feat: move clickhouse load to luigi

v03_pipeline/api/request_handlers.py

@@ -1,8 +1,6 @@
-import json
 from collections.abc import Callable
 from typing import Any
 
-import hailtop.fs as hfs
 import luigi
 import luigi.execution_summary
 
@@ -12,62 +10,20 @@
     PipelineRunnerRequest,
     RebuildGtStatsRequest,
 )
-from v03_pipeline.lib.core import DatasetType, ReferenceGenome
+from v03_pipeline.lib.core import DatasetType, FeatureFlag
 from v03_pipeline.lib.logger import get_logger
 from v03_pipeline.lib.misc.clickhouse import (
     delete_family_guids,
     rebuild_gt_stats,
 )
-from v03_pipeline.lib.misc.retry import retry
-from v03_pipeline.lib.paths import (
-    clickhouse_load_success_file_path,
-    metadata_for_run_path,
+from v03_pipeline.lib.tasks.write_clickhouse_load_success_file import (
+    WriteClickhouseLoadSuccessFileTask,
 )
 from v03_pipeline.lib.tasks.write_success_file import WriteSuccessFileTask
 
 logger = get_logger(__name__)
 
 
-@retry()
-def fetch_run_metadata(
-    reference_genome: ReferenceGenome,
-    dataset_type: DatasetType,
-    run_id: str,
-) -> tuple[list[str], list[str]]:
-    # Run metadata
-    with hfs.open(
-        metadata_for_run_path(
-            reference_genome,
-            dataset_type,
-            run_id,
-        ),
-        'r',
-    ) as f:
-        metadata_json = json.load(f)
-        project_guids = metadata_json['project_guids']
-        family_guids = list(metadata_json['family_samples'].keys())
-    return project_guids, family_guids
-
-
-@retry()
-def write_success_file(
-    reference_genome: ReferenceGenome,
-    dataset_type: DatasetType,
-    run_id: str,
-):
-    with hfs.open(
-        clickhouse_load_success_file_path(
-            reference_genome,
-            dataset_type,
-            run_id,
-        ),
-        'w',
-    ) as f:
-        f.write('')
-    msg = f'Successfully loaded {reference_genome.value}/{dataset_type.value}/{run_id}'
-    logger.info(msg)
-
-
 def run_loading_pipeline(
     lpr: LoadingPipelineRequest,
     run_id: str,
@@ -79,6 +35,13 @@ def run_loading_pipeline(
             WriteSuccessFileTask(
                 run_id=run_id,
                 **lpr.model_dump(exclude='request_type'),
+            )
+            if FeatureFlag.CLICKHOUSE_LOADER_DISABLED
+            else (
+                WriteClickhouseLoadSuccessFileTask(
+                    run_id=run_id,
+                    **lpr.model_dump(exclude='request_type'),
+                )
             ),
         ],
         detailed_summary=True,

v03_pipeline/bin/clickhouse_loader.py

@@ -1,12 +1,12 @@
 #!/usr/bin/env python3
+import json
 import signal
 import sys
 import time
 
 import hailtop.fs as hfs
 
-from v03_pipeline.api.request_handlers import fetch_run_metadata, write_success_file
-from v03_pipeline.lib.core import FeatureFlag
+from v03_pipeline.lib.core import DatasetType, FeatureFlag, ReferenceGenome
 from v03_pipeline.lib.logger import get_logger
 from v03_pipeline.lib.misc.clickhouse import (
     drop_staging_db,
@@ -15,6 +15,8 @@
 from v03_pipeline.lib.misc.runs import get_run_ids
 from v03_pipeline.lib.paths import (
     clickhouse_load_fail_file_path,
+    clickhouse_load_success_file_path,
+    metadata_for_run_path,
 )
 
 logger = get_logger(__name__)
@@ -31,6 +33,26 @@ def signal_handler(*_):
 signal.signal(signal.SIGTERM, signal_handler)
 
 
+def fetch_run_metadata(
+    reference_genome: ReferenceGenome,
+    dataset_type: DatasetType,
+    run_id: str,
+) -> tuple[list[str], list[str]]:
+    # Run metadata
+    with hfs.open(
+        metadata_for_run_path(
+            reference_genome,
+            dataset_type,
+            run_id,
+        ),
+        'r',
+    ) as f:
+        metadata_json = json.load(f)
+        project_guids = metadata_json['project_guids']
+        family_guids = list(metadata_json['family_samples'].keys())
+    return project_guids, family_guids
+
+
 def main():
     reference_genome, dataset_type, run_id = None, None, None
     while True:
@@ -69,7 +91,15 @@ def main():
                         project_guids,
                         family_guids,
                     )
-                    write_success_file(reference_genome, dataset_type, run_id)
+                    with hfs.open(
+                        clickhouse_load_success_file_path(
+                            reference_genome,
+                            dataset_type,
+                            run_id,
+                        ),
+                        'w',
+                    ) as f:
+                        f.write('')
         except Exception:
             logger.exception('Unhandled Exception')
             if reference_genome and dataset_type and run_id:

v03_pipeline/bin/pipeline_worker_test.py

@@ -27,12 +27,12 @@ def output(self):
 
 class PipelineWorkerTest(MockedDatarootTestCase):
     @patch('v03_pipeline.lib.misc.slack._safe_post_to_slack')
-    @patch('v03_pipeline.api.request_handlers.WriteSuccessFileTask')
+    @patch('v03_pipeline.api.request_handlers.WriteClickhouseLoadSuccessFileTask')
     @patch('v03_pipeline.bin.pipeline_worker.logger')
     def test_process_queue(
         self,
         mock_logger,
-        mock_write_success_file_task,
+        mock_write_clickhouse_load_success_file_task,
         mock_safe_post_to_slack,
     ):
         raw_request = {
@@ -43,7 +43,7 @@ def test_process_queue(
             'reference_genome': ReferenceGenome.GRCh38.value,
             'dataset_type': DatasetType.SNV_INDEL.value,
         }
-        mock_write_success_file_task.return_value = MockCompleteTask()
+        mock_write_clickhouse_load_success_file_task.return_value = MockCompleteTask()
         os.makedirs(
             loading_pipeline_queue_dir(),
             exist_ok=True,
@@ -62,12 +62,12 @@ def test_process_queue(
         )
 
     @patch('v03_pipeline.lib.misc.slack._safe_post_to_slack')
-    @patch('v03_pipeline.api.request_handlers.WriteSuccessFileTask')
+    @patch('v03_pipeline.api.request_handlers.WriteClickhouseLoadSuccessFileTask')
     @patch('v03_pipeline.bin.pipeline_worker.logger')
     def test_process_failure(
         self,
         mock_logger,
-        mock_write_success_file_task,
+        mock_write_clickhouse_load_success_file_task,
         mock_safe_post_to_slack,
     ):
         raw_request = {
@@ -78,7 +78,7 @@ def test_process_failure(
             'reference_genome': ReferenceGenome.GRCh38.value,
             'dataset_type': DatasetType.SNV_INDEL.value,
         }
-        mock_write_success_file_task.return_value = MyFailingTask()
+        mock_write_clickhouse_load_success_file_task.return_value = MyFailingTask()
         os.makedirs(
             loading_pipeline_queue_dir(),
             exist_ok=True,

v03_pipeline/lib/tasks/load_complete_run_to_clickhouse.py

@@ -0,0 +1,64 @@
+import json
+
+import hailtop.fs as hfs
+import luigi
+import luigi.util
+
+from v03_pipeline.lib.misc.clickhouse import (
+    ClickHouseTable,
+    TableNameBuilder,
+    load_complete_run,
+    logged_query,
+)
+from v03_pipeline.lib.paths import metadata_for_run_path
+from v03_pipeline.lib.tasks.base.base_loading_run_params import (
+    BaseLoadingRunParams,
+)
+from v03_pipeline.lib.tasks.write_success_file import WriteSuccessFileTask
+
+
+@luigi.util.inherits(BaseLoadingRunParams)
+class LoadCompleteRunToClickhouse(luigi.Task):
+    attempt_id = luigi.IntParameter()
+
+    def requires(self) -> luigi.Task:
+        return [self.clone(WriteSuccessFileTask)]
+
+    def complete(self):
+        table_name_builder = TableNameBuilder(
+            self.reference_genome,
+            self.dataset_type,
+            self.run_id,
+        )
+        max_key_src = logged_query(
+            f"""
+            SELECT max(key) FROM {table_name_builder.src_table(ClickHouseTable.ANNOTATIONS_MEMORY)}
+            """,
+        )[0][0]
+        return logged_query(
+            f"""
+            SELECT EXISTS (
+                SELECT 1
+                FROM {table_name_builder.dst_table(ClickHouseTable.ANNOTATIONS_MEMORY)}
+                WHERE key = %(max_key_src)
+            );
+            """,
+            {'max_key_src': max_key_src},
+        )[0][0]
+
+    def run(self):
+        with hfs.open(
+            metadata_for_run_path(
+                self.reference_genome,
+                self.dataset_type,
+                self.run_id,
+            ),
+        ) as f:
+            family_guids = list(json.load(f)['family_samples'].keys())
+        load_complete_run(
+            self.reference_genome,
+            self.dataset_type,
+            self.run_id,
+            self.project_guids,
+            family_guids,
+        )

v03_pipeline/lib/tasks/write_clickhouse_load_success_file.py

@@ -0,0 +1,32 @@
+import luigi
+import luigi.util
+
+from v03_pipeline.lib.paths import clickhouse_load_success_file_path
+from v03_pipeline.lib.tasks.base.base_loading_run_params import (
+    BaseLoadingRunParams,
+)
+from v03_pipeline.lib.tasks.files import GCSorLocalTarget
+from v03_pipeline.lib.tasks.load_complete_run_to_clickhouse import (
+    LoadCompleteRunToClickhouse,
+)
+
+
+@luigi.util.inherits(BaseLoadingRunParams)
+class WriteClickhouseLoadSuccessFileTask(luigi.Task):
+    attempt_id = luigi.IntParameter()
+
+    def output(self) -> luigi.Target:
+        return GCSorLocalTarget(
+            clickhouse_load_success_file_path(
+                self.reference_genome,
+                self.dataset_type,
+                self.run_id,
+            ),
+        )
+
+    def requires(self) -> luigi.Task:
+        return self.clone(LoadCompleteRunToClickhouse)
+
+    def run(self):
+        with self.output().open('w') as f:
+            f.write('')