_` for `fetch` or `curl` so we should be consistent here
+ post = request.method == 'POST'
if (
- is_post
+ post
and request.headers.get('content-type') == 'application/x-www-form-urlencoded'
and request.raw_body != b''
):
- raise BRE('The body must be empty for a POST request of content-type '
- '`application/x-www-form-urlencoded` to this endpoint')
+ raise BRE('POST requests to this endpoint must have an empty body if '
+ 'they specify a `Content-Type` header of '
+ '`application/x-www-form-urlencoded`')
query_params = request.query_params or {}
_hoist_parameters(query_params, request)
- if is_post:
+ if post:
query_params.setdefault('wait', '1')
if token_or_key is None:
query_params.setdefault('filters', '{}')
@@ -1707,7 +1710,7 @@
catalog=validate_catalog,
format=validate_manifest_format,
filters=validate_filters,
- **({'wait': validate_wait} if is_post else {}))
+ **({'wait': validate_wait} if post else {}))
# Now that the catalog is valid, we can provide the default format that
# depends on it
default_format = app.metadata_plugin.manifest_formats[0].value| ), | ||
| params.query( | ||
| 'wait', | ||
| schema.optional(int), |
There was a problem hiding this comment.
Why was this changed from int to string? Why was the default logic added? I don't think we have a use case for there not being a default.
Meta comment: I try not to ask rhetorical questions, but you often treat my questions as such and assume that I already have an answer particular answer or preference which you then implement. Sometimes you get it right but sometimes you don't. It would be better to just answer my questions. In general, feel free to provide the answer during a PL slot and for the questions in the paragraph above, a PL slot would most certainly be the best venue.
There was a problem hiding this comment.
Why was this changed from int to string?
This was a mistake. I was confused by the string form of the param in the request. I've set it back to int in the schema.
Why was the
defaultlogic added? I don't think we have a use case for there not being a default.
We do not set a wait default value for the /repository/files/{file_uuid} and /fetch/repository/files/{file_uuid} endpoints.
Also, a default value is added for the initial manifest request (POST /manifest/files), but not for the subsequent requests to GET /manifest/files/{token}
REVIEW: Why was the link removed?
Links are not supported in this section of the Swagger
From: https://service.azul.data.humancellatlas.org/#/Manifests/put_manifest_files
There was a problem hiding this comment.
A "default" is a value that takes effect when no explicit value is specified. When an endpoint accepts a wait parameter, that parameter must be optional so as to not complicate "regular" use cases in which the wait can be performed client-side. For every optional parameter a default needs to be specified. Anything else would lead to incorrect or incomplete documentation. If you are unable to implement this directive for any reason, or if you think I am wrong, please request a PL slot so we can clear this up.
There was a problem hiding this comment.
Added a default value for the wait parameter in both the /repository/files and /manifest/files endpoints.
hannes-ucsc
left a comment
There was a problem hiding this comment.
Please push the commits individually.
src/azul/chalice.py
Outdated
| def current_request(self) -> AzulRequest: | ||
| return self.app.current_request | ||
|
|
||
| def server_side_sleep(self, max_seconds: int | float) -> float: |
There was a problem hiding this comment.
| def server_side_sleep(self, max_seconds: int | float) -> float: | |
| def server_side_sleep(self, max_seconds: float) -> float: |
Time in Python is always a float. Supporting int may seem like a convenience but just leads to sloppy use of this function.
|
|
||
| :return: The actual amount of time slept in seconds | ||
| """ | ||
| remaining_time = self.lambda_context.get_remaining_time_in_millis() / 1000 |
There was a problem hiding this comment.
Add validation of type and range of the argument.
| retry_after = body.get('Retry-After') | ||
| if retry_after is not None: | ||
| time_slept = self.server_side_sleep(retry_after) | ||
| body['Retry-After'] = round(retry_after - time_slept) |
There was a problem hiding this comment.
| body['Retry-After'] = round(retry_after - time_slept) | |
| body['Retry-After'] = ceil(retry_after - time_slept) |
| if wait is None: | ||
| azul_url.args['wait'] = '0' |
There was a problem hiding this comment.
This indicates that Azul now yields a URL with wait set to the default. I'd prefer not to do that.
| elif wait == '1': | ||
| time_slept = self.server_side_sleep(float(retry_after)) | ||
| retry_after = round(retry_after - time_slept) | ||
| retry_after = ceil(retry_after - time_slept) |
There was a problem hiding this comment.
