Merge branch 'main' into joseph/rewrite-test-results

.github/workflows/ci.yml

@@ -22,11 +22,11 @@ jobs:
   lint:
     name: Run Lint
 
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
 
   build:
     name: Build Worker
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
 
     secrets: inherit
     with:
@@ -35,14 +35,14 @@ jobs:
   codecovstartup:
     name: Codecov Startup
     needs: build
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
 
     secrets: inherit
 
   test:
     name: Test
     needs: [build]
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
 
     secrets: inherit
     with:
@@ -52,7 +52,7 @@ jobs:
     name: Build Self Hosted Worker
     needs: [build, test]
 
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
     secrets: inherit
     with:
       repo: ${{ vars.CODECOV_IMAGE_V2 || 'codecov/self-hosted-worker' }}
@@ -61,7 +61,7 @@ jobs:
     name: Push Staging Image
     needs: [build, test]
     if: ${{ github.event_name == 'push' && (github.event.ref == 'refs/heads/main' || github.event.ref == 'refs/heads/staging') && github.repository_owner == 'codecov' }}
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
     secrets: inherit
     with:
       environment: staging
@@ -71,7 +71,7 @@ jobs:
     name: Push Production Image
     needs: [build, test]
     if: ${{ github.event_name == 'push' && github.event.ref == 'refs/heads/main' && github.repository_owner == 'codecov' }}
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
     secrets: inherit
     with:
       environment: production
@@ -82,7 +82,7 @@ jobs:
     needs: [build-self-hosted, test]
     secrets: inherit
     if: ${{ github.event_name == 'push' && github.event.ref == 'refs/heads/main' && github.repository_owner == 'codecov' }}
-    uses: codecov/gha-workflows/.github/workflows/[email protected].23
+    uses: codecov/gha-workflows/.github/workflows/[email protected].24
     with:
       push_rolling: true
       repo: ${{ vars.CODECOV_IMAGE_V2 || 'codecov/self-hosted-worker' }}

database/models/core.py

@@ -16,7 +16,7 @@
 from helpers.config import should_write_data_to_storage_config_check
 
 
-class User(CodecovBaseModel):
+class User(CodecovBaseModel, MixinBaseClass):
     __tablename__ = "users"
     id_ = Column("id", types.BigInteger, primary_key=True)
 
@@ -136,7 +136,7 @@ class Repository(CodecovBaseModel):
     updatestamp = Column(types.DateTime)
     yaml = Column(postgresql.JSON)
     deleted = Column(types.Boolean, nullable=False, default=False)
-    branch = Column(types.Text)
+    branch = Column(types.Text, default="main")
     image_token = Column(
         types.Text,
         default=lambda: "".join(
@@ -184,7 +184,7 @@ class GithubAppInstallation(CodecovBaseModel, MixinBaseClass):
     # replacement for owner.integration_id
     # installation id GitHub sends us in the installation-related webhook events
     installation_id = Column(types.Integer, server_default=FetchedValue())
-    name = Column(types.Text, server_default=FetchedValue())
+    name = Column(types.Text, default=GITHUB_APP_INSTALLATION_DEFAULT_NAME)
     # if null, all repos are covered by this installation
     # otherwise, it's a list of repo.id values
     repository_service_ids = Column(
@@ -199,7 +199,7 @@ class GithubAppInstallation(CodecovBaseModel, MixinBaseClass):
         Owner, foreign_keys=[ownerid], back_populates="github_app_installations"
     )
 
-    is_suspended = Column(types.Boolean, server_default=FetchedValue())
+    is_suspended = Column(types.Boolean, default=False)
 
     def repository_queryset(self, dbsession: Session):
         """Returns a query set of repositories covered by this installation"""
@@ -364,7 +364,7 @@ class Branch(CodecovBaseModel):
     __tablename__ = "branches"
 
     repoid = Column(types.Integer, ForeignKey("repos.repoid"), primary_key=True)
-    updatestamp = Column(types.DateTime)
+    updatestamp = Column(types.DateTime, default=datetime.now)
     branch = Column(types.Text, nullable=False, primary_key=True)
     base = Column(types.Text)
     head = Column(types.Text, nullable=False)

database/tests/factories/core.py

@@ -112,7 +112,7 @@ class Meta:
         ).hexdigest()
     )
     ci_passed = True
-    pullid = 1
+    pullid = None
     timestamp = datetime(2019, 2, 1, 17, 59, 47)
     author = factory.SubFactory(OwnerFactory)
     repository = factory.SubFactory(RepositoryFactory)

helpers/checkpoint_logger/__init__.py

@@ -17,79 +17,22 @@
 )
 
 import sentry_sdk
-from shared.metrics import Counter, Histogram, metrics
+from shared.metrics import metrics
 
-logger = logging.getLogger(__name__)
+from helpers.checkpoint_logger.prometheus import PROMETHEUS_HANDLER
 
+logger = logging.getLogger(__name__)
 
 T = TypeVar("T", bound="BaseFlow")
 TSubflows: TypeAlias = Mapping[T, Iterable[tuple[str, T]]]
 
 
-CHECKPOINTS_TOTAL_BEGUN = Counter(
-    "worker_checkpoints_begun",
-    "Total number of times a flow's first checkpoint was logged.",
-    ["flow"],
-)
-CHECKPOINTS_TOTAL_SUCCEEDED = Counter(
-    "worker_checkpoints_succeeded",
-    "Total number of times one of a flow's success checkpoints was logged.",
-    ["flow"],
-)
-CHECKPOINTS_TOTAL_FAILED = Counter(
-    "worker_checkpoints_failed",
-    "Total number of times one of a flow's failure checkpoints was logged.",
-    ["flow"],
-)
-CHECKPOINTS_TOTAL_ENDED = Counter(
-    "worker_checkpoints_ended",
-    "Total number of times one of a flow's terminal checkpoints (success or failure) was logged.",
-    ["flow"],
-)
-CHECKPOINTS_ERRORS = Counter(
-    "worker_checkpoints_errors",
-    "Total number of errors while trying to log checkpoints",
-    ["flow"],
-)
-CHECKPOINTS_EVENTS = Counter(
-    "worker_checkpoints_events",
-    "Total number of checkpoints logged.",
-    ["flow", "checkpoint"],
-)
-
-CHECKPOINTS_SUBFLOW_DURATION = Histogram(
-    "worker_checkpoints_subflow_duration_seconds",
-    "Duration of subflows in seconds.",
-    ["flow", "subflow"],
-    buckets=[
-        0.05,
-        0.1,
-        0.5,
-        1,
-        2,
-        5,
-        10,
-        30,
-        60,
-        120,
-        180,
-        300,
-        600,
-        900,
-        1200,
-        1800,
-        2400,
-        3600,
-    ],
-)
-
-
 def _error(msg, flow, strict=False):
     # When a new version of worker rolls out, it will pick up tasks that
     # may have been enqueued by the old worker and be missing checkpoints
     # data. At least for that reason, we want to allow failing softly.
     metrics.incr("worker.checkpoint_logger.error")
-    CHECKPOINTS_ERRORS.labels(flow=flow.__name__).inc()
+    PROMETHEUS_HANDLER.log_errors(flow=flow.__name__)
     if strict:
         raise ValueError(msg)
     else:
@@ -331,12 +274,12 @@ class MyEnum(str, Enum):
 
     def log_counters(obj: T) -> None:
         metrics.incr(f"{klass.__name__}.events.{obj.name}")
-        CHECKPOINTS_EVENTS.labels(flow=klass.__name__, checkpoint=obj.name).inc()
+        PROMETHEUS_HANDLER.log_checkpoints(flow=klass.__name__, checkpoint=obj.name)
 
         # If this is the first checkpoint, increment the number of flows we've begun
         if obj == next(iter(klass.__members__.values())):
             metrics.incr(f"{klass.__name__}.total.begun")
-            CHECKPOINTS_TOTAL_BEGUN.labels(flow=klass.__name__).inc()
+            PROMETHEUS_HANDLER.log_begun(flow=klass.__name__)
             return
 
         is_failure = hasattr(obj, "is_failure") and obj.is_failure()
@@ -345,14 +288,14 @@ def log_counters(obj: T) -> None:
 
         if is_failure:
             metrics.incr(f"{klass.__name__}.total.failed")
-            CHECKPOINTS_TOTAL_FAILED.labels(flow=klass.__name__).inc()
+            PROMETHEUS_HANDLER.log_failure(flow=klass.__name__)
         elif is_success:
             metrics.incr(f"{klass.__name__}.total.succeeded")
-            CHECKPOINTS_TOTAL_SUCCEEDED.labels(flow=klass.__name__).inc()
+            PROMETHEUS_HANDLER.log_success(flow=klass.__name__)
 
         if is_terminal:
             metrics.incr(f"{klass.__name__}.total.ended")
-            CHECKPOINTS_TOTAL_ENDED.labels(flow=klass.__name__).inc()
+            PROMETHEUS_HANDLER.log_total_ended(flow=klass.__name__)
 
     klass.log_counters = log_counters
     return klass
@@ -373,7 +316,7 @@ class CheckpointLogger(Generic[T]):
     reconstructed from its serialized data allowing you to begin a flow on one host
     and log its completion on another (as long as clock drift is marginal).
 
