Change logging and add secrets

.github/workflows/daily-csv-upload.yaml

@@ -2,6 +2,7 @@ name: Upload current-full.csv and current-federal.csv
 run-name: Upload current-full.csv and current-federal.csv for branch ${{ github.head_ref }}
 
 on:
+  pull_request:
   workflow_dispatch:
     inputs:
       environment:
@@ -42,9 +43,24 @@ jobs:
             const environment = (github && github.event && github.event.inputs) ? github.event.inputs.environment : 'za';
             core.setOutput('environment', environment);
 
+  wait-for-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Wait for deploy to complete
+        uses: fountainhead/action-wait-for-check@v1.0.0
+        id: wait-for-deploy
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          checkName: "deploy"
+          ref: ${{ github.event.pull_request.head.sha }}
+          # the maximum time to wait for the check to complete, in seconds
+          timeoutSeconds: 600 
+          # the time to wait between checks, in seconds
+          intervalSeconds: 10 
+
   upload-reports:
     runs-on: ubuntu-latest
-    needs: [variables]
+    needs: [variables, wait-for-deploy]
     env:
       CF_USERNAME: CF_{{ needs.variables.outputs.environment }}_USERNAME
       CF_PASSWORD: CF_{{ needs.variables.outputs.environment }}_PASSWORD

src/api/views.py

@@ -12,7 +12,7 @@ from login_required import login_not_required
 
 from cachetools.func import ttl_cache
 
-from registrar.utility.s3_bucket import S3ClientHelper
+from registrar.utility.s3_bucket import S3ClientError, S3ClientHelper
 
 
 DOMAIN_FILE_URL = "https://raw.githubusercontent.com/cisagov/dotgov-data/main/current-full.csv"
@@ -103,31 +103,32 @@ def available(request, domain=""):
 
 @require_http_methods(["GET"])
 @login_not_required
-def get_current_full(request, file_path="migrationdata/current-full.csv"):
+def get_current_full(request, file_name="current-full.csv"):
     """This will return the file content of current-full.csv which is the command
     output of generate_current_full_report.py. This command iterates through each Domain
     and returns a CSV representation."""
-    return serve_file(file_path, "current-full.csv")
+    return serve_file(file_name)
 
 
 @require_http_methods(["GET"])
 @login_not_required
-def get_current_federal(request, file_path="migrationdata/current-federal.csv"):
+def get_current_federal(request, file_name="current-federal.csv"):
     """This will return the file content of current-federal.csv which is the command
     output of generate_current_federal_report.py. This command iterates through each Domain
     and returns a CSV representation."""
-    return serve_file(file_path, "current-federal.csv")
+    return serve_file(file_name)
 
 
-def serve_file(file_path, file_name):
-    """Downloads a file based on a given filepath. Returns a 404 if not found."""
+def serve_file(file_name):
+    """Downloads a file based on a given filepath. Returns a 500 if not found."""
     s3_client = S3ClientHelper()
-    # TODO - #1403, grab from the S3 instance instead
-    # TODO - check if file exists in s3, not here
-    if os.path.exists(file_path):
-        # Serve the CSV file
-        file = s3_client.get_file(file_name)
-        response = FileResponse(file)
+    # Serve the CSV file. If not found, an exception will be thrown.
+    # This will then be caught by flat, causing it to not read it - which is what we want.
+    try:
+        file = s3_client.get_file(file_name, decode_to_utf=True)
+    except S3ClientError as err:
+        # TODO - #1317: Notify operations when auto report generation fails
+        raise err
+
+    response = HttpResponse(file)
     return response
-    else:
-        return HttpResponse("File not found", status=404)

src/registrar/config/settings.py

@@ -58,6 +58,11 @@ secret_key = secret("DJANGO_SECRET_KEY")
 secret_aws_ses_key_id = secret("AWS_ACCESS_KEY_ID", None)
 secret_aws_ses_key = secret("AWS_SECRET_ACCESS_KEY", None)
 
+aws_s3_region_name = secret("AWS_S3_ACCESS_KEY_ID", None)
+secret_aws_s3_key_id = secret("AWS_S3_SECRET_ACCESS_KEY", None)
+secret_aws_s3_key = secret("AWS_S3_REGION", None)
+secret_aws_s3_bucket_name = secret("AWS_S3_BUCKET_NAME", None)
+
 secret_registry_cl_id = secret("REGISTRY_CL_ID")
 secret_registry_password = secret("REGISTRY_PASSWORD")
 secret_registry_cert = b64decode(secret("REGISTRY_CERT", ""))
@@ -257,6 +262,13 @@ AUTH_USER_MODEL = "registrar.User"
 AWS_ACCESS_KEY_ID = secret_aws_ses_key_id
 AWS_SECRET_ACCESS_KEY = secret_aws_ses_key
 AWS_REGION = "us-gov-west-1"
+
+# Configuration for accessing AWS S3
+AWS_S3_ACCESS_KEY_ID = secret_aws_s3_key_id
+AWS_S3_SECRET_ACCESS_KEY = secret_aws_s3_key
+AWS_S3_REGION = aws_s3_region_name
+AWS_S3_BUCKET_NAME = secret_aws_s3_bucket_name
+
 # https://boto3.amazonaws.com/v1/documentation/api/latest/guide/retries.html#standard-retry-mode
 AWS_RETRY_MODE: Final = "standard"
 # base 2 exponential backoff with max of 20 seconds:

src/registrar/management/commands/generate_current_federal_report.py

@@ -4,6 +4,7 @@ import os
 
 from django.core.management import BaseCommand
 from registrar.utility import csv_export
+from registrar.utility.s3_bucket import S3ClientHelper
 
 
 logger = logging.getLogger(__name__)
@@ -23,20 +24,32 @@ class Command(BaseCommand):
 
     def handle(self, **options):
         """Grabs the directory then creates current-federal.csv in that directory"""
+        file_name = "current-federal.csv"
         # Ensures a slash is added
         directory = os.path.join(options.get("directory"), "")
         check_path = options.get("checkpath")
+
         logger.info("Generating report...")
+        try:
+            self.generate_current_federal_report(directory, file_name, check_path)
+        except Exception as err:
+            # TODO - #1317: Notify operations when auto report generation fails
+            raise err
+        else:
+            logger.info(f"Success! Created {file_name}")
 
-        self.generate_current_federal_report(directory, check_path)
-        logger.info(f"Success! Created {directory}current-federal.csv")
+    def generate_current_federal_report(self, directory, file_name, check_path):
+        """Creates a current-full.csv file under the specified directory, 
+        then uploads it to a AWS S3 bucket"""
+        s3_client = S3ClientHelper()
+        file_path = os.path.join(directory, file_name)
         
-    def generate_current_federal_report(self, directory, check_path):
-        """Creates a current-full.csv file under the specified directory"""
-        # TODO - #1403, push to the S3 instance instead
-        file_path = os.path.join(directory, "current-federal.csv")
+        # Generate a file locally for upload
         with open(file_path, "w") as file:
             csv_export.export_data_federal_to_csv(file)
 
         if check_path and not os.path.exists(file_path):
             raise FileNotFoundError(f"Could not find newly created file at '{file_path}'")
+
+        # Upload this generated file for our S3 instance        
+        s3_client.upload_file(file_path, file_name)

src/registrar/management/commands/generate_current_full_report.py

@@ -28,23 +28,29 @@ class Command(BaseCommand):
         # Ensures a slash is added
         directory = os.path.join(options.get("directory"), "")
         check_path = options.get("checkpath")
+
         logger.info("Generating report...")
-
+        try:
             self.generate_current_full_report(directory, file_name, check_path)
-
-        file_path = os.path.join(directory, file_name)
-        logger.info(f"Success! Created {file_path}")
+        except Exception as err:
+            # TODO - #1317: Notify operations when auto report generation fails
+            raise err
+        else:
+            logger.info(f"Success! Created {file_name}")
 
     def generate_current_full_report(self, directory, file_name, check_path):
-        """Creates a current-full.csv file under the specified directory"""
+        """Creates a current-full.csv file under the specified directory, 
+        then uploads it to a AWS S3 bucket"""
         s3_client = S3ClientHelper()
-        # TODO - #1403, push to the S3 instance instead
         file_path = os.path.join(directory, file_name)
-        # TODO - Don't genererate a useless file
+        
+        # Generate a file locally for upload
         with open(file_path, "w") as file:
             csv_export.export_data_full_to_csv(file)
 
         if check_path and not os.path.exists(file_path):
             raise FileNotFoundError(f"Could not find newly created file at '{file_path}'")
 
+        # Upload this generated file for our S3 instance        
         s3_client.upload_file(file_path, file_name)
+

src/registrar/utility/s3_bucket.py

@@ -19,8 +19,8 @@ class S3ClientHelper:
                 "s3",
                 region_name=settings.AWS_S3_REGION,
                 aws_access_key_id=settings.AWS_S3_ACCESS_KEY_ID,
-                aws_secret_access_key=settings.AWS_S3_SECRET_ACCESS_KEY
-                #config=settings.BOTO_CONFIG,
+                aws_secret_access_key=settings.AWS_S3_SECRET_ACCESS_KEY,
+                config=settings.BOTO_CONFIG,
             )
         except Exception as exc:
             raise S3ClientError("Could not access the S3 client.") from exc
@@ -49,9 +49,10 @@ class S3ClientHelper:
         """Gets a file to our S3 instance and returns the file content"""
         try:
             response = self.boto_client.get_object(Bucket=self.get_bucket_name(), Key=file_name)
+        except self.boto_client.exceptions.NoSuchKey as exc:
+            raise S3ClientError("File was not found") from exc
         except Exception as exc:
-            raise S3ClientError("Couldn't get file") from exc
-
+            raise S3ClientError("Couldn't get file, an unspecified error occured") from exc
         file_content = response["Body"].read()
         if decode_to_utf:
             return file_content.decode("utf-8")