The change from round to ceil is currently in a commit labeled "Add a default value for the /repository/files wait parameter". Please explain why going from round to ceil is related to adding a default or isolate that change in its own commit.
There was a problem hiding this comment.
Moved change to commit Fix rounding of /repository/file retry-after value
| as a property of a JSON object in the body of the request. This can be | ||
| useful in case the value of the `filters` query parameter causes the URL | ||
| to exceed the maximum length of 8192 characters, resulting in a 413 | ||
| Request Entity Too Large response. | ||
| as a property of a JSON object in the body of the request. This is | ||
| referred to as *parameter hoisting* and can be useful in case the value | ||
| of the `filters` query parameter causes the URL to exceed the maximum | ||
| length of 8192 characters, resulting in a 413 Request Entity Too Large | ||
| response. | ||
|
|
||
| The request `%s %s?filters={…}`, for example, is equivalent to `%s %s` | ||
| with the body `{"filters": "{…}"}` in which any double quotes or | ||
| backslash characters inside `…` are escaped with another backslash. That | ||
| escaping is the requisite procedure for embedding one JSON structure | ||
| inside another. | ||
| with a `Content-Type` header of `application/json` and the body | ||
| `{"filters": "{…}"}` in which any double quotes or backslash characters | ||
| inside `…` are escaped with another backslash. That escaping is the | ||
| requisite procedure for embedding one JSON structure inside another. |
There was a problem hiding this comment.
Correct me if I am wrong but it seems like these two hunks are unrelated documentation improvement that I requested at some point, but that have nothing to do with "Add[ing] support for POST requests to the manifest endpoint", as is the title of the commit the hunks are part of.
There was a problem hiding this comment.
Moved change to commit Refine parameter hoisting note
Fixes regression from fb58b01. The problem manifested as an error with the `app` property in an AppController subclass, and this wasn't noticed until now due to the `app` property not being accessed in a controller that had this problem, namely ManifestController.
…vocation (#5918) Add support for POST requests to the manifest endpoint
…vocation (#5918) Add a wait parameter option to the manifest endpoint


Linked issues: #5918
Checklist
Author
developissues/<GitHub handle of author>/<issue#>-<slug>1 when the issue title describes a problem, the corresponding PR
title is
Fix:followed by the issue titleAuthor (partiality)
ptag to titles of partial commitspartialor completely resolves all linked issuespartiallabelAuthor (reindex)
rtag to commit title or the changes introduced by this PR will not require reindexing of any deploymentreindex:devor the changes introduced by it will not require reindexing ofdevreindex:anvildevor the changes introduced by it will not require reindexing ofanvildevreindex:anvilprodor the changes introduced by it will not require reindexing ofanvilprodreindex:prodor the changes introduced by it will not require reindexing ofprodreindex:partialand its description documents the specific reindexing procedure fordev,anvildev,anvilprodandprodor requires a full reindex or carries none of the labelsreindex:dev,reindex:anvildev,reindex:anvilprodandreindex:prodAuthor (API changes)
APIor this PR does not modify a REST APIa(A) tag to commit title for backwards (in)compatible changes or this PR does not modify a REST APIapp.pyor this PR does not modify a REST APIAuthor (upgrading deployments)
make docker_images.jsonand committed the resulting changes or this PR does not modifyazul_docker_images, or any other variables referenced in the definition of that variableutag to commit title or this PR does not require upgrading deploymentsupgradeor does not require upgrading deploymentsdeploy:sharedor does not modifydocker_images.json, and does not require deploying thesharedcomponent for any other reasondeploy:gitlabor does not require deploying thegitlabcomponentdeploy:runneror does not require deploying therunnerimageAuthor (hotfixes)
Ftag to main commit title or this PR does not include permanent fix for a temporary hotfixanvilprodandprod) have temporary hotfixes for any of the issues linked to this PRAuthor (before every review)
develop, squashed fixups from prior reviewsmake requirements_updateor this PR does not modifyrequirements*.txt,common.mk,Makefile,Dockerfileorenvironment.bootRtag to commit title or this PR does not modifyrequirements*.txtreqsor does not modifyrequirements*.txtmake integration_testpasses in personal deployment or this PR does not modify functionality that could affect the IT outcomePeer reviewer (after approval)
Note that when requesting changes, the PR must be assigned back to the author.
System administrator (after approval)
demoorno demono demono sandboxN reviewslabel is accurateOperator
reindex:…labels andrcommit title tagno demodevelopOperator (deploy
.sharedand.gitlabcomponents)_select dev.shared && CI_COMMIT_REF_NAME=develop make -C terraform/shared apply_keep_unusedor this PR is not labeleddeploy:shared_select dev.gitlab && CI_COMMIT_REF_NAME=develop make -C terraform/gitlab applyor this PR is not labeleddeploy:gitlab_select anvildev.shared && CI_COMMIT_REF_NAME=develop make -C terraform/shared apply_keep_unusedor this PR is not labeleddeploy:shared_select anvildev.gitlab && CI_COMMIT_REF_NAME=develop make -C terraform/gitlab applyor this PR is not labeleddeploy:gitlabdeploy:gitlabdeploy:gitlabSystem administrator (post-deploy of
.gitlabcomponent)dev.gitlabare complete or this PR is not labeleddeploy:gitlabanvildev.gitlabare complete or this PR is not labeleddeploy:gitlabOperator (deploy runner image)
_select dev.gitlab && make -C terraform/gitlab/runneror this PR is not labeleddeploy:runner_select anvildev.gitlab && make -C terraform/gitlab/runneror this PR is not labeleddeploy:runnerOperator (sandbox build)
sandboxlabel or PR is labeledno sandboxdevor PR is labeledno sandboxanvildevor PR is labeledno sandboxsandboxdeployment or PR is labeledno sandboxanvilboxdeployment or PR is labeledno sandboxsandboxdeployment or PR is labeledno sandboxanvilboxdeployment or PR is labeledno sandboxsandboxor this PR does not remove catalogs or otherwise causes unreferenced indices indevanvilboxor this PR does not remove catalogs or otherwise causes unreferenced indices inanvildevsandboxor this PR is not labeledreindex:devanvilboxor this PR is not labeledreindex:anvildevsandboxor this PR is not labeledreindex:devanvilboxor this PR is not labeledreindex:anvildevOperator (merge the branch)
pif the PR is also labeledpartialOperator (main build)
devanvildevdevdevanvildevanvildev_select dev.shared && make -C terraform/shared applyor this PR is not labeleddeploy:shared_select anvildev.shared && make -C terraform/shared applyor this PR is not labeleddeploy:shareddevanvildevOperator (reindex)
devor this PR is neither labeledreindex:partialnorreindex:devanvildevor this PR is neither labeledreindex:partialnorreindex:anvildevdevor this PR is neither labeledreindex:partialnorreindex:devanvildevor this PR is neither labeledreindex:partialnorreindex:anvildevdevor this PR is neither labeledreindex:partialnorreindex:devanvildevor this PR is neither labeledreindex:partialnorreindex:anvildevdevor this PR does not require reindexingdevanvildevor this PR does not require reindexinganvildevdevor this PR does not require reindexingdevanvildevor this PR does not require reindexinganvildevdevor this PR does not require reindexingdevanvildevor this PR does not require reindexinganvildevdevor this PR does not require reindexingdevdevor this PR does not require reindexingdevdeploy_browserjob in the GitLab pipeline for this PR indevor this PR does not require reindexingdevanvildevor this PR does not require reindexinganvildevdeploy_browserjob in the GitLab pipeline for this PR inanvildevor this PR does not require reindexinganvildevOperator (mirroring)
devor this PR does not require mirroringdevanvildevor this PR does not require mirroringanvildevdevor this PR does not require mirroringdevanvildevor this PR does not require mirroringanvildevdevor this PR does not require mirroringdevanvildevor this PR does not require mirroringanvildevOperator
deploy:shared,deploy:gitlab,deploy:runner,API,reindex:partial,reindex:anvilprodandreindex:prodlabels to the next promotion PRs or this PR carries none of these labelsdeploy:shared,deploy:gitlab,deploy:runner,API,reindex:partial,reindex:anvilprodandreindex:prodlabels, from the description of this PR to that of the next promotion PRs or this PR carries none of these labelsShorthand for review comments
Lline is too longWline wrapping is wrongQbad quotesFother formatting problem