From 43940e23c0323a42d17873920025b64f9e9f2723 Mon Sep 17 00:00:00 2001
From: John Hooks <bitmachina@outlook.com>
Date: Fri, 15 Dec 2023 02:52:08 -0800
Subject: [PATCH] docs: add schema.md

---
 docs/architecture.md | 14 ++++++++
 docs/schema.md       | 78 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 92 insertions(+)
 create mode 100644 docs/architecture.md
 create mode 100644 docs/schema.md

diff --git a/docs/architecture.md b/docs/architecture.md
new file mode 100644
index 0000000..280a409
--- /dev/null
+++ b/docs/architecture.md
@@ -0,0 +1,14 @@
+# Architecture
+
+- Tokenization / Lexer
+  - HTML tokenization
+  - Stemming
+- Index
+  - Repository
+  - Cron
+- Ranking
+- Query
+  - Parser / Lexer
+  - Repository
+- Stop Words
+- Config
diff --git a/docs/schema.md b/docs/schema.md
new file mode 100644
index 0000000..cc1f741
--- /dev/null
+++ b/docs/schema.md
@@ -0,0 +1,78 @@
+# Schema
+
+The database base structure used to index WordPress documents for search.
+
+## Document
+
+Has many words through the `document_word` table.
+
+Has many meta words through the `document_meta` table.
+
+Has many hits through the `query_hit` table.
+
+The biggest documents to index will be posts.
+
+There will need to be a way to indicated document type.
+
+|id|index_id|
+|--|--------|
+
+## Index
+
+Used to group indexed documents and track queries.
+
+Has many documents.
+
+|id|name|description|type|
+|--|----|-----------|----|
+
+## DocumentWord
+
+Ranking the importance of a word in the document based on meta words could be done during the index process, if we know how to rank the words at the time of indexing.
+
+|id|doc_id|word_id|position|
+|--|------|-------|--------|
+
+## DocumentMeta
+
+Meta word table. `meta_id` is an enum of meta word types.
+
+|id|doc_id|meta_id|position|
+|--|------|-------|--------|
+
+## Word
+
+Has many documents through `document_word` table.
+
+|id|value|
+|--|-----|
+
+## Gram
+
+Unique segments of words.
+
+Have many words through the `gram_word` table.
+
+|id|value|length|
+|--|-----|------|
+
+## GramWord
+
+Map partial word to full word
+
+|id|word_id|gram_id|hit_count|
+|--|-------|-------|---------|
+
+## Query
+
+User input queries.
+
+|id|value|
+|--|-----|
+
+## Query Hit
+
+Match successful queries to specific words contained in a document, independent of location with in the document. The concept is to have permanent link between query -> word -> document, used for ranking based on previous results.
+
+|id|query_id|word_id|document_id|count|
+|--|--------|-------|-----------|-----|