-    See `UploadFlow` for an example of defining a flow. It's recomended that you
+    See `UploadFlow` for an example of defining a flow. It's recommended that you
     define your flow with the decorators in this file:
     - `@success_events()`, `@failure_events()`: designate some events as terminal
       success/fail states of your flow.
@@ -489,9 +432,9 @@ def submit_subflow(self: _Self, metric: str, start: T, end: T) -> _Self:
         if duration:
             sentry_sdk.set_measurement(metric, duration, "milliseconds")
             duration_in_seconds = duration / 1000
-            CHECKPOINTS_SUBFLOW_DURATION.labels(
-                flow=self.cls.__name__, subflow=metric
-            ).observe(duration_in_seconds)
+            PROMETHEUS_HANDLER.log_subflow(
+                flow=self.cls.__name__, subflow=metric, duration=duration_in_seconds
+            )
 
         return self
 

helpers/checkpoint_logger/prometheus.py

@@ -0,0 +1,94 @@
+from shared.metrics import Counter, Histogram
+
+# Main Counter
+CHECKPOINTS_TOTAL_BEGUN = Counter(
+    "worker_checkpoints_begun",
+    "Total number of times a flow's first checkpoint was logged.",
+    ["flow"],
+)
+CHECKPOINTS_TOTAL_SUCCEEDED = Counter(
+    "worker_checkpoints_succeeded",
+    "Total number of times one of a flow's success checkpoints was logged.",
+    ["flow"],
+)
+CHECKPOINTS_TOTAL_FAILED = Counter(
+    "worker_checkpoints_failed",
+    "Total number of times one of a flow's failure checkpoints was logged.",
+    ["flow"],
+)
+CHECKPOINTS_TOTAL_ENDED = Counter(
+    "worker_checkpoints_ended",
+    "Total number of times one of a flow's terminal checkpoints (success or failure) was logged.",
+    ["flow"],
+)
+CHECKPOINTS_ERRORS = Counter(
+    "worker_checkpoints_errors",
+    "Total number of errors while trying to log checkpoints",
+    ["flow"],
+)
+CHECKPOINTS_EVENTS = Counter(
+    "worker_checkpoints_events",
+    "Total number of checkpoints logged.",
+    ["flow", "checkpoint"],
+)
+CHECKPOINTS_SUBFLOW_DURATION = Histogram(
+    "worker_checkpoints_subflow_duration_seconds",
+    "Duration of subflows in seconds.",
+    ["flow", "subflow"],
+    buckets=[
+        0.05,
+        0.1,
+        0.5,
+        1,
+        2,
+        5,
+        10,
+        30,
+        60,
+        120,
+        180,
+        300,
+        600,
+        900,
+        1200,
+        1800,
+        2400,
+        3600,
+    ],
+)
+
+
+class PrometheusCheckpointLoggerHandler:
+    """
+    PrometheusCheckpointLoggerHandler is a class that is responsible for all
+    Prometheus related logs. This checkpoint logic is responsible for logging
+    metrics to any checkpoints we define. This class is made with the intent
+    of extending different checkpoints for metrics for different needs. The
+    methods in this class are mainly used by the CheckpointLogger class.
+    """
+
+    def log_begun(self, flow: str):
+        CHECKPOINTS_TOTAL_BEGUN.labels(flow=flow).inc()
+
+    def log_failure(self, flow: str):
+        CHECKPOINTS_TOTAL_FAILED.labels(flow=flow).inc()
+
+    def log_success(self, flow: str):
+        CHECKPOINTS_TOTAL_SUCCEEDED.labels(flow=flow).inc()
+
+    def log_total_ended(self, flow: str):
+        CHECKPOINTS_TOTAL_ENDED.labels(flow=flow).inc()
+
+    def log_checkpoints(self, flow: str, checkpoint: str):
+        CHECKPOINTS_EVENTS.labels(flow=flow, checkpoint=checkpoint).inc()
+
+    def log_errors(self, flow: str):
+        CHECKPOINTS_ERRORS.labels(flow=flow).inc()
+
+    def log_subflow(self, flow: str, subflow: str, duration: int):
+        CHECKPOINTS_SUBFLOW_DURATION.labels(flow=flow, subflow=subflow).observe(
+            duration
+        )
+
+
+PROMETHEUS_HANDLER = PrometheusCheckpointLoggerHandler()