From cf9961711d90569a2a20e009f20956902abc5397 Mon Sep 17 00:00:00 2001
From: Jennifer Chang <jennifer.chang.bioinform@gmail.com>
Date: Wed, 6 Mar 2024 14:36:48 -0800
Subject: [PATCH] Refactor upload configs to match pathogen-repo-guide

https://github.com/nextstrain/pathogen-repo-guide/tree/f825132e441a672586ec1748af4cdb149669f7c3/ingest/build-configs/nextstrain-automation
---
 .../nextstrain-automation/config.yaml         | 37 ++++++++++---------
 .../nextstrain-automation/upload.smk          | 19 +++++-----
 2 files changed, 29 insertions(+), 27 deletions(-)

diff --git a/ingest/build-configs/nextstrain-automation/config.yaml b/ingest/build-configs/nextstrain-automation/config.yaml
index 9f00a7e..2e819bb 100644
--- a/ingest/build-configs/nextstrain-automation/config.yaml
+++ b/ingest/build-configs/nextstrain-automation/config.yaml
@@ -1,22 +1,23 @@
-# Optional configs used by Nextstrain team
-# Params for uploads
-upload:
-  # Upload params for AWS S3
-  s3:
-    # AWS S3 Bucket with prefix
-    dst: 's3://nextstrain-data/files/workflows/zika'
-    # Mapping of files to upload, with key as remote file name and the value
-    # the local file path relative to the ingest directory.
-    files_to_upload:
-      genbank.ndjson.xz: data/genbank.ndjson
-      all_sequences.ndjson.xz: data/sequences.ndjson
-      metadata.tsv.gz: results/metadata.tsv
-      sequences.fasta.xz: results/sequences.fasta
-      alignment.fasta.xz: data/alignment.fasta
-      insertions.csv.gz: data/insertions.csv
-      translations.zip: data/translations.zip
+# This configuration file should contain all required configuration parameters
+# for the ingest workflow to run with additional Nextstrain automation rules.
 
-    cloudfront_domain: 'data.nextstrain.org'
+# Custom rules to run as part of the Nextstrain automated workflow
+# The paths should be relative to the ingest directory.
+custom_rules:
+  - build-configs/nextstrain-automation/upload.smk
+
+# Nextstrain CloudFront domain to ensure that we invalidate CloudFront after the S3 uploads
+# This is required as long as we are using the AWS CLI for uploads
+cloudfront_domain: "data.nextstrain.org"
+
+# Nextstrain AWS S3 Bucket with pathogen prefix
+# Replace <pathogen> with the pathogen repo name.
+s3_dst: "s3://nextstrain-data/files/workflows/<pathogen>"
+
+files_to_upload:
+  genbank.ndjson.zst: data/genbank.ndjson
+  metadata.tsv.zst: results/metadata.tsv
+  sequences.fasta.zst: results/sequences.fasta
 
 # Toggle for Slack notifications
 send_slack_notifications: True
diff --git a/ingest/build-configs/nextstrain-automation/upload.smk b/ingest/build-configs/nextstrain-automation/upload.smk
index 60c5c9b..1c82699 100644
--- a/ingest/build-configs/nextstrain-automation/upload.smk
+++ b/ingest/build-configs/nextstrain-automation/upload.smk
@@ -1,13 +1,14 @@
 """
-This part of the workflow handles uploading files to a specified destination.
+This part of the workflow handles uploading files to AWS S3.
 
-Uses predefined wildcard `file_to_upload` determine input and predefined
-wildcard `remote_file_name` as the remote file name in the specified destination.
+Files to upload must be defined in the `files_to_upload` config param, where
+the keys are the remote files and the values are the local filepaths
+relative to the ingest directory.
 
-Produces output files as `data/upload/{upload_target_name}/{remote_file_name}.done`.
+Produces a single file for each uploaded file:
+    "results/upload/{remote_file}.upload"
 
-Currently only supports uploads to AWS S3, but additional upload rules can
-be easily added as long as they follow the output pattern described above.
+The rule `upload_all` can be used as a target to upload all files.
 """
 import os
 
@@ -27,7 +28,7 @@ def _get_upload_inputs(wildcards):
     `send_notifications` is True.
     """
     inputs = {
-        "file_to_upload": config["upload"]["s3"]["files_to_upload"][
+        "file_to_upload": config["files_to_upload"][
             wildcards.remote_file_name
         ],
     }
@@ -52,8 +53,8 @@ rule upload_to_s3:
         "data/upload/s3/{remote_file_name}.done",
     params:
         quiet="" if send_notifications else "--quiet",
-        s3_dst=config["upload"].get("s3", {}).get("dst", ""),
-        cloudfront_domain=config["upload"].get("s3", {}).get("cloudfront_domain", ""),
+        s3_dst=config["s3_dst"],
+        cloudfront_domain=config["cloudfront_domain"],
     shell:
         """
         ./vendored/upload-to-s3 \