From bae889ddf219a49e7c43c414b260ff862b7e4f07 Mon Sep 17 00:00:00 2001
From: pyr0ball <pyroballpcs@gmail.com>
Date: Wed, 13 May 2026 15:51:19 -0700
Subject: [PATCH] feat: add context_facts, context_documents, context_chunks
 tables to schema

---
 app/ingest/pipeline.py       | 29 +++++++++++++++++++++++++++++
 tests/context/__init__.py    |  0
 tests/context/test_schema.py | 24 ++++++++++++++++++++++++
 3 files changed, 53 insertions(+)
 create mode 100644 tests/context/__init__.py
 create mode 100644 tests/context/test_schema.py

diff --git a/app/ingest/pipeline.py b/app/ingest/pipeline.py
index bddaf43..cf1a16a 100644
--- a/app/ingest/pipeline.py
+++ b/app/ingest/pipeline.py
@@ -62,6 +62,35 @@ CREATE TABLE IF NOT EXISTS received_bundles (
 );
 CREATE INDEX IF NOT EXISTS idx_bundles_bundled ON received_bundles(bundled_at);
 CREATE INDEX IF NOT EXISTS idx_bundles_type    ON received_bundles(issue_type);
+
+CREATE TABLE IF NOT EXISTS context_facts (
+    id           TEXT PRIMARY KEY,
+    category     TEXT NOT NULL,
+    key          TEXT NOT NULL,
+    value        TEXT NOT NULL,
+    source       TEXT,
+    created_at   TEXT NOT NULL
+);
+CREATE INDEX IF NOT EXISTS idx_facts_category ON context_facts(category);
+CREATE INDEX IF NOT EXISTS idx_facts_key      ON context_facts(key);
+
+CREATE TABLE IF NOT EXISTS context_documents (
+    id           TEXT PRIMARY KEY,
+    filename     TEXT NOT NULL,
+    doc_type     TEXT NOT NULL,
+    full_text    TEXT NOT NULL,
+    file_size    INTEGER,
+    uploaded_at  TEXT NOT NULL
+);
+
+CREATE TABLE IF NOT EXISTS context_chunks (
+    id           TEXT PRIMARY KEY,
+    document_id  TEXT NOT NULL REFERENCES context_documents(id) ON DELETE CASCADE,
+    chunk_index  INTEGER NOT NULL,
+    text         TEXT NOT NULL,
+    embedding    BLOB
+);
+CREATE INDEX IF NOT EXISTS idx_chunks_doc ON context_chunks(document_id);
 """
 
 
diff --git a/tests/context/__init__.py b/tests/context/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/tests/context/test_schema.py b/tests/context/test_schema.py
new file mode 100644
index 0000000..69b0327
--- /dev/null
+++ b/tests/context/test_schema.py
@@ -0,0 +1,24 @@
+"""Verify the three new context tables are created by ensure_schema."""
+import sqlite3
+from pathlib import Path
+import pytest
+from app.ingest.pipeline import ensure_schema
+
+
+def test_context_tables_created(tmp_path):
+    db = tmp_path / "t.db"
+    ensure_schema(db)
+    conn = sqlite3.connect(str(db))
+    tables = {r[0] for r in conn.execute(
+        "SELECT name FROM sqlite_master WHERE type='table'"
+    ).fetchall()}
+    conn.close()
+    assert "context_facts"     in tables
+    assert "context_documents" in tables
+    assert "context_chunks"    in tables
+
+
+def test_context_schema_idempotent(tmp_path):
+    db = tmp_path / "t.db"
+    ensure_schema(db)
+    ensure_schema(db)  # second call must not raise