From d9a8c44fdddacd09a2a14139ae673ad386232f3b Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@archive.org>
Date: Fri, 16 Oct 2020 18:15:41 -0700
Subject: proposal: kafka update pipeline(s)

---
 proposals/kafka_update_pipeline.md | 47 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)
 create mode 100644 proposals/kafka_update_pipeline.md

(limited to 'proposals/kafka_update_pipeline.md')

diff --git a/proposals/kafka_update_pipeline.md b/proposals/kafka_update_pipeline.md
new file mode 100644
index 0000000..86ee167
--- /dev/null
+++ b/proposals/kafka_update_pipeline.md
@@ -0,0 +1,47 @@
+
+Want to receive a continual stream of updates from both fatcat and SIM
+scanning; index the updated content; and push into elasticsearch.
+
+
+## Message Types
+
+Scholar Update Request JSON
+- `key`: str
+- `type`: str
+    - `fatcat_work`
+    - `sim_issue`
+- `updated`: datetime, UTC, of event resulting in this request
+- `work_ident`: str (works)
+- `fatcat_changelog`: int (works)
+- `sim_item`: str (items)
+
+"Heavy Intermediate" JSON (existing schema)
+- key
+- `fetched`: Optional[datetime], UTC, when this doc was collected
+
+Scholar Fulltext ES JSON (existing schema)
+
+
+## Kafka Topics
+
+fatcat-ENV.work-ident-updates
+    6x, long retention, key compaction
+    key: doc ident
+scholar-ENV.sim-updates
+    6x, long retention, key compaction
+    key: doc ident
+scholar-ENV.update-docs
+    12x, short retention (2 months?)
+    key: doc ident
+
+## Workers
+
+scholar-fetch-docs-worker
+    consumes fatcat and/or sim update requests, individually
+    constructs heavy intermediate
+    publishes to update-docs topic
+
+scholar-index-docs-worker
+    consumes updated "heavy intermediate" documents, in batches
+    transforms to elasticsearch schema
+    updates elasticsearch
-- 
cgit v1.2.3