From 6af463ed44aea0b7a4bc8c86dfbdd638206041cb Mon Sep 17 00:00:00 2001
From: Per Stark <perstark.se@gmail.com>
Date: Wed, 16 Oct 2024 09:39:15 +0200
Subject: [PATCH] database plan

---
 database.md                | 59 ++++++++++++++++++++++++++++++++++++++
 src/models/text_content.rs |  2 ++
 2 files changed, 61 insertions(+)
 create mode 100644 database.md

diff --git a/database.md b/database.md
new file mode 100644
index 0000000..5990988
--- /dev/null
+++ b/database.md
@@ -0,0 +1,59 @@
+# SurrealDB only
+
+Right now we have the FileInfo stored in "files"
+
+- Change the uuid to Uuid type, and have the database layer still use String. Means parsing and unparsing but thats fine.
+
+```
+pub struct FileInfo {
+    pub uuid: String,
+    pub sha256: String,
+    pub path: String,
+    pub mime_type: String,
+}
+```
+
+We create TextContent objects, which we should store?
+
+- We store the "snippets" along with the vectors, but it would make sense to store the whole textcontent, at least for not enormous files?
+
+```
+pub struct TextContent {
+    pub id: Uuid,
+    pub text: String,
+    pub file_info: Option<FileInfo>,
+    pub instructions: String,
+    pub category: String,
+}
+```
+
+We create KnowledgeSource, which we will store
+
+- Add a uuid to we can link the textcontent and files to the knowledge sources?
+
+```
+pub struct KnowledgeSource {
+    pub id: String,
+    pub title: String,
+    pub description: String,
+    pub relationships: Vec<Relationship>,
+}
+```
+
+We will create embeddings and vector representations of TextContent, possibly split up and store in vector DB
+
+```
+pub struct VectorEmbeddingOfTextContent {
+    pub id: Uuid,
+    pub vectors: Vec<u8>(or something),
+    pub text_content: String,
+    pub category: String,
+}
+```
+
+______________________________________________________________________
+
+## Goals
+
+- Smooth operations when updating, removing and adding data
+- Smooth queries where one can search, get a vector snippet, which links to a graph node and its edges, and also the fulltext document.
diff --git a/src/models/text_content.rs b/src/models/text_content.rs
index f145107..4deaee1 100644
--- a/src/models/text_content.rs
+++ b/src/models/text_content.rs
@@ -84,6 +84,7 @@ impl TextContent {
     }
 
     /// Stores analysis results in a graph database.
+    #[allow(dead_code)]
     async fn store_in_graph_db(&self, _analysis: &AnalysisResult) -> Result<(), ProcessingError> {
         // TODO: Implement storage logic for your specific graph database.
         // Example:
@@ -95,6 +96,7 @@ impl TextContent {
     }
 
     /// Splits text and stores it in a vector database.
+    #[allow(dead_code)]
     async fn store_in_vector_db(&self) -> Result<(), ProcessingError> {
         // TODO: Implement text splitting and vector storage logic.
         // Example: