[FEATURE] Creating a new command to export database data to CSV. (#4572)

* [FEATURE] Creating a new command to export database data to CSV.

Issue #4561

* Updating queries, bucket

* Renaming querries

.github/workflows/export-data-to-csv.yml

@@ -0,0 +1,55 @@
+---
+name: Export Data to CSV
+on:
+  # schedule:
+  # Monthly, on the 5th, at 8am UTC (3am EST)
+  #  - cron: '0 8 5 * *'
+  workflow_dispatch:
+    inputs:
+      environment:
+        required: true
+        type: choice
+        description: The environment the workflow should run on.
+        options:
+          - dev
+          - staging
+          - preview
+          - production
+
+jobs:
+  scheduled-data-export:
+    if: ${{ github.event_name == 'schedule' }}
+    strategy:
+      matrix:
+        environments: ["production"]  # For now, just do the scheduled job on production to save space
+    name: Run data export on ${{ inputs.environment }}
+    runs-on: ubuntu-latest
+    environment: ${{ matrix.environments }}
+    env:
+      space: ${{ matrix.environments }}
+    steps:
+      - name: Run Command
+        uses: cloud-gov/cg-cli-tools@main
+        with:
+          cf_username: ${{ secrets.CF_USERNAME }}
+          cf_password: ${{ secrets.CF_PASSWORD }}
+          cf_org: gsa-tts-oros-fac
+          cf_space: ${{ env.space }}
+          command: cf run-task gsa-fac -k 2G -m 2G --name export_data_to_csv --command "python manage.py export_data"
+
+  dispatch-data-export:
+    if: ${{ github.event.inputs.environment != '' }}
+    name: Run data export on ${{ inputs.environment }}
+    runs-on: ubuntu-latest
+    environment: ${{ inputs.environment }}
+    env:
+      space: ${{ inputs.environment }}
+    steps:
+      - name: Run Command
+        uses: cloud-gov/cg-cli-tools@main
+        with:
+          cf_username: ${{ secrets.CF_USERNAME }}
+          cf_password: ${{ secrets.CF_PASSWORD }}
+          cf_org: gsa-tts-oros-fac
+          cf_space: ${{ env.space }}
+          command: cf run-task gsa-fac -k 2G -m 2G --name export_data_to_csv --command "python manage.py export_data"

backend/requirements.txt

@@ -1193,6 +1193,14 @@ six==1.16.0 \
     #   orderedmultidict
     #   pyjwkest
     #   python-dateutil
+sling==1.3.4 \
+    --hash=sha256:70541da99b48313a0cfd5e61c52d984e4d50113b7945308697d32cd50a6bb183
+    # via -r ./requirements/requirements.in
+sling-linux-arm64==1.3.4 \
+    --hash=sha256:08977df437742b415232e8c09c2f7f43181a3b764bef601fa6ebd369f6777783
+sling-linux-amd64==1.3.4 \
+    --hash=sha256:d5f3b71fb191e8b9663f88890c464d84ab4c36df3a5fd6148a51cdd2ea9a5099
+    # via sling
 sqlalchemy==2.0.36 \
     --hash=sha256:03e08af7a5f9386a43919eda9de33ffda16b44eb11f3b313e6822243770e9763 \
     --hash=sha256:0572f4bd6f94752167adfd7c1bed84f4b240ee6203a95e05d1e208d488d0d436 \

backend/requirements/requirements.in

@@ -36,6 +36,7 @@ python-slugify
 pyyaml
 requests>=2.32.3
 sqlalchemy
+sling
 sqlparse>=0.5.0
 types-python-dateutil==2.9.0.20240821
 uritemplate

backend/support/management/commands/export_data.py

@@ -0,0 +1,138 @@
+import os
+import logging
+
+from config import settings
+from datetime import datetime
+from django.core.management.base import BaseCommand, CommandError
+from sling import Replication, ReplicationStream
+
+from support.decorators import newrelic_timing_metric
+from dissemination.summary_reports import restricted_model_names
+
+logger = logging.getLogger(__name__)
+
+S3_CONNECTION = f"""{{
+            "type": "s3",
+            "bucket": "{settings.AWS_PRIVATE_STORAGE_BUCKET_NAME}",
+            "access_key_id": "{settings.AWS_PRIVATE_ACCESS_KEY_ID}",
+            "secret_access_key": "{settings.AWS_PRIVATE_SECRET_ACCESS_KEY}",
+            "endpoint": "{settings.AWS_S3_ENDPOINT_URL}"
+         }}
+    """
+DB_URL = os.environ.get("DATABASE_URL")
+FAC_DB_URL = (
+    f"{DB_URL}?sslmode=disable" if settings.ENVIRONMENT in ["LOCAL", "TEST"] else DB_URL
+)
+DEFAULT_OPTIONS = {
+    "target_options": {
+        "format": "csv",
+        "compression": "gzip",
+        "file_max_rows": 0,
+    }
+}
+
+
+class StreamGenerator:
+    EXCLUDE_NONPUBLIC_QUERY = (
+        "select * from {table_name} where report_id in ("
+        " select dg.report_id from public.dissemination_general dg"
+        " where dg.audit_year = '{audit_year}' and dg.is_public = 'true' )"
+    )
+
+    UNRESTRICTED_QUERY = (
+        "select * from {table_name} where report_id in ("
+        " select dg.report_id from public.dissemination_general dg"
+        " where dg.audit_year = '{audit_year}')"
+    )
+
+    def __init__(self, table_name, friendly_name, query_override=None):
+        self.table_name = table_name
+        self.friendly_name = friendly_name
+
+        restricted_tables = [
+            "dissemination_" + model for model in restricted_model_names
+        ]
+        default_query = (
+            self.EXCLUDE_NONPUBLIC_QUERY
+            if table_name in restricted_tables
+            else self.UNRESTRICTED_QUERY
+        )
+        self.query = query_override or default_query
+
+    def generate_stream(self, audit_year):
+        return (
+            f"{self.table_name}.{audit_year}",
+            ReplicationStream(
+                object=f"bulk_export/{{MM}}/{audit_year}_{self.friendly_name}.csv",
+                sql=self.query.format(
+                    table_name=self.table_name, audit_year=audit_year
+                ),
+                mode="full-refresh",
+                target_options={"format": "csv"},
+            ),
+        )
+
+
+STREAM_GENERATORS = [
+    StreamGenerator(
+        friendly_name="General",
+        table_name="dissemination_general",
+        query_override="select * from dissemination_general where audit_year = '{audit_year}'",
+    ),
+    StreamGenerator(
+        friendly_name="AdditionalEIN", table_name="dissemination_additionalein"
+    ),
+    StreamGenerator(
+        friendly_name="AdditionalUEI", table_name="dissemination_additionaluei"
+    ),
+    StreamGenerator(
+        friendly_name="CorrectiveActionPlans", table_name="dissemination_captext"
+    ),
+    StreamGenerator(
+        friendly_name="FederalAward", table_name="dissemination_federalaward"
+    ),
+    StreamGenerator(friendly_name="Finding", table_name="dissemination_finding"),
+    StreamGenerator(
+        friendly_name="FindingText", table_name="dissemination_findingtext"
+    ),
+    StreamGenerator(friendly_name="Note", table_name="dissemination_note"),
+    StreamGenerator(
+        friendly_name="PassThrough", table_name="dissemination_passthrough"
+    ),
+    StreamGenerator(
+        friendly_name="SecondaryAuditor", table_name="dissemination_secondaryauditor"
+    ),
+]
+
+
+@newrelic_timing_metric("data_export")
+def _run_data_export():
+    logger.info("Begin exporting data")
+    # We may want to consider instead of hardcoding 2016 only export the past X years.
+    # This will only export data that exists, so doing +2 just incase some data is in early
+    years = range(2016, datetime.today().year + 2)
+    streams = {}
+    for stream_generator in STREAM_GENERATORS:
+        for year in years:
+            streams.update([stream_generator.generate_stream(year)])
+
+    replication = Replication(
+        source="FAC_DB",
+        target="BULK_DATA_EXPORT",
+        streams=streams,
+        defaults=DEFAULT_OPTIONS,
+        env=dict(FAC_DB=FAC_DB_URL, BULK_DATA_EXPORT=S3_CONNECTION),
+        debug=settings.DEBUG,
+    )
+    logger.info(f"Exporting {len(streams)} streams")
+    replication.run()
+    logger.info("Successfully exported data")
+
+
+class Command(BaseCommand):
+    def handle(self, *args, **kwargs):
+        try:
+            _run_data_export()
+        except Exception as ex:
+            logger.error("An error occurred while exporting data", exc_info=ex)
+            raise CommandError("Error while exporting